University of Southern California Dissertations and Theses

The measurement of physical functioning in a longitudinal study of Americans 70 years of age and older

(USC Thesis Other)

The measurement of physical functioning in a longitudinal study of Americans 70 years of age and older

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy subm itted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. ProQuest Information and Learning 300 North Zeeb Road, Ann Arbor, M l 48106-1346 USA 800-521-0600 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Copyright 2000 THE MEASUREMENT OF PHYSICAL FUNCTIONING IN A LONGITUDINAL STUDY OF AMERICANS 70 YEARS OF AGE AND OLDER by Melissa Tabbarah A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Gerontology) August 2000 Melissa Tabbarah Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 3018133 Copyright 2000 by Tabbarah, Melissa All rights reserved. ___ ® UMI UMI Microform 3018133 Copyright 2001 by Bell & Howell Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. Bell & Howell Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UNIVERSITY OF SOUTHERN CALIFORNIA IHSOftADUAI* SCHOOL UMvmmrrAU LOS a n o b a c a u p o rm a nm This dissertation, written by • • • • Jt Wm. 3Af c] »F. 9&. ---------------------- under die directum of k Dissertation CommUtee, end approved by edits member % has been presented to and accepted by The Graduate School, in partial fulfilment of re• qvirements for die degree of DOCTOR OF PHILOSOPHY O — cf G ndu,* Shtdits Date „..JuLy..23«..2Q Q Q DISSERTATION COMMITTEE Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This dissertation is dedicated to my teachers. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGMENTS Patient perseverance. A simple phrase that describes the process involved in climbing this “Everest” o f my life. When my feet questioned walking further toward the summit. I looked around me and I found you smiling. You inspired me when the terrain looked impossible to traverse. You are my family, friends, mentors, and students: Hassan Tabbarah. M.D.; Kay Tabbarah, R.N.; Amanda Postalakis, J.D.; Stephen Postalakis. J.D.; Reem Tabbarah, M.S.W./M.P.H.; Christopher Paul, A.B.D.; Christy Beaudin, Ph.D.; Jennifer Chisholm-Huddleston, M.P.H.; Vandna Pandita, M.P.H.; Sasha Bucur, A.B.D.; Susan Slockdale. A.B.D.; Damien Ellwood; Lauren Evans: Rick Russell, Ph.D.; Gamward Quan; Judy Yip. Ph.D.; Jung Ki Kim, A.B.D.; Hiroshi Ueda, Ph.D.; Jeff Hyde. A.B.D.: Susan Stewart. Ph.D.; Rhea Simonsen; Barney Simonsen; Anna Simonsen; Heather Moore: Timothy Moore. D.D.S.; Abdul Abukurah, M.D.; Hazzar Abukurah; George Iskandar, M.D.; Eileen Iskandar; Darrel Harrington. M.D.; Misty Nitta-Yee, M.S.G.; Nick Nitta-Yee. M.S.G.; Nicole Zitta; Ondrez Bures; Will Douglass: Kevin Tomczyk; Eileen M.Crimmins, Ph.D.: Teresa E. Seeman. Ph.D.; James M. Ferris, Ph.D.: Merril D. Silverstein, Ph.D.; Jon Pynoos, Ph.D.; Phoebe S. Liebig. Ph.D.; and the students o f the introductory statistics classes. The successful completion of a dissertation is not possible without the support of a dissertation committee. The members of my committee were Eileen M. Crimmins, Ph.D.; Teresa E. Seeman, Ph.D.; and James M. Ferris, Ph.D. Lead by Eileen, this committee provided valuable suggestions on how to improve this work. I especially thank Eileen and Teresa because they consistently found the time to encourage and promote my scholarship. In addition, during my tenure at the University o f Southern iii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. California, Eileen provided immeasurable opportunities to: 1) publish in peer reviewed journals, 2) teach, and 3) explore related areas o f interest. The generous extension of herself for the advancement o f her students is extraordinary. I must also thank the National Institute on Aging. Without their generous financial support, this work could not have been realized. While I was writing this dissertation, I was encouraged by two special people: Christopher Paul, A.B.D. and Christy Beaudin, Ph.D.. Christopher picked me up and dusted me off when I fell; listened to my anxieties and frustrations; reminded me to live a “balanced" life; and extended a gentle nudge when I needed to work harder. He is my mountain o f emotional support and it would have been particularly challenging to complete this dissertation without him. Christy is the older sister I always wanted. Her mentorship, generosity, and laughter helped me smile and encouraged me to think “out of the box" in order to create opportunities for myself and those around me. My family taught me the importance of a dream and my father taught me how to persevere. 1 am, indeed, privileged. To realize a dream, excellence, integrity, a positive attitude, and some gentle encouragement are required. While my father promoted the characteristics o f integrity and excellence, my mother showed me the necessity of cultivating a positive attitude. In addition, when I needed to step away from this work, my sisters extended the gentle encouragement and humor that facilitated my ability to return to this work with renewed energy and confidence. Together, you, my family, friend, mentors, and students, have taught me that the realization o f any dream is possible. Thank-you. I could not have climbed this mountain without you. iv Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS PAGE DEDICATION ii ACKNOWLEDGEMENTS iii LIST OF FIGURES ix LIST OF TABLES xi ABSTRACT xvi CHAPTER I INTRODUCTION 1 A. Focus o f Study 3 B. Significance of Contribution 5 C. Overview of Chapters 6 1. Conceptual Framework and Literature Review 6 2. Description o f Data and Measures 7 3. Discrepancies Between Self-Reported and 8 Performance-Based Measures of Physical Functioning 4. Sources of Discrepancies Within Different Domains 8 of Self-Reported Measures o f Physical Functioning 5. The Effect o f Change in Cognitive Performance on 9 Change in Performance-Based Measures o f Physical Functioning 6. Conclusion 9 CHAPTER II CONCEPTUAL FRAMEWORK AND LITERATURE REVIEW 11 A. Introduction 11 B. History o f Functional Assessment 12 C. Types o f Functional Assessment: Their Advantages and Disadvantages 14 V Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS (Continued) PAGE CHAPTER II (Continued) D. Conceptual Frameworks 23 1. International Classification o f Impairments. 23 Disabilities, and Handicaps (ICIDH) 2. The Nagi Framework 26 3. The Disablement Process 30 4. The Revised ICIDH (ICIDH-2) 33 E. Self-Reported and Performance-Based Measures o f 44 Physical Functioning F. Sources o f Discrepancies in the Cross-Section 62 1. Demographic Characteristics 62 2. Health Characteristics 67 3. Psychological Attributes 74 G. Sources of Discrepancies Over Time 76 1. Depression 77 2. Psychological Attributes 78 H. Other Potential Risk Factors: Behavioral Characteristics 79 I. The Relationship Between Cognitive Impairment and 83 Physical Functioning J. Risk Factors Specific to Cognitive Impairment and 88 Physical Functioning K. Conclusion 92 CHAPTER III DESCRIPTION OF DATA AND MEASURES 95 A. Introduction 95 B. Description o f Data and Sample Used 95 C. Measurement o f Constructs 97 1. Physical Functioning 97 vi Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS (Continued) PAGE CHAPTER III (Continued) 2. Demographic Characteristics 101 3. Health Conditions and Events 102 4. Cognitive Performance 103 5. Depression 104 6. Health Behaviors 104 7. Psychological Characteristics 107 D. Descriptive Statistics o f the Sample 107 E. Summary 115 CHAPTER IV DISCREPANCIES BETWEEN SELF-REPORTED AND 117 PERFORMANCE-BASED MEASURES OF PHYSICAL FUNCTIONING A. Introduction 117 B. Method 119 C. Results 140 D. Discussion 153 CHAPTER V SOURCES OF DISCREPANCIES WITHIN DIFFERENT DOMAINS 158 OF SELF-REPORTED MEASURES OF PHYSICAL FUNCTIONING A. Introduction 158 B. Discrepancies Within Summary Measures 158 C. Method 165 D. Results 175 1. Cross-Sectional 178 2. Longitudinal 183 E. Discussion 190 vii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS (Continued) PAGE CHAPTER VI THE EFFECT OF CHANGE IN COGNITIVE PERFORMANCE 194 ON CHANGE IN PERFORMANCE BASED MEASURES OF PHYSICAL FUNCITONING A. Introduction 194 B. Method 196 C. Results 208 1. Part 1: Cases Studies 209 2. Part 2: Predictors o f change for 12 Performance 222 Based Measures D. Discussion 228 CHAPTER VII CONCLUSION 231 A. Summary and Discussion of Findings 234 B. Implications for Future Research 240 REFERENCES 246 v iii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES Figure II-1. Figure II-2. Figure II-3. Figure II-4. Figure II-5. Figure III-1. Figure III-2. Figure IV-l. Figure IV-2. Figure IV-3. Figure IV-4. Figure IV-5. Figure IV-6. Figure IV-7. The International Classification o f Impairments, Disabilities, and Handicaps (WHO, 1980) TheN agi Scheme (Nagi, 1965, 1969, 1976, 1977) Comparison o f the Nagi and Wood models The Disablement Process (Verbrugge and Jette. 1994) Person-environment interaction between dimensions of ICIDH-2 (WHO, 1999) Distribution of Cognitive Score Distribution o f Depression Score A priori comparisons of self-reported and performance-based measures Gait Function: Self-reported ability to walk across a room and performance on the timed walking o f a 10 foot course. Mac Arthur Study (1988. 1991. and 1995) Lower Body Strength: Self-reported ability to use the toilet and performance on timed 5 repeated chair stands, MacArthur Study (1988. 1991. and 1995) Lower Body Strength: Self-reported ability to stoop, crouch, or kneel and performance on the timed 5 repeated chair stands. MacArthur Study (1988. 1991. and 1995) Fine Motor Dexterity of Gross Function: Self-reported ability to use the toilet and strength o f grip on a dynamometer. MacArthur Study (1988, 1991, 1995) Fine Motor Dexterity o f Gross Function: Self-reported ability to use the toilet and performance on hand signature. MacArthur Study (1988, 1991, 1995) Fine Motor Dexterity: Self-reported ability to write/handle small objects and strength o f grip on a dynamometer, MacArthur Study (1988, 1991, 1995) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES (Continued) PAGE Figure IV-8. Fine Motor Dexterity: Self-reported ability to write/handle small 128 objects and performance on timed hand signature, MacArthur Study (1988, 1991, 1995) Figure IV-9. Validity in a self-report measure 139 Figure IV-10. Validity o f change in a self-reported measure 139 Figure VII-1. An alternative theoretical model o f physical dysfunction 243 x Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table II-1. Table 11-2. Table 11-3. Table III-1. Table III-2. Table III-3. Table III-4. Table III-5. Table IV -1. Table IV-2. Table IV-3. Table IV-4. Table IV-5. LIST OF TABLES PAGE Items assessed on mobility measures 20 (from Duncan and Sudenski, 1994) ICIDH-2 draft one digit level classification 39 (from Simeonsson et al.. 2000) Summary o f research studies by data source, measures, 45 method, and findings Descriptive characteristics o f performance-based measures o f 108 physical functioning for all three years (1988. 1991. 995) Descriptive characteristics of self-reported measures of physical 110 functioning Descriptive characteristics of performance-based measures o f 111 physical functioning Descriptive characteristics of change (1988 to 1995) in 112 self-reported and performance-based measures o f physical functioning (n=722) Descriptive characteristics of those included and excluded in the study 115 Descriptive statistics for performance-based measures in 1988 130 Descriptive statistics for performance-based measures in 1995 130 Descriptive statistics for performance-based measures o f physical 132 functioning in 1988, 1995, and 1988 to 1995 Descriptive statistics for self-reported measures of physical 133 functioning in 1988, 1995, and 1988 to 1995 Cross-tabulations between baseline levels o f health 135 conditions/events and the three dependent variables o f interest: 1) over-reporting the onset of disability walking; 2) over-reporting the onset of disability toileting (with grip strength); and 3) over-reporting the onset o f disability toileting (with hand signature). xi Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES (Continued) Table IV-6. Table IV-7. Table IV-8. Table IV-9. Table IV-10. Table IV-11. Table IV -12. Table IV-13. Table IV -14. Table IV -15. Table V -l. Table V-2. Table V-3. Agreement between reported dysfunction and observed performance o f gait function in 1995 Agreement between reported dysfunction and observed performance o f lower body strength in 1995 Agreement between reported dysfunction and observed performance o f fine motor dexterity o f gross motor function, hand signature in 1995 Agreement between reported dysfunction and observed performance o f fine motor dexterity of gross motor function, grip strength in 1995 Agreement between reported dysfunction and observed performance o f gait function (1988-1995) Agreement between reported dysfunction and observed performance o f lower body strength (1988-1995) Agreement between reported dysfunction and observed performance o f fine motor dexterity o f gross function, hand signature (1988-1995) Agreement between reported dysfunction and observed performance o f fine motor dexterity o f gross function, grip strength (1988-1995) Correlations between hypothesized covariates and over-reporting onset o f physical limitations Logistic regression o f over-reporting the onset of physical limitations (1988-1995) Descriptive statistics of physical functioning for the study Descriptive characteristics o f the study Mean and standard deviation o f physical performance by the number o f ADL items individuals report being unable/ need help with in 1995 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES (Continued) Table V-4. Table V-5. Table V-6. Table V-7. Table V-8. Table V-9. Table V-10. Table V -U . Mean and standard deviation o f physical performance by the number o f gross motor items individuals report being unable with in 1995 Mean and standard deviation o f physical performance by the number o f physical activity items individuals report being unable with in 1995 Percentage of explained variance in self-reported performance-based measures o f physical functioning and explained variance added by groups o f demographic, cognitive, health, and behavioral measures in 1995 Unstandardized, standardized regression coefficients and standard 182 errors for individual demographic, cognitive, health, and behavioral characteristics with summary self-report levels o f dysfunction as the outcome and controlling for performance- based levels in 1995 Correlation coefficients between absolute residuals (obtained with 183 regression analysis o f self-report and performance-based measures) and demographic, cognitive, health, and behavioral characteristics in 1995 Mean and standard deviation score of change in physical 184 performance by the number o f ADL items individuals report becoming unable/need help with from 1988 to 1995 Mean and standard deviation score o f change in physical 184 performance by the number o f gross motor items individuals report becoming unable with from 1988 to 1995 Mean and standard deviation score o f change in physical 184 performance by the number o f physical activity items individuals report becoming unable/need help with from 1988 to 1995 PAGE 179 179 180 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES (Continued) PAGE Table V -l 2. Percentage of explained variance o f change in self-report by 185 change in performance-based measures o f physical functioning and explained variance added by change in cognitive functioning and baseline measures o f demographic, cognitive, depressive, health, behavioral, and psychological characteristics from 1988 to 1995 Table V-13. Unstandardized, standardized regression coefficients and standard 187 errors for individual demographic, cognitive, depressive, health, behavioral, and psychological characteristics with change in summary self-report levels o f dysfunction as the outcome and controlling for change in performance-based levels Table V -l4. Correlation coefficients between absolute residuals (obtained with 189 regression analysis o f change in self-report and change in performance-based measures) and demographic, cognitive, depressive, health, behavioral, and psychological characteristics from 1988 to 1995 Table V I-1. Sample size for each physical performance exercise 197 Table VI-2. Percentage of persons who become "unable" to perform specific tasks 204 Table VI-3. Summary of values assigned to persons who became "unable" on respective measures 207 Table VI-4. Strengths and weaknesses o f the comparative analytical strategies 210 Table VI-5. Predictors o f change on 5 repeated chair stands 211-212 Table VI-6. Predictors of change on foot tapping 213-214 Table VI-7. Predictors of change on turning in a circle 215-216 Table VI-8. Predictors o f change on walking at a normal pace 217-218 Table VI-9. Predictors o f change on walking at a fast pace 219-220 XIV Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES (Continued) PAGE Table VI- Table VI- 0. Predictors o f change in performance-based measures where decrement is based on lower values at follow-up 1. Predictors o f change in performance-based measures where decrement is based on higher values at follow-up 223-224 227-228 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT This study employs a longitudinal community based data o f older Americans to examine the measurement o f physical functioning. In this study, the association between individual and summary measures o f self-reported and performance-based measures o f physical functioning is examined. In addition, the demographic, health, psychological, and behavioral characteristics that contribute to their observed discordance are identified. Finally, this study investigates whether change in cognitive performance is related to decrement on more physically demanding or novel tasks. A combination o f approaches suggest that: a) although a relatively weak longitudinal association exists between individual items selected on an a priori basis, a moderate longitudinal association exists between summary measures of three domains o f reported (ADL, gross mobility, and physical activity) and performance-based measures o f physical functioning; b) the observed discrepancies within these different domains o f reported functioning are differentially influenced by demographic, health, psychological, and behavioral characteristics; and c) change in cognitive performance mirrors change on physical performance-based measures, irrespective o f the type of functioning. The findings from this investigation suggest that researchers who rely on self-reported measures to assess the physical functioning o f older persons should consider the demographic, health, psychological and behavioral characteristics o f their subjects because these characteristics may influence these self-reported measures. Furthermore, specific preventive strategies directed toward functional limitations associated with cognitive impairment may serve to improve an individual’s intrinsic ability to function. xvi Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter I Introduction For more than 37 years, researchers have focused on the assessment of physical functioning among older persons (Katz et al„ 1963). Measuring functional ability, however, is complex and a “gold standard" for measurement o f physical functioning does not exist. Since 1963 when McBride employed an assessment tool that evaluated tissue damage, behavioral characteristics o f a body part or organ, behaviors o f the individual, and interactions with the person's environment, the measurement o f physical functioning has been complex (McBride, 1963). This complexity has perpetuated due to the explosion o f programs and research interests with varied purposes and goals. For instance, program administrators use this information in different settings to administer their varied programs and services (Halpem and Fuhrer. 1984). In addition, researchers with disparate interests (e.g.. basic disease, injury, disability, and comprehensive data collection) have employed diverse measures in their respective studies. Thus, the proliferation o f functional assessment measures has created several problems: 1) poor ability to communicate among professional researchers, practitioners, and program directors; 2) lack o f comparability among research efforts; and 3) the frustration among policy makers who are confused about the inability to consistently describe the functional status o f citizens. Several theoretical models have been developed to describe how individuals reach progressively poorer states o f physical functioning. The original and recently revised models proposed by the World Health Organization (the International Classification o f Impairments, Disabilities, and Handicaps, WHO 1980 and 1999), the Nagi l Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. framework (1965, 1969, 1976,1977), and the Disablement Process by Verbrugge and Jette (1994). all define specific pathways through which individuals progress toward physical dysfunction. When operationalizing these models, however, it remains unclear whether to consider self-reported measures o f physical functioning equivalent to performance-based measures of physical functioning. The present investigation thereby treates what the ICIDH (1980) calls “disability” and what Nagi (1965, 1969. 1976. 1977) and Verbrugge and Jette (1994) call “functional limitations" as the same concept and compares self-reported measures to performance-based measures o f physical functioning. Although existing studies suggest that a moderate association exists between reported and performance based measures in the cross-section (Jette and Branch. 1985: Kelly-Hayes et al.. 1992; Sager et al., 1992: Myers etal.. 1993: Cress et al„ 1995: Reuben et al., 1995; Kempen, Van Heuvelen et al.. 1996; Kempen. Steverink et al.. 1996: Merril et al., 1997; Ferrer et al.. 1999) and longitudinally (Myers et al.. 1993; Mendes de Leon et al., 1996; Kempen et al., 1999), the direction o f the disagreement remains unclear. In addition, factors that contribute their discordance need further investigation. For instance although demographic, health, and psychological characteristics have been identified as sources o f discrepancies, contradictions exist specific to the direction o f the discrepancy. For instance, among community residents 63 to 94 years o f age, Kelly- Hayes et al. (1992) showed that individuals 75 and older over over-report (i.e.. higher reported levels o f dysfunction than performance levels) compared to younger individuals. Although this finding was confirmed by Kempen. Van Heuvelen et al., (1996), Ferrer et * al. (1999) showed that individuals 75 and older under-report (i.e., lower reported levels o f dysfunction than performance levels) compared to younger community Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. residents (72 to 74 years o f age). Furthermore, although studies have shown that behavioral characteristics (e.g., body mass, smoking cigarettes, and alcohol consumption) influence various measures o f physical functioning, we have yet to examine whether these behavioral characteristics influence the discordance between reported and performance-based measures o f physical functioning. A. Focus o f Study The studies examining the potential sources of discrepancies between self- reported and performance-based measures of physical functioning leave many questions unanswered. Although they suggest that demographic, health, and psychological attributes account for the discrepancies between reported and performed measures o f physical functioning, information is needed on how these characteristics influence discrepancies between change in reported and performed measures o f physical functioning. Although this information has recently been provided for depressive symptomatology (Kempen, Sullivan et al. 1999), the first aim o f this investigation is ask: 1) What is the concordance between change in reported and performed measures of physical functioning?; and 2) How do demographic, health, behavioral, and psychological characteristics influence the observed discrepancies between change in reported and preformed measures o f physical functioning? Furthermore, to the author’s knowledge, we have yet to clearly understand how specific factors influence discrepancies within different domains o f reported functioning. Three commonly used domains o f reported functioning are: 1) activities o f daily living, ADL (Katz et al., 1963); 2) gross motor functioning (Rosow and Breslau, 1963); 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and 3) what Allaire et al. (1990) calls “physical activity” (Nagi 1976). As reported measures are more commonly used than performance-based measures to assess the physical functioning o f older (and younger) persons, the information collected from such an investigation could be used to improve our selection o f the best domain of reported functioning suited for specific circumstances. Alternatively, if restricted to a specific domain o f reported physical functioning, this information could serve to guide professionals on how specific factors may confound their results. This investigation thereby asks: How do specific factors influence discrepancies (between reported and performed measures of physical functioning) within different domains o f reported physical functioning? Two o f the most feared states o f the elderly are cognitive and physical dysfunction because they can lead to both social and physical dependency. We have yet to examine, however, whether change in cognitive performance is related to decrement on more physically demanding or novel tasks. Research suggests that performances on many information-processing tasks assessing cognitive or perceptual process is slowed with age (Cerella, 1985). Based on the findings by Kaufman and colleagues (1989) that aging seems to decrease fluid intelligence (e.g., tasks requiring sensorimotor coordination, new learning, and speedy performance— Cattell, 1943) while preserving crystallized intelligence (e.g., tasks requiring language skills and the use o f established habits— Cattell, 1943); Albert and Kaplan (1980) suggest that decreased fluid intelligence is related to deficits in functions of initiative, flexibility and planning. Although untested, Barberger-Gateau and Fabrigoule (1997) propose a classification scheme for the physical performance o f older persons: a loss in efficiency for attentional 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. demanding or novel tasks; and good performance on more automatic or over-learned skills. Consequently, the final question o f this investigation asks: How does change in cognitive performance relate to change on an array o f physical exercises? B. Significance o f Contribution The present study explores these research questions by using the MacArthur Research Network on Successful Aging Community Study, a longitudinal, community- based study o f older Americans (70-79 at baseline). Although comparative studies between self-reported and performance-based measures exist for populations in the United States. Canada, the Netherlands, and Spain, the vast majority of these studies have relied on cross-sectional data. With the availability o f repeated measures from the MacArthur Study, the analyses presented here on the change in reported and performed measures o f physical functioning will provide more informative answers about the physical functioning o f older persons. The findings from this investigation will provide insight toward our understanding o f the observed discrepancies between reported and performed measures o f physical functioning. This information will serve to: 1) guide researchers in selecting measures to assess the physical functioning among older persons; 2) improve our ability to describe the functional status o f older citizens; 3) assist clinicians and other health professionals in targeting prevention programs that can slow or postpone the progression o f physical dysfunction; and 4) guide the development of alternative theoretical models o f disability. From this information, it is anticipated that the problems resulting from the proliferation o f functional assessment measures will be mitigated. For instance, 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. by providing a clearer understanding o f what factors contribute to the discrepancies between self- reported and performance-based measures o f physical functioning, the ability to communicate among professional researchers, practitioners, program directors, and policy-makers will improve. Furthermore, although the findings from this investigation will not deter the use o f different types of functional assessment measures, it is anticipated that researchers will gain insight on why comparability is not found among disparate studies examining the physical functioning o f older persons. C. Overview o f Chapters This study is primarily interested in developing a better understanding o f the relationship between self-report and performance-based measures of physical functioning among a community based sample o f older Americans. Issues specific to the research questions are addressed in the following chapters. A brief description of these chapters is presented below to guide the reader. Chapter II Conceptual Framework and Literature Review This chapter begins by providing a history o f the field o f functional assessment beginning in the mid-1800s when research focused on measuring physical limitations in the general population. Alternative types o f self-reported and performance-based measures o f physical functioning are then presented with reference to their strengths and weaknesses. Four major conceptual frameworks o f disability are provided to better understand the physical constructs being measured. These theoretical models include the: 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Handicaps (ICIDH, WHO, 1980), 2) Nagi framework (Nagi, 1965, 1969, 1976, 1977), 3) Disablement Process (Verbugge and Jette, 1994), and 4) recently revised draft version o f the ICIDH (ICIDH-2). Existing studies examining the relationship between self- reported and performance-based measures of physical functioning are then reviewed. The potential sources (e.g. demographic, health, psychological, and behavioral characteristics) for the observed discordance between self-reported and performance-based measures are also identified. The present investigation also considers the relationship between cognitive impairment and physical functioning. Consequently, studies investigating the relationship between cognitive impairment and physical functioning are reviewed; as are the risk factors associated with this relationship. Chapter III Description o f Data and Measures This chapter presents a detailed description o f the data and sample used for the present study. The data used in this study come from the MacArthur Research Network on Successful Aging Community Study, a longitudinal, three-site cohort study o f high functioning, disability free Americans 70-79 in 1988. These persons were followed over seven years and re-interviewed in 1991 and 1995. Detailed information is provided on how constructs o f interest are measured and descriptive statistics specific to these constructs are shown. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter IV Discrepancies Between Self-Reported and Performance-Based Measures o f Physical Functioning The purpose o f this chapter is to further our understanding of the longitudinal association between self-reported and performance-based measures o f physical functioning and identify sources o f discrepancies. As the MacArthur Study was not originally designed to examine the concordance between self-reported and performance- based measures, comparative self-reported and performance-based measures are not one o f the same. Thus, in order to select individual measures of physical functioning for comparison, descriptive statistics are presented for seven a priori matched pairs o f self- reported and performance-based measures. The accuracy of the change in reported functional problems is estimated by calculating sensitivity and specificity for the comparisons selected. Multivariate analyses are then used to test the effect of demographic, health, behavioral, and psychological characteristics on the observed discrepancies between change in self-reported and performance-based measures. Chapter V Sources o f Discrepancies Within Different Domains of Self-Reported Measures o f Physical Functioning This chapter explores whether different determinants influence observed discrepancies within different domains o f physical functioning. The three domains o f reported physical functioning examined are: 1) activities o f daily living, ADL (Katz et al. 1963). 2) gross motor functioning (Rosow and Breslau. 1963). and 3) what Allaire et al (1990) calls "physical activity” (Nagi 1976). This chapter focuses on the demographic, health, behavioral, and psychological characteristics responsible for the observed 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. discrepancies between these domains o f reported functioning and performance-based measures o f physical functioning. Multivariate analyses are conducted both in the cross- section and over time. Chapter VI The Effect o f Change in Cognitive Performance on Change in Performance-Based Measures of Physical Functioning This chapter examines whether change in cognitive performance is related to decrement on more physically demanding or novel tasks. Focus is placed on the cognitive, health, behavioral and demographic characteristics predictive o f change on twelve physical performance exercises. Using the MacArthur Study, the classification proposed by Barberger-Gateau and Fabrigoule (1997) is tested. This classification suggests that: a loss in efficiency for attentional demanding or novel tasks and good performance on more automatic or over-learned skills. Consequently, risk factors for older persons declining on specific types o f physical performance activities are identified. Chapter Vll Conclusion This chapter, summarizes the research findings and provides direction for future studies. Specific findings are discussed in a summary organized around the questions that have guided this study. Furthermore, the author suggests an alternative conceptual model, based on the well accepted models o f Wood (ICIDH, 1980), Nagi (1965. 1966. 1976. 1977) and Verbrugge and Jette (1994) for future researchers to consider when describing the process by which individuals become disabled. This chapter concludes with recommendations for future research. These proposed studies should serve to 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. improve the ability o f public health professionals to target programs and services aimed at curtailing the onset of physical disability. 10 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter II Conceptual Framework and Literature Review A. Introduction Relatively little is known about the relationship between self-reported and performance-based measures o f physical functioning. The aim o f this chapter is to establish the conceptual framework to examine this relationship. First, a brief history of the field o f functional assessment is provided and the term functional assessment is defined. Subsequently, the strengths and weaknesses o f reported and performed measures are discussed. Four major conceptual frameworks o f physical dysfunction are then presented to better understand the constructs of physical functioning. These theoretical models include the: 1) International Classification o f Impairments. Disabilities, and Handicaps (ICIDH, WHO, 1980), 2) Nagi framework (Nagi, 1965. 1969. 1976. 1977). 3) Disablement Process (Verbugge and Jette. 1994), and 4) recently revised draft version o f the ICIDH (ICIDH-2). Existing studies examining the relationship between self- reported and performance-based measures o f physical functioning are then reviewed with particular attention to the factors shown to influence the discordance between these types o f measures. Although we have yet to examine whether health behaviors, such as alcohol consumption, smoking behavior, and body mass, influence the discordance between reported and performed functioning, rationale for their inclusion in this investigation is also provided. The present investigation also considers the relationship between cognitive impairment and physical functioning. Consequently, studies investigating this relationship are reviewed; as are the risk factors associated with this relationship. ll Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Finally, this chapter summarizes the content o f the chapter and asks four research questions specific to the investigation. B. History o f Functional Assessment Research in the m id-I800s focused on measuring physical limitations due to impairment by identifying and classifying the prevalence o f functional limitations in the general population (U.S. Department o f Education, 1982; Wood and Badley, 1980). These measures relied on health surveys that examined restriction in activity due to physical impairment. With the passage o f the Federal Employees Compensation Act in 1908, civil employees received medical benefits and reimbursement for any temporary or permanent physical inability to return to work. As a result of this legislation, attempts to measure residual functioning following an injury or chronic disease began. The primary purposes o f assessment was measuring the loss o f function. Between 1930 and 1958 there were many assessment procedures developed to measure function (Frey, 1984). Kessler (1931) developed the first set of standards to measure functional abilities. For instance, he suggested using physical units (pounds or degrees) to describe muscle strength and range of motion. In 1963. McBride offered a 10-factor system organized in two major sections: disabling physical impairments and disabling functional deficiencies: Disabling Physical Impairments: 1. The anatomical and physiological mass tissue damage resulting in limited motion (weakened back, osteophytosis, fibrosis) 2. The clinical manifestations (pain, tenderness, fatigue) 3. The restrictions toward work restoration (lifting, stooping, pulling, pushing, etc.) 4. The restrictions related to working conditions (coldness, dampness, irregular hours, etc.) 12 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5. The intangibles o f reactionary interference with recovery (limited opportunities for rehabilitation, unpredictable aggravation, progressive degenerative changes, recover}' slowed by age) McBride, 1963. p. 43 Disabling Functional Deficiencies: 1. Quickness o f action 2. Coordination o f skills 3. Strength, stability 4. Security, self-confidence 5. Endurance McBride, 1963. p. 43 From 1940-1960. three legislative changes contributed to the surge o f functional assessment activity. In 1943 and then in 1954. vocational rehabilitation-related legislation influenced the development o f functional assessment methodologies. The Barden-LaFollette Act of 1943 extended the quantity and quality of rehabilitation services to the physically and mentally impaired by allowing for the inclusion of provisions for the rehabilitation o f disabled veterans under the Veterans Administration. In 1954. the Vocational Rehabilitation Act Amendment (PL 83-565) and the Medical Facilities Survey and Construction Act (PL 83-482) allowed for a 300% increase in the capacity of client services, planing and construction grants for rehabilitation facilities and training grants for developing professional capacity in counseling, medicine, nursing, physical therapy, occupational therapy, psychology, and social work (Wessen. 1965). Finally, the Social Security Amendments o f 1956 (PL 84-880) required medical certification o f disability for certain social security programs. These legislative changes thereby acted to improve various aspects o f an individual's aptitudes: physical, social, psychological, and vocational. 13 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Functional assessment, during this period, focused on daily activity measurements. In 1945, Deaver coined the term "activities o f daily living" (ADL) as the wide range o f behavior patterns considered necessary for a physically impaired person to meet the demands o f daily life (U.S. Department of Education. 1982). These behaviors included an individual's ability to eat, drink, toilet, dress, bathe, use the toilet, and walk. Unfortunately, however, as both medical and non-medical professionals attempted to develop standards for assessment, confusion resulted. Bruette and Overs (1968) reviewed twelve ADL scales that were developed between 1951 and 1966. They found that no single ADL appeared across all twelve scales. In 1973. Donaldson et al. found that only one activity (dressing) appeared across the twenty-five scales they reviewed. Hedrick et al. (1981) found that among eleven scales, only three activities (eating, dressing, and bathing) were represented. From 1960 to 1983. federal legislation in rehabilitation significantly influenced the development of functional assessment. The Office of Technology and Assessment (U.S. Congress, 1982) reported that by 1982 there were over 100 federal programs serving the disabled population. These programs covered income maintenance, health and medical care, social services, educational services, and vocational rehabilitation or independent living (Frey, 1984). As the number of these programs have grown, so too has the popularity o f functional assessment. C. Types o f Functional Assessment: Their Advantages and Disadvantages The field o f functional assessment is permeated by complexity. Since 1963 when McBride employed an assessment tool that evaluated tissue damage, behavioral 14 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. characteristics o f a body part or organ, behaviors of the individual, and interactions with the persons’ environment, the measurement o f physical functioning has been complex. This complexity has perpetuated due to the explosion of programs and research interests with varied purposes and goals (Halpem and Fuhrer. 1984). A possible rationale for the imprecision found in the field might be attributable to the lack o f a concise and commonly accepted definition of the term “functional assessment." Again, perhaps this is due to the array of users who employ the term. These users include policymakers, program administrators, researchers, service providers, evaluators, and those evaluated. Each o f these individuals may interpret and use the information collected from an assessment in distinct ways. Halpem and Fuhrer's (1984) definition o f functional assessment is chosen because of its inclusion o f seven key elements. They defined functional assessment as “the measurement o f purposeful behavior in interaction with the environment, which is interpreted according to the assessment's intended uses” (Halpem and Fuhrer. 1984. p.3). The seven key components in their definition o f functional assessment are: measurement, purposeful, behavior, environment, interpreted, and intended use. • Measurement includes reported and observed or tested performance collected to assess physical functioning. • The term “purposeful” suggests that the individual collecting the information has emitted a goal or objective. • Behavior is the entity being measured. • The environment plays an important role in shaping an individual's behavior. It is therefore important to refer to the places where the behavior is taking place. • The interaction between an individual’s behavior and the environment in a functional assessment should also be considered. 1 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. • The interpretation of the information collected should regard issues of reliability and validity when choosing the application or use of this information because alternative application of this information may require different data collection. Halpem and Fuhrer, 1984, p. 3 As Halpem and Fuhrer (1984) suggested, an integral component to functional assessment is measurement, either reported or observed. The activities of daily living (sometimes referred to as the ADLs or BADLs-basic activities of daily living) were first proposed in 1959 by Katz and his colleagues (Katz et al., 1959). These measures were originally developed as benchmarks for gauging the effects of rehabilitation with post stroke patients (McDowell and Newell, 1996). Since their original development, a number of scales have been developed to measure ADLs in different formats (self-report, proxy-report. and objective evaluation). Other self-reported measures o f physical functioning include measures of instrumental activities o f daily living (Lawton and Brody. 1969). gross functional mobility (Rosow and Breslau. 1966) and what Allaire et al. (1990) calls "physical activity'' (Nagi, 1976) measures o f physical activity. The Rosow and Bresalu (1966) gross-motor scale contains three items: working around the house, walking up/down a flight o f stairs, and walking half a mile. Items on the Nagi scale (1976) include pushing/pulling large objects, stooping/crouching/kneeling, carrying ten pounds or more, reaching/extending arms above shoulder, and writing/handling small objects. There has also been a tremendous surge in the development o f objective or performance-based measures. Functional tests to assess individual mobility and balance have been developed by many researchers (Deeg et al., 1994; Duncan and Studenski et al., 1994; Winograd et al., 1994; Guralnik et al., 1994; Reuben and Siu, 1990; 16 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Granger et al., 1986; Mathias et al., 1986; Tinetti et al.. 1986; Berg et al.. 1989). Although some o f these measures rely on standardized metrics (seconds, kilograms) to identify the level o f functional impairment, some measures also rely on an observers assessment o f the level of difficulty a subject has with activities of daily living (a lot. a little, some, or no difficulty). Functional tests to assess individual mobility and balance have been developed and used by investigators to evaluate stability (Duncan and Studenski et al., 1994). These test include the Get Up and Go Test (Mathias et al.. 1986). the Tinetti Performance- Oriented Mobility Assessment (Tinetti, 1986). the Berg Balance Scale (Berg et al.. 1989). the Duke Progressive Mobility Skills Test (Hogue et al., 1990). the Physical Performance Mobility Examination (Winograd et al.. 1994). and the Established Population for Epidemilogic Studies o f the Elderly (EPESE) Mobility Assessment (Grualnik et al.. 1993). Duncan and Studenski (1994) provided an excellent outline o f these tests which is replicated here and shown in Table II-l. The Get Up and Go Test Mathias. Mayak. and Isaacs( 1986) developed the Get Up and Go Test to identify balance impairment in the elderly. This test requires that subjects stand up from a chair, walk a short distance, turn around, return and sit down. Performance on these functional tasks is graded on a 5-point ordinal scale. The authors reported that poor performance on these items may b due to impairment o f balance. Interrater reliability was established for this test, and test function was correlated with sway and other clinical measures o f balance. The test-retest reliability, sensitive to change, and predictive validity o f the Get Up and Go Test have not been established. Podsiadlo and Richardson (1991) modified the Get Up and Go Test by timing performance rather than by assessing quality of performance (see Table II-1). It takes approximately five minutes to administer. The Tinetti Performance-Oriented Assessment o f Mobility This test is a performance-based measure o f balance and gait. Tinetti's (1986) original balance assessment included 15 performance activities which were rated dichotomously as normal or abnormal. The Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tinetti measure has recently been expanded to include the items listed in Table II-1. These activities are scored on a 3-point ordinal scale. The Tinetti measure is reliable, valid and predictive of falls, fall injuries, and nursing home placement. The sensitivity of the measure to change is currently being investigated. [On average, this test requires twenty minutes to perform.] The Berg Balance Scale Berg et al. (1989) developed a balance measure to assess function in elderly subjects. It includes 14 items (see Table II-1) and is scored using a 5-poing ordinal scale. The reliability and validity o f the measure has been assessed in several different populations. It has been compared with laboratory measures o f postural sway and other clinical measure o f balance and mobility in 31 elderly subjects (Berg et al., 1992). It correlates with other balance measures (Tinetti, Barthel Mobility Index, and the Timed Get Up and Go Test) as well as with motor function. Both intertester and test-retest reliability have been established. In two longitudinal studies o f 114 elderly subjects and 70 acute stroke patients the Berg Balance assessment was predictive o f falls and could discriminate subjects by their use o f assistive devices; it was also sensitive to changes in functional status (Berg, 1992). The measure is simple, can be performed in any setting and takes an average o f 10 to 15 minutes to administer. Although the measure has only been administered by professional staff, it could be performed buy nonprofessional staff with minimal training. The test probably has a ceiling effect in fit elderly subjects. The Duke Progressive Mobility Skills Test Thirteen mobility items that capture static and dynamic balance comprise this test (see Table II-l) (Hogue et al.. 1990). The items are scored on a 3- point ordinal scale and are hierarchically organized...The measure is simple, requires approximately 10 minutes to administer and may be administered in any environment. It may be administered by nonprofessional staff, but training is required. The Mobility Skills may not be able to discriminate function among fit older adults. The Physical Performance and Mobility Examination (PPME) The PPME (Winograd et al., 1994) tests bed mobility, transfer skill, multiple stands from a chair standing balance, step-up, and ambulation, see Table II-l. The scoring is either pass-fail or a 3-level ordinal scoring. This mobility measure was developed to assess hospitalized older adults. The PPME correlates with physical activities o f daily living and physical performance on the Medical Outcomes Scale (Stewart and Walre, 1992). The sensitivity o f this measure to change has not been assessed. This brief assessment is portable and can be administered by nonprofessional staff. 18 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The EPSE Mobility Assessment Guralnik et al., (1994) have evaluated the use o f a mobility assessment that has been employed in the Established Populations for Epidemiologic Studies of the Elderly (EPESE (See Table II-l). This assessment includes timed side-by-side, semi-tandem, and tandem stands, timed 8-foot walk, and time to rise from chair 5 times. The measure has been employed in over 5,000 community-dwelling individuals over the age of 71. The measure may be scored as continuous (timed) or as categorical variables, using nonperformance as poorest category and quartiles o f performance as four additional categories. The reliability o f the measure is high and it has demonstrated construct validity. The categorical scoring o f tasks is predictive for nursing home admission and death. The instrument takes approximately 10 minutes to administer. It can be administered by nonprofessional staff in any environment. The sensitivity of the measure to change has not been assessed. There are many advantages and disadvantages to both self-reported and performance-based measures that have been documented by previous researchers. Evans (1993, p. 150) outlined four advantages to self-reported measures: 1) a number of scales are available and data on their characteristics have been published for some scales; 2) many scales are suited for administration to a proxy respondent so that persons who are not able to respond themselves are not selectively excluded from assessment; 3) most scales provide a wide or even global assessment o f function and can be administered in a relatively short time; and 4) reliable administration o f these measures requires less training than does administration of direct performance tests of physical function. Self-report measures, however, can be imprecise. For instance, an individual's assessment o f his or her functioning or overall health status can vary from the assessment o f a family member o f professional caregiver (Guralnik. Branch et al.. 1989). Furthermore, Guralnik, Branch et al., (1989) suggested that the accuracy of the measurement can be compromised when these instruments are not clearly defined in terms o f the activity being assessed or the possible response categories. For example, when individuals are asked if they need help bathing, it may not be clear what 19 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le 11-1. Ite m s a s s e s s e d o n m o b ility m e a s u re s (f ro m D u n c a n a n d S u d e n s k i, 1 9 9 4 ) T im e d T in n e tti B e r g B a la n c e D u k e P h y s ic a l P e rfo rm a n c e E s ta b lis h e d P o p u la tio n s G e t U p a n d G o P e rfo rm a n c e - A s s e s s m e n t M o b ility S k ills M o b ility A s s e s s m e n t fo r E p id e m io lo g ic S tu d ie s O r ie n te d A s s e s s m e n t o f th e E ld e rly ( M a th ia s et a l., 1 9 8 6 ) ( T in e tti, 1 9 8 6 ) (B e r g e t a l., 1 9 8 9 ) ( H o g u e e t a l., 1 9 9 0 ) ( W in o g r a d e t a l., 1 9 9 3 ) (G u ra ln ik e t a l., 1 9 9 4 ) G e t u p f r o m c h a ir C h a ir S ittin g to s ta n d in g S ittin g b la n c e B e d m o b ility T im e d s id e -b y s id e s ta n d W a lk 3 m e te rs S ittin g b a la n c e S ta n d in g S ittin g re a c h T r a n s f e r s k ills S e m ita n d e m s ta n d T u r n S it to s ta n d u n s u p p o r te d T r a n s f e r M u ltip le c h a ir s ta n d s T a n d e m s ta n d s W a lk b a c k 3 m e te rs S ta n d to sit S ittin g u n s u p p o r te d R is in g fro m a c h a ir S ta n d in g b a la n c e T im e d 8 - fo o t w a lk s S it d o w n B e d S ta n d in g to s ittin g S ta n d in g b a la n c e S te p - u p T im e d c h a ir ris e S ta n d to sit T r a n s f e r P ic k u p o b je c t A m b u la tio n S it to lie S ta n d in g -e y e s fro m flo o r L ie to sit c lo s e d W a lin g S it to s ta n d S ta n d in g -f e e t T u rn in g S ta n d in g to g e th e r A b r u p t s to p T a n d e m R e a c h in g fo rw a r d O b s ta c le (e y e s o p e n ) R e trie v in g o b je c t S ta n d in g re a c h S e m ita n d e m fro m flo o r S ta irs P u ll a t w a is t T u r n in g to lo o k L e a n b a c k w a r d b e h in d T o e s ta n d in g T u r n in g 3 6 0 O n e let s ta n d d e g re e s G a it— fla t s u rf a c e P la c in g a lte r n a te In itia tio n fo o t P a th d e v ia tio n o n sto o l T u r n in g S ta n d in g M is s e d s te p s o n e fo o t in fro n t S te p o v e r o b s ta c le S ta n d in g o n o n e G a i t — u n e v e n s u rf a c e fo o l In itia tio n P a th d e v ia tio n T u r n in g M is s e d s te p s S te p o v e r o b s ta c le specific tasks within the behavior o f bathing is being specified. Similarly, if individuals are not provided with a set o f guidelines on how to assess the level o f their difficulty, they may have problems determining whether they have a little, some, or a lot of difficulty for a given task such as transferring from a bed to a chair. Individuals who are cognitively impaired may also have difficulty in assessing their physical functioning (Guralnik, Branch et al.. 1989). Individuals must have the cognitive ability to perform these reported tasks (e.g., ''toileting") and the ability to document or report on how well they are able to perform. For this reason, proxies are frequently used to assess the physical functioning o f cognitively impaired measures, when reported measures of physical functioning are implemented. The use o f proxies, however, is not without it's limitations, as well. Proxies can only report on the observed activities and cannot speculate on what the physical functioning of the individual would be like if cognitive impairment was not present. Like self-reported measures, performance-based measures have advantages and disadvantages. Performance-based measures are advantageous in that they have greater face validity and maybe reproducibility (Mungall and Hainsworth. 1979; Guyatt et al.. 1984; Guralnik. Branch et al., 1989; Dorevitch et al.. 1992) because they are relying on objective standards o f measurement (i.e., number of seconds to walk a specified distance or grip strength in kilograms). Whereas reported measures may differ from day to day according to the mood or interpretation of the individual, because performed measures rely on objective standards of measurement, the face validity of these measures is stronger than that o f reported measures. Although not empirically tested, because performance-based measures are repeatedly executed in the same way. it has been 21 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. suggested that these measures also have greater reproducibility than reported instruments. It has been suggested that performance-based measures have greater sensitivity to change and are less influenced by poor cognitive functioning, culture, language and education than self-reported measures (Guralnik. Branch et al.. 1989: Linn et al.. 1980). Unfortunately, however, performance-based measure can require more time, special training o f examiners, adequate space and special equipment (Guralnik, Branch et al.. 1989). In addition, the subject's level o f function, as measured, may not reflect his or her actual performance in daily life (Law and Letts. 1989). Performance measures may be limited by their dependence upon the subject's motivation to perform (Cress et al.. 1995); or the subject's motivation may inadvertently increase their risk of unnecessary harm or injury. Researchers have also compared the use o f reported and performed measures for longitudinal analyses. Branch et a. (1983. 1986) suggests that change in self and proxy reported functioning probably results from some degree o f unreliability in measurement due to the subjective elements intrinsic in self reported items such as the subject's level o f expectations. Thus, researchers tend to prefer performed measures. Performance- based measures tend to avoid this pitfall and assess change over time on a continuous scale (e.g.. number o f seconds, weight in kilograms, etc.) rather than broad categorical changes (e.g.. no difficulty, some difficulty, a lot o f difficulty, cannot do) (Guralnik. Branch et al., 1989: Tinned et al.. 1986). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. D. Conceptual Frameworks The confusion in functional assessment could be due to a lack o f common definitions o f the terms •'impairments,” “disability,” and "handicap.” The Bureau o f Social Science Research, Inc (BSSR, 1981) compared definitions from: 1) the Social Security Act. 2) Section 504 o f the Rehabilitation Act Amendments o f 1978. They note that the Social Security Act considers a person eligible for certain benefits only if the person has participated in gainful employment for one year because o f impairment. The Section 504 regulation defines handicap as a physical or mental impairment that substantially limits major life activities. Thus a person who is viewed as impaired, disabled, or handicapped in one context may not be similarly viewed in another. There have been, however, some attempts to clarify these concepts. The four models reviewed here are: the International Classification o f Impairments, Disabilities and Handicaps (ICIDH), the Nagi framework (1965. 1969. 1976. 1977), and the Disablement Process proposed by Verbrugee and Jette in 1994. and the revised draft o f the ICIDH (ICIDH-2. http://www.who.ch/icidh). 1. The International Classification o f Impairments, Disabilities, and Handicaps (ICIDH) The conceptual framework o f the ICIDH was developed by Dr. Philip Wood (Wood. 1980; Wood and Bradley, 1978, 1981). The four main components o f this classification are disease, impairment, disability and handicap (see Figure II-l). These concepts are defined as (WHO, 1980): Disease: Intrinsic pathology or disorder. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Impairment: Loss or abnormality o f psychological, physiological, or anatomical structure o f function at organ level. Disability: Restriction or lack o f ability to perform an activity in a normal manner. Handicap: Disadvantage due to impairment or disability that limits or prevents fulfillment of a normal role (depends on age, sex, sociocultural factors for the person). For the three major concepts (impairment, disability, and handicap) a taxonomic listing o f entries exists in each chapter of the ICIDH. For instance, in the impairment code, there are 1009 entries intended to document "any loss or abnormality of psychological, physiological, or anatomical structure of functioning (WHO. 1980)." These entries are listed in a hierarchical fashion so that meaning is based on a single digit entry. For a more detailed classification, coding o f a second digit after a decimal point is possible. For instance, 1-56 designates visual field impairment and I 56.5 designates slight impairment o f the visual fields (filed diameter 120 degrees or less). There are 322 items in the disability code. Disabilities were assumed to be manifested along a gradient o f severity that can be coded on a zero to six scale from do disability to complete disability. Finally, the handicap code is oriented to scale the relative disadvantage experienced in survival roles from none (0) to extreme (8) and unspecified (9). In 1993 the ICIDH was updated (Bradley, 1993) and published as a supplement to the International Classification of Diseases (ICD) (WHO, 1993). While the ICIDH was designed to classify the consequences of chronic health conditions (Chapireau and Colvez, 1998), the ICD was developed to explain health problems. In the ICD, diseases, disorders, and complaints are classified according to etiology or organ systems, 2 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. F ig u r e 11-1. T h e I n te rn a tio n a l C la s s if ic a tio n o f Im p a ir m e n ts , D is a b ilitie s , a n d H a n d ic a p s (W H O , 1 9 8 0 ) H A N D IC A P ( d is a d v a n ta g e d u e to im p a irm e n t o r d is a b ility th a t lim its o r p r e v e n ts fu lfillm e n t o f a n o rm a l ro le - d e p e n d s o n a g e , se x s o c io c u ltu ra l fa c to rs fo r th e p e r s o n ) D IS E A S E ( th e in tr in s ic p a th o lo g y o r d is o r d e r ) IM P A I R M E N T (lo s s o r a b n o r m a lity o f p s y c h o lo g ic a l, o r a n a to m ic a l s tr u c tu r e o r fu n c tio n a t o r g a n le v e l) D IS A B IL I T Y - » (r e s tr ic tio n o r la c k o f a b ility to p e r fo rm a n a c tiv ity in n o rm a l m a n n e r) CO <Ol for epidemiological and statistical purposes concerning morbidity and mortality. In contrast, the ICIDH uses concepts to record health problems without taking into consideration their cause (Halbertsma. 1992a: Halbertsma. 1992b). The ICIDH is accepted nearly worldwide as a tool for recognizing and describing disablement (Schuntermann. 1996). It has changed policy, planning and administrative roles of governments, organizations, and individuals to the concepts o f impairment, disability and handicap (Voorwerk. 1993). 2. The Nagi Framework Similar to the ICIDH model developed by Wood (1980). Nagi (1965. 1969, 1976. 1977) developed a model to describe the consequences o f chronic injury and disease. Nagi (1976. 1977) suggested that a conceptual model begins with the disruption of normal body system processes, see Figure II-2. This is called active pathology. When active pathology cannot be cured or controlled by the organisms physiological systems, then impairment results. Impairment refers to a ‘'physiological, anatomical, or mental loss or other abnormality or some combination of these" (Nagi, 1977, p.26). Nagi suggested that although impairment and active pathology can overlap, the distinction between them is clearer when impairment does not have an active pathology associated with it. Thus, although active pathology is always accompanied by impairment, impairment is not necessarily accompanied by pathology. Furthermore, impairments can affect organs, systems, or body parts. A couple o f examples o f impairments include vision loss and restricted range o f arm motion. Nagi considered impairment as lower levels o f functional limitations. Higher levels of functional limitations are defined as the inability o f the organism as a whole to perform activities. The higher level o f 26 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. F ig u r e 11-2. T h e N a g i S c h e m e ( N a g i, 1 9 6 5 , 1 9 6 9 , 1 9 7 6 , 1 9 7 7 ) A C T IV E P A T H O L O G Y (in te r r u p tio n o r in te rf e re n c e w ith a n d e f fo r ts o f th e o r g a n is m to r e g a in n o rm a l s ta te ) IM P A I R M E N T (a n a to m ic a l, p h y s io lo g ic a l, m e n ta l m e n ta l, o r e m o tio n a l a b n o r m a litie s o r lo s s i j -j F U N C T I O N A L L I M I T A T IO N S (lim ita tio n in p e r fo rm a n c e a t th e le v e l o f th e w h o le o r g a n is m o r p e r s o n ) - » D IS A B IL IT Y (lim ita tio n in p e rfo rm a n c e o f s o c ia lly d e fin e d ro le s a n d ta s k s w ith in a s o c io c u ltu ra l a n d p h y s ic a l e n v ir o n m e n t) limitations are called functional limitations. These include walking, climbing, lifting, and reasoning. Thus Nagi distinguished between two separate types o f limitations. When functional limitations result in the inability to perform an expected social role, such as employment, the individual is defined as disabled. Nagi defines disability as " a form of inability or limitation in performing roles and tasks expected o f an individual within a social environment” (Nagi, 1977, p.27). These tasks and roles include self-care, education, family relations, personal relations, recreation, and employment. Nagi pointed out that not every impairment resulted in disability and that similar patterns of disability can result from different functional limitations and impairments. There are some important similarities and differences between the ICIDH and the model developed by Nagi. First, although the specific labels differ somewhat. Nagi and Wood employed similar definitions, as seen in Figure II-3. Disease and active pathology are similar as are the two definitions o f impairment. The Wood definition o f disability is also similar to N agi's definition o f higher functional limitations. Both refer to the consequences o f impairment that are manifested at the level o f the person. Nagi's definition o f disability and W ood's definition of handicap differ, however. Although both consider the effect o f the environment on the individual. Wood refers to a disadvantage placed on the individual by the environment. The individual has relatively little direct control over this variable. Nagi, however, dismissed the idea o f disadvantage by placing emphasis on the individual’s inability to perform up to the expectations in the environment in which he or she lives. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. F ig u r e 11-3. C o m p a ris o n o f th e N a g i a n d th e W o o d m o d e ls C o n te n t D is r u p tio n o f n o rm a l b o d y p r o c e s s e s P h y s io lo g ic a l, a n a to m ic a l, o r p s y c h o lo g ic a l lo ss , a b n o r m a lity o r in ju ry R e s tr ic tio n o r la c k o f a b ility to p e r fo rm a h u m a n a c tiv ity In a b ility o r la c k o f o p p o r tu n ity to p e r fo rm a s o c ia lly e x p e c te d ro le N a g i’s L a b e l A c tiv e P a th o lo g y Im p a ir m e n t F u n c tio n a l L im ita tio n D is a b ility W o o d ’s L a b e l D is e a s e Im p a ir m e n t D is a b ility H a n d ic a p sO 3. The Disablement Process In 1994. Verbrugge and Jette proposed a model called the Disablement Process, see Figure II-4. This model is founded on both the ICIDH and the Nagi theoretical models. The Disablement Process, however, introduces factors that may speed up or slow down the movement along the path toward greater dysfunction. These factors include social, psychological, and environmental forces. The main pathway proposed by Verbrugge and Jette (1994) contains four concepts: pathology, impairments, functional limitations, and disability. Pathology is defined as "the biochemical and physiological abnormalities that are detected and medically labeled as disease, injury or congenital/developmental conditions” (Verbrugge and Jette, 1994. p.3). Examples of pathology include Alzheimer’s disease, osteoarthritis, and cerebral palsy. Impairments are '"dysfunctions and significant structural abnormalities in specific body systems" (Verbrugge and Jette. 1994. p.3). They use the word significant to mean that the abnormality can have consequences for physical, mental, or social functioning. Functional limitations are “restrictions in performing fundamental physical and mental actions used in daily life by one's age-sex group" (Verbrugge and Jette. 1994. p.3). These actions include walking, lifting objects, and climbing stairs. Finally, they define disability as “the difficulty o f doing activities in any domain o f life" (Verbugge and Jette. 1994, p.3). The domains of life they refer to include ADL. IADL. basic activities of daily living (BADL), and other types of activities specific to hobbies, recreational, leisure, religious, and socialization. Verbrugge and Jette (1994) address the similarity between functional limitations and disability by using the words “action” and “activity" coined by LaPlante (1990). They suggest that functional limitations and disability refer to 3 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. F ig u re 11-4 T h e D is a b le m e n t P r o c e s s (V e rb r u g g e a n d J e tte , 1 9 9 4 ) THE MAIN PATHWAY P A T H O L O G Y -» ( d ia g n o s e s o f d is e a s e , in ju ry , c o n g e n ita l/ d e v e lo p m e n ta l c o n d itio n ) I M P A I R M E N T -» ( d y s f u n c tio n s a n d s tr u c tu ra l a b n o rm a litie s in s p e c itlc b o d y s y s te m s : m u s c u lo s k e le ta l, c a r d io v a s c u la r , n e u ro lo g ic a l, e tc .) / R IS K F A C T O R S ( p r e d is p o s in g c h a r a c te r is tic s : d e m o g ra p h ic , s o c ia l, lif e s ty le , b e h a v io r a l, p s y c h o lo g ic a l, e n v ir o n m e n ta l, b io lo g ic a l) E X T R A - I N D I V I D U A L F A C T O R S (m e d ic a l c a r e a n d re h a b ilita tio n ; m e d ic a tio n ; o th e r th e r a p e u tic re g im e n s ; e x te r n a l s u p p o rt; b u ilt, p h y s ic a l, a n d s o c ia l e n v ir o n m e n t) i F U N C T I O N A L L I M I T A T IO N S ( r e s tr ic tio n s in b a s ic p h y s ic a l a n d m e n ta l a c tio n s : a m b u la te , re a c h , s to o p , c lim b s ta irs , p r o d u c e in te llig ib le s p e e c h , e tc .) D IS A B IL I T Y (d if f ic u lty d o in g a c tiv itie s o f d a ily life: j o b , h o u s e h o ld m a n a g e m e n t, p e rs o n a l c a r e , h o b b ie s , a c tiv e re c re a tio n , c lu b s , s o c ia liz in g w ith fr ie n d s a n d k in , c h ild c a r e , e r r a n d s , s le e p , trip s , e tc .) -------------------------- I N T R A - IN D I V I D U A L F A C T O R S (life s ty le a n d b e h a v io r a l c h a n g e s ; p s y c h o s o c ia l a ttr ib u te s a n d c o p in g ; a c tiv ity a c c o m m o d a tio n s ) ■ ‘different behaviors, not different ways o f measuring the same behavior’’ (Verbrugge and Jette. 1994, p.5). In addition to the main pathway from pathology to disability, the Disablement Process, described by Verbrugge and Jette (1994). includes other social and psychological concepts that contribute to the process. Risk factors are defined as "...dem ographic, social, life-style, behavioral, psychological, environmental, and biological characteristics of an individual that can affect the presence and severity of impairment, functional limitations and disability” (Verbrugge and Jette. 1994. p. 8). These factors are pre-disposing in that they exist at or before the outset o f the disablement process. Furthermore, these characteristics are usually permanent qualities of the individual because they "...prom pt chronic conditions and enduring impacts" (Verbrugge and Jette. 1994. p.8). Actions taken in response to disease or dysfunction can also reduce or increase the progression toward disability. These factors are defined as interventions, and exacerbators. Interventions made by individuals or others serve as “buffers.” These interventions include "medical care and rehabilitation, medications and other therapeutic regimens, external supports (personal assistance, special equipment and devices), modifications, the built/physical/social environment, lifestyle and behavior changes, psychosocial attributes, and coping, and activity accommodations" (Verbrugge and Jette. 1994. p.8). These factors are not predisposing but are inserted during the disablement process in the effort to slow or reverse the process. Furthermore, the timing o f their effects “may be immediate, delayed or cum ulative...thereby making the effects of specific interventions problematic in non-experimental research” (Verbrugge and 12 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Jette, 1994, p.8). Verbrugge and Jette go on to suggest that these interventions can operate from within the individual. "intra-individual." or outside the individual, "extra- individual.1 ' Thus, interventions can act on any o f the four main features o f the main pathway. For example, while medical interventions influence pathology and impairments; health interventions influence functional limitations and disability; and social interventions influence disability. Exacerbtors can promote or maintain dysfunction in several ways. Verbrugge and Jette (1994) suggest that exacerbators may happen in three ways. First, interventions, such as surgery, can make matters worse. In addition, individuals may adopt behaviors in response to their dysfunction that may aggravate their limitations or disability. For instance, individuals may increase their alcohol intake. These two types o f exacerbators are surface in response to the disablement process. Society also places impediments on limited or disabled individuals. For example, inflexible work hours or architectural barriers. Verbrugge and Jette (1994) refer to this final type o f exacerbators as the essence of "handicap'1 or "social disadvantage.1 1 This type ofexacerbator is considered predisposing. Overall, all three types o f exacerbators have negative consequences for functioning. 4. The Revised ICIDH (ICIDH-2) The ICIDH has recently been revised in the effort to address some of the theoretical problems concerning the concepts and terms of the classification (Wiersma. 1986; Chapireau, 1994; Foulgeyrollas 1993a; Foulgeyrollas 1993b; Fougeyrollas 1995). First, three criticism are outlined and then the revised draft version o f the ICIDH 33 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (ICIDH-2) is presented. The three criticisms discussed here concern the: 1) the definition of the term "handicap”, 2) the explicit neglect o f environmental factors in the classification, and 3) the causal interpretation o f the theoretical model. The Definition of the Term Handicap. The WHO defined handicap as "a disadvantage (WHO, 1980, p.30).” It was described as a separate level because: For instance, one individual with rheumatoid arthritis may be only mildly disabled and yet at a sever disadvantage, whereas another person with the same disease who is much more severely disabled may, perhaps because o f greater support from the family or social network, experience considerably less disadvantage. If handicap is the prime area of social concern, not all those with activity restriction are necessarily at a disadvantage...and methodological obstacles need not compel social action on disadvantage to be determined by measures o f disability alone. WHO, 1980, p. 41 As many have stated, handicaps seem merely complex disabilities (Haber. 1985; Wiersma. 1986; Colverz and Robine. 1986: Grimby et al.. 1988; Orgogozo. 1994). Wiersma suggests that: "In essence, it is a different way o f conceptualizing and operationalizing the same thing (Wiersma, 1996 p. 102)." More specifically, Wiersma (1986) stated that an overlap between the concepts o f disability and handicap exist: To quite a degree, the handicap section o f the ICIDH is a summary of parts o f the impairment and disability sections. Two difficulties deserve emphasis. First the ICIDH states that the concepts o f handicap is associated with that o f values, while implying that disability is not so related. The question then arises whether disability, particularly disabilities in relations, are free o f value. The answer is N o... Secondly, the ICIDH describes handicap as a social phenomenon and the classification scheme as being directed not to individuals or their attributes but to circumstances... Yet the details of the handicap dimensions do not refer to circumstances but explicitly to the individuals' abilities and competence. Wiersma. 1986, p.4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Contrary to this position, others have suggested that disability is different from handicap. Fougelloras (1993b) proposed, that a person with a health problem has no handicap if this person is integrated in all the areas o f life they wish to be integrated in. It is only if this integration is not possible that a handicap is realized. Therefore, if we have to decide whether a person has or does not have a handicap, we have to look at the set of his/her areas o f life to determine in what areas o f their life they are not integrated. Thus, a handicap is attributed to an individual's area o f life (Schuntermann. 1996). Handicap is characterized by a discordance between the individual's performance or status and the expectations o f the particular group of which he is a member. Handicap is thus a social phenomenon, representing the social and environmental consequences for the individuals stemming from the presence of impairment and disabilities. WHO. 1980 The Explicit Neglect o f Environmental Factors. The classification has also been criticized because o f its explicit neglect of the environmental factors of disablement. Fougeyrollas (1995) states: "The ICIDH model graphic diagram did not include the environment factors and thus handicap was still considered or understood in a medical perspective as an individual characteristic." Fougeyrollas proposed that while disabilities can be assessed in the environment o f functional rehabilitation, the intervention for handicaps must occur in the real world environment (Whiteneck and Fougeyrollas. 1996: Fougeyrollas, 1996). Chapireau and Covez (1998) suggest that Fougeyrolla's criticism is due to the misunderstanding about the definition o f handicap. They state: "When seen as a disadvantage, it [handicap] appears to be the end result of several factors, including the environment” (Chapireau and Covez, 1998, p.63). The WHO manual states: 35 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. handicap thus reflects interaction with and adaptation to the individual’s surrounds (p. 14). This plane reflects the response of society to the individual’s experience be this expressed in attitudes, such as the engendering o f stigma, or in behavior, which may include specific instruments such as legislation (p. 26). Handicap is thus a social phenomenon, representing the social and environmental consequences for the individual, stemming from the presence o f impairments and disabilities (P-26). WHO. 1980 Brown also states: It is ironic that handicap has been so criticized, because the concept was offered as a measure precisely to move the focus from characteristics o f the individual to those environmental and or societal characteristics that limit the individual (p.61). Brown. 1993 Causal Interpretation o f the Theoretical Model. It has been suggested that the ICIDH is deeply flawed because it is modeled as different levels o f health resulting from some aspect o f morbidity (e.g., disease, trauma, mental illness, or age-related conditions) (Bickenback et al., 1999). The classification suggests that people are disadvantaged because o f their disabilities alone and that handicaps are caused by impairments and disability: Disadvantage accrues as a result of [the individual] being unable to conform to the norms o f his universe. Handicap is thus a social phenomenon, representing the social and environmental consequences for the individual stemming from the presence o f impairments and disabilities. WHO. 1980. p. 29 The model shows one-way arrows liking disease to impairment to disability to handicap. The manual describes disability as “resulting from an impairment” and handicap as “resulting form an impairment or a disability” (WHO, 1980). This linear model implies that dysfunction results from a sequence o f stages determined by medical conditions. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The model seems short sighted. For instance, the model suggests that the term disability refers to the term impairment, alone. There are, however, disabilities which are part o f a disorder (e.g., compulsive-obsessive behavior) such that disability could refer to the term disease, as well. Thus, the causal interpretation o f Wood’s model o f consequences of diseases seems problematic. In response to these criticisms, and because o f the wide acceptance of the ICIDH, the World Health Organization initiated revisions o f the ICIDH in 1990 (Badley, 1993). The revisers o f the ICIDH seem to have taken these criticisms to heart. The ICIDH-2-B2 (beta version) consists of a "biopsychosocial” model that synthesizes the medical and social approaches to disablement (http://www.who.ch/icidh). This 1999 draft represents changes in structure and content in preparation o f the final version expected in 2001. First, the ICIDH-2 draft differs from the 1980 version in that the dimensions of impairment, disabilities and handicaps are re-defined as body structure and function, activities, and participation. The body dimension "comprises two classifications, one for functions o f body systems, and one for the body structure" (ICIDH-2-B2, http://www.who.ch/icidh. section 3). While the body functions "are the physiological and psychological functions of the body systems, the body structures are anatomic parts o f the body such as organs, limbs, and their components" (ICIDH-2-B2 http://w A vw .w ho.ch/icidh. section 4). The activity dimension "covers the complete range o f activities performed by an individual” (ICIDH-2-B2, http: 7w vvA v.u ho.eh/icidh. section 3). The participation dimension "classified areas o f life in which an individual is involved has access to , has social opportunities or barriers” (ICIDH-2-B2, http://vvA vw.who.ch/icidh. section 3). The draft ICIDH-2 also contains a fourth 3 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. component, a listing o f environmental factors, that can be used with the other classifications or separately to identify the determinants o f disablement at the body, person, or person-in-context levels o f human functioning. Simeonsson et al. (2000) has provided a summary o f these dimensions that is extended here, see Table II-2. Each dimension is viewed as an interaction between the individual and their social and physical environment. (Bickenback et al., 1999). Thus, together, this classification reflects a dynamic instead o f a linear relationship reflecting the interaction o f these components, see Figure II-5. The ICIDH-2 uses an alphanumeric system. For instance if we wanted to classify bathing, the classification is first given the letter "a” because it is an "activity" of walking. If the component fell within the dimensions o f body function, body structure, participation, or environment, they would be assigned a letter b, s, p, and e. respectively. These letters are then followed by a numeric code that starts with the chapter number (1 digit), a second level o f the classification (2 digits), and the third and fourth levels o f the classification (each one digit). Qualifiers are then appended after a decimal point to provide additional information and each code should be accompanied by at least one qualifier. The first qualifier specified the magnitude o f the functioning or disability in that category. The second qualifier is dimension specific (i.e., within the dimension of activities it is specific to assistance, within the dimension o f body function and structure it is specific to localization, within the dimension o f participation it is specific to subjective satisfaction, and dimension o f environment is still under development). The first qualifier or “uniform” qualifier is coded in the same manner for all components: 0=no problem, l=m ild problem, 2=moderate problem, 3=severe problem, 38 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le 11-2. IC ID H -2 d r a ft o n e d ig it le v e l c la s s if ic a tio n (f ro m S im e o n s s o n e t a l., 2 0 0 0 ) B o d y f u n c t io n s B o d y s t r u c t u r e A c tiv itie s P a r t i c i p a t i o n E n v i r o n m e n t a l f a c t o r s M e n ta l fu n c tio n s S tr u c tu r e o f th e n e r v o u s s y s te m A c tiv itie s o f le a r n in g & a p p ly in g k n o w le d g e P a r tic ip a tio n in p e r s o n a l m a in te n a n c e P ro d u c ts & te c h n o lo g y S e n s o r y fu n c tio n s E y e , e a r , a n d re la te d s tr u c tu r e s C o m m u n ic a tio n a c tiv itie s P a rtic ip a tio n in m o b ility N a tio n a l e n v ir o n m e n t & h u m a n m a d e c h a n g e s to e n v iro n m e n t V o ic e & s p e e c h fu n c tio n s S tr u c tu r e s in v o lv e d in v o ic e & s p e e c h M o v e m e n t a c tiv itie s P a rtic ip a tio n in e x c h a n g e o f in fo rm a tio n S u p p o rt & re la tio n s h ip s F u n c tio n s o f th e c a r d io v a s c u la r , h e m a to lo g ic a l, im m u n o lo g ic a l, & re s p ira to ry s y s te m S tr u c tu r e o f th e c a r d io v a s c u la r , im m u n o lo g ic a l & re s p ira to ry s y s te m s A c tiv itie s o f m o v in g a r o u n d P a rtic ip a tio n in s o c ia l r e la tio n s h ip A ttitu d e s , v a lu e s , & b e lie fs F u n c tio n s o f th e d ig e s tiv e , m e ta b o lic , & e n d o c r in e s y s te m s S tr u c tu r e s r e la te d to th e d ig e s tiv e , m e ta b o lis m & e n d o c r in e s y s te m s S e l f c a r e a c tiv itie s P a rtic ip a tio n in h o m e life & a s s is ta n c e to o th e r s S e rv ic e s G e n ito u r in a r y & r e p r o d u c tiv e fu n c tio n s S tr u c tu r e r e la te d to g e n ito u r in a r y s y s te m D o m e s tic a c tiv itie s P a rtic ip a tio n in e d u c a tio n S y s te m s & p o lic ie s N e u r o m u s c u lo s k e le ta l & m o v e m e n t re la te d fu n c tio n s S tr u c tu r e r e la te d to m o v e m e n t In te rp e r s o n a l a c tiv itie s P a rtic ip a tio n in w o rk & e m p lo y m e n t F u n c tio n s o f th e s k in & r e la te d s tr u c tu re s S k in & r e la te d s tru c tu re s P e rfo rm in g ta s k s & m a jo r life a c tiv itie s P a r tic ip a tio n in e c o n o m ic life U * N O Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. F ig u r e 11-5. P e r s o n - e n v ir o n m e n t in te ra c tio n b e tw e e n d im e n s io n s o f IC ID H -2 (W H O , 1 9 9 9 ) H e a lth C o n d itio n s ( d is o r d e r o r d is e a s e ) B o d y F u n c tio n s & S tru c tu re A c tiv ity E n v ir o n m e n ta l F a c to rs P e rs o n a l F a c to rs P a rtic ip a tio n o 4=complete problem, 8=not specified, and 9=not applicable. Thus, the classification a5101.2 indicates a moderate difficulty with bathing the whole body and there is moderate difficulty without the use o f assistive devices and the classification aS lO l. 1 1 indicates that there is a mild difficulty with bathing the whole body with the use of devices (WHO, 1999). Despite the criticisms o f the original ICIDH. we should remember that the WHO is an operational organization and not a scientific one. W ood's model is therefore a pragmatic one. The paradigm he proposed argued that the classical medical model of illness is too narrow' because it fails to reflect illness related consequences like impairment, disabilities, and handicaps. Even in its original form, the ICIDH contributed to the field o f disability by providing a basis for differentiated interventions based on the level o f physical dysfunction. For instance, while preventive measures could be applied at the level of impairment; rehabilitation could be applied at the level o f disability; and legislation or educational efforts could be applied at the level o f handicap (Bettinghaus. 1980). While the original ICIDH had a more narrow scope, emphasizing conditions based on physical and motor problems that apply to primarily adults, the revised beta- version if the ICIDH-2 has increased the cataloguing o f components that contribute to the expression of disability. This increased comprehensiveness and complexity may contribute to both the strengthening and weakening of the ICIDH. While the classification o f the old system was much shorter than the new one, the success o f this new classification will depend on the implementation o f this classification to a broad array o f purposes. Unfortunately, for the purpose o f this investigation, the ICIDH-2 is difficult to operationalize. For instance, the reported measures o f physical 41 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. functioning could be placed within the dimension o f "'activities” or "participation." If the ICIDH-2 is to be used for scientific investigation, the definitions and distinctions o f these two dimensions (activities and participation) is needed to better operationalize these constructs. The original ICIDH (WHO, 1980) and the Nagi framework (1965, 1969, 1976. 1977), and the Disablement Process (Verbrugge and Jette. 199) are similar in that they refer to the consequences of impairment at the person level. Although the ICIDH called this level "disability" and the Nagi framework and the Disablement Process called this level "functional limitations" the definitions o f these concepts are similar. O f the theoretical concepts presented here, only that o f Verbrugge and Jette, explicitly suggested that functional limitations should be tested by "'self reports or proxy reports o f doing an action; an interviewers observation o f the subject doing an action with a rating o f his or her performance, and equipment based performance, including timed tasks" (Verbugee and Jette. 1994, p.3). O f the potential reported measures one could use to examine the relationship between reported and performance-based measures of "functional limitation" a discrepancy surfaces. While it seems acceptable to compare the reported measures o f Nagi (1976) (e.g., pushing/pulling large objects, stooping/crouching/kneeling, carrying ten pounds or more, reaching/extending arms above shoulder, and writing/handling small objects) or Rosow and Breslau (1966) items of gross motor functioning (i.e., working around the house, walking up/down a flight of stairs, and walking half a mile) with performance-based measures; Verbugge and Jette (1994) and Nagi (1965. 1968, 1976, and 1977) suggest that the reported ADL measures are actually defining the concept o f “disability." Thus it is not appropriate to compare reported measures o f ADL 4 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. with performance-based measures o f physical functioning because we would be comparing measures o f different concepts, (ie., "disability” and "functional limitation”). Verbugge and Jette (1994) state that "ADLs are activities a person does as a member of society, even if they occur in the private rather than public setting!” (Verbugge and Jette. 1994. p.6). Similarly, although Nagi (1965. 1968, 1976. 1977) states that disability is reached when functional limitations result in the inability to perform expected social roles, he sees the role o f self-care as a "social role.” The author disagrees with exclusion o f ADL from the concept o f "functional limitation” made by Nagi (1965, 1969, 1976. 1977) and Verbugge and Jette (1994). Functional limitation is defined by Verbugge and Jette as "restrictions in performing fundamental physical and mental actions used in daily life by one's age-sex group” (Verbrugge and Jette, 1994, p.3). According to this definition, the author suggests that ADL are "physical and mental actions used in daily life” and should be included in the definition o f functional limitations, not disability. Verbugge and Jette (1994) go on to state that they distinguish between the concepts o f functional limitations and disabilities by the usage o f the terms "action” and "activity.” This point o f differentiation is too fine to distinguish between the states o f functional limitation and disability. The author prefers to distinguish the concepts of functional limitations and disability by incorporating the role o f the environment as Wood (ICIDH, 1980) does. The ICIDH (WHO, 1980) definition o f "handicap” is probably a better definition o f "disability.” Recall that Wood defines handicap as a disadvantage placed on the individual by the environment. This "disadvantage” is what should be called "disability.” Thus, for the present investigation, all three types o f reported measures (Nagi, gross-motor 4 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. functioning and ADL) and performance-based measures are treated as "functional limitation.” The relationship between these self-reported and performance-based measures o f "functional limitation” are thereby investigated. E. Self-Reported and Performance-Based Measures o f Physical Functioning Studies have already begun to explore the relationship between reported and performed measures of physical functioning. Table II-3 summarizes the data, measurement, method, and findings o f previous studies examining this relationship. In general, although rates o f agreement between self-reported and performance-based measures o f physical functioning seem to vary (for both individual and summary items), overall rates o f agreement hover around eighty percent. O f these studies, only three had a longitudinal component (Myers et al.. 1993: Mendes de Leon et al.. 1996; Kempen et al.. 1999). Although most o f these studies have relied on community-based samples (Kelly- Hayes et al.. 1992; Myers et al.. 1993; Reuben et al., 1995: Kempen Van Heuvelen et al. 1996; Merrill et al., 1997; Ferrer et al.. 1999), a few have sampled hospitalized patients (Dorevitch et al.. 1992; Sager et al.. 1992; and Elam et al.. 1991) and others have relied on population based samples (Jette and Branch, 1985; Kempen, Steverink et al.. 1996; Kempen. Sullivan et al., 1999). A more detailed description o f these studies is provided here. Among community-based populations o f older persons, varying rates of agreement were also found between reported and observed measures o f physical functioning. Kelly-Hayes et al. (1992) found particularly high rates o f agreement between individual self-reported and performance-based measures o f six ADLs (from 4 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le H -3 . S u m m a r y o f te n r e s e a r c h s tu d ie s b y d a ta s o u r c e , m e a s u r e s , m e th o d , a n d fin d in g s S t u d y D a t a S o u r c e M e a s u r e s M e t h o d F i n d i n g s E la m 73 in d iv id u a ls 6 0 S e lf -R e p o rte d (S R ): C r o s s P e rc e n t A g re e m e n t: e t a l.. a n d o l d e r w h o w e r e 5 A D L s : w a lk , d r e s s , e a t, tr a n s f e r a n d u s e ta b u la tio n s W a lk in g 9 0 .4 % 1991 h o s p ita liz e d a t th e te le p h o n e ( J e tte a n d B r a n c h , 1 9 8 1 ) T e le p h o n e u s e 8 4 .5 % B a p tis t M e m o ria l Classified: in d e p e n d e n t ( a b le w ith o u t a s s is ta n c e T r a n s f e r r in g 7 9 .2 % H o s p ita l fro m o th e r p e r s o n s ) o r d e p e n d e n t ( c a n ’t d o o r n e e d a s s is ta n c e fro m o th e r p e r s o n s ) E a tin g 7 2 .2 % D re s s in g 5 0 .0 % C r o s s - S e c tio n a l P e rfo rm a n c e -B a s e d (P B ): 5 ite m s : w a lk 2 5 fe e t, b u tto n s h ir t, tr a n s f e r d r ie d b e a n s w ith a s p o o n fr o m 1 d is h to a n o th e r, tr a n s f e r fro m b e d to c h a ir , m a k e a te le p h o n e c a ll. Classified: in d e p e n d e n t o r d e p e n d e n t! re q u ire u s e o f e q u ip m e n t, o th e r p e r s o n s , o r b o th ). D o r e v itc h 1 SO p a tie n ts S e lf - R e p o r te d ( S R k C r o s s A g r e e m e n t b e tw e e n s e lf-r e p o r te d et a)., a tte n d in g a 13 r e p o r te d ite m s . ta b u la tio n s a n d p e r fo rm a n c e -b a s e d m e a s u re s 1 9 92 g e r ia tr ic d a y Classified: in d e p e n d e n t ( p e r f o r m a lo n e ) o r r a n g e d fro m 6 7 % to 1 0 0% : h o s p ita l d e p e n d e n t ( n e e d a s s is ta n c e o r n o t a t a ll). D r in k in g 1 0 0 % E a tin g 9 8 % C r o s s - S e c tio n a l P e r f o r m a n c e - B a s e d (P B ): S a m e 13 ite m s p e r f o r m e d to th e b e s t o f th e ir a b ility . Classified: in d e p e n d e n t (p e r f o r m a l o n e ) o r d e p e n d e n t (n e e d a s s is ta n c e o r n o t a t a ll). D r e s s in g u p p e r b o d y 9 4 % D re s s in g lo w e r b o d y 8 2 % G r o o m in g 9 4 % W a s h in g 6 7 % C o n tr o llin g u r in a tio n 8 5 % C o n tr o llin g b o w e ls 9 6 % C h a ir tra n s fe rs 9 4 % T o ile t tra n s fe rs 9 4 % B a th /s h o w e r tra n s fe rs 8 9 % W a lk in g 5 0 m e te rs 8 7 % W a lk in g u p /d o w n s ta irs 7 7 % O v e r a ll,8 8 .9 % a g re e m e n t; 6 .1 % o v e r- re p o rt a n d 5 .0 % u n d e r-re p o rt. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le 11-3 ( c o n t.) . S u m m a r y o f te n r e s e a r c h s tu d ie s b y d a ta s o u rc e , m e a s u r e s , m e th o d , a n d f in d in g s S t u d y D a t a S o u r c e M e a s u r e s M e t h o d F i n d i n g s K e lly -H a y e s C o m m u n ity - b a s e d S e lf - R e o o r te d fS R ): C r o s s P e rc e n t A g re e m e n t e t a l., 1 9 9 2 s a m p le o f 1 4 5 3 6 A D L s : d r e s s in g , g r o o m in g , e a tin g . ta b u la tio n s T r a n s f e r r in g 9 6 .6 % p a r tic ip a n ts 6 3 to tr a n s f e r r in g , w a lk in g , s ta ir c lim b in g ( K a tz e t E a tin g 9 6 .4 % 9 4 y e a r s o f a g e a l., 1 9 7 9 a n d M a h o n e y a n d B a r th e l, 1 9 6 5 ). Classified: u s e s n o h e lp to p e r fo rm , u s e s D re s s in g 9 6 .1 % G r o o m in g 9 5 .5 % C r o s s - S e c tio n a l e q u ip m e n t, u s e s o th e r p e r s o n s , d o e s n o t p e r f o r m . P e r f o r m a n c e - B a s e d (P B ): P r o f e s s io n a ls o b s e r v e d th e p e r f o r m a n c e o f th e s e 6 A D L S . Classified: d i d a c tiv ity , re q u ire d e q u ip m e n t, r e q u ir e d h u m a n a s s is ta n c e . W a lk in g 9 3 .5 % S ta ir C lim b in g 9 3 .5 % O v e r a ll, in d iv id u a ls te n d to o v e r re p o r t 8 9 % o f th e tim e . S a g e r 3 0 2 in d iv id u a ls 7 0 S e lf - R e p o r te d (S R I: C r o s s P e rc e n t A g re e m e n t: e t a l., 1 9 9 2 a n d o l d e r a d m itte d 5 A D L S : d r e s s , b a th e , e a t, to ile t. ta b u la tio n s E a tin g 9 4 % to S t. M a r y ’s tr a n s f e r fr o n t b e d to c h a ir. T o ile tin g 8 4 % H o s p ita l M e d ic a l Classified: in d e p e n d e n t, n e e d h e lp , u n a b le to d o . T r a n s f e r r in g 8 2 % C e n te r P e r f o r m a n c e - B a s e d IP B ): D re s s in g 6 4 % B a th in g 6 3 % C r o s s - S e c tio n a l S a m e 5 A D L s u s in g F u n c tio n a l I n d e p e n d e n c e M e a s u re ( F IM ) ( G r a n g e r e t a l., 1 9 8 6 ). Classified: in d e p e n d e n t, n e e d h e lp , u n a b le to d o . O v e r a ll r a te s o f a g re e m e n t b e tw e e n s e lf -r e p o r te d a n d p e r f o r m a n c e - b a s e d m e a s u re s w a s 7 8 % : O n a ll 5 ite m s 4 5 % O n 4 ite m s 2 j % O n 3 ite m s 17% O n 2 ite m s 10% O n 1 ite m 4 % O n n o n e 1% 4 - O n Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le H -3 (c o n t ). S u m m a r y o f t e n r e s e a rc h s tu d ie s b y d a t a s o u rc e , m e a s u re s , m e th o d , a n d fin d in g s S tu d y D a ta S o u r c e M e a su r e s M e th o d F in d in g s M y e r s e t a l . , 1 9 9 3 18 2 in d iv id u a ls 6 0 - 9 2 y e a rs o f a g e r e s id in g in th e c o m m u n ity C r o s s - S e c tio n L o n g itu d in a l (1 - y e a r f o llo w -u p ) S e lf - R e p o r te d ( S R ): 15 lA D L s ' Classified', n o d if f ic u lty , s o m e d if f ic u lty , a lo t o f d iff ic u lty . P e r f o r m a n c e - B a s e d ( P B ): 14 I A D L s 2 th a t m ir r o r th e s e lf -r e p o r te d ite m s . Classified: n o d if f ic u lt, s o m e d iff ic u lty , and a lo t o f d if f ic u lty /u n a b le t o d o . C r o s s - In th e C r o s s -S e c tio n : ta b u la tio n s P e rc e n t A g re e m e n t: C o o k in g lig h t m e a ls 9 0 % C o o k in g fu ll m e a ls 8 0 % U s in g th e p h o n e 8 4 % T e llin g tim e 8 2 % W a lk in g 5 5 % In d iv id u a ls u n d e r-re p o rt: C o o k in g lig h t m e a ls C o o k in g fu ll m e a ls T e llin g tim e W a lk in g In d iv id u a ls o v e r- re p o rt: U s in g th e p h o n e L o n g itu d in a l: B e lo w 8 0 % fo r 10 o f th e 15 IA D L m a tc h in g B e lo w 6 0 % fo r w a lk in g a n d _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ u s in g th e te le p h o n e d ir e c to r y Notes: 1 15 r e p o r te d I A D L s u s e d in th e s tu d y b y M y e r s e t a l., ( 1 9 9 3 ) : w rite a s h o rt le tte r, te llin g tim e fro m b e d s id e c lo c k , re a d in g n u m b e rs in te le p h o n e b o o k , u s in g a te le p h o n e , o p e n in g c h i l d p r o o f p ill b o x , o p e n in g flit to p p ill b o ttle , o p e n in g p u s h /tu r n p ill b o ttle , o p e n in g s c re w /to p p ill b o ttle , o p e n in g b l is te r p a c k , r e a d in g la b e l fro m b o ttle , p r e p a r in g fu ll m e a ls (e .g ., m e a t v e g e ta b le s ) , p r e p a r in g lig h t m e a ls (e .g ., s o u p /s a n d w ic h , s w e e p in g flo o r, r e a c h in g o n to s h e l f a t e y e le v e l, a n d w a lk in g . 2 14 p e r f o r m e d lA D L s u s e d th e s tu d y b y M y e r s e t a l., ( 1 9 9 3 ) : w ritin g , te llin g tim e , lo c a tin g a p h o n e n u m b e r in a p h o n e b o o k , m a k in g a te le p h o n e c a ll, r e m o v in g to p o f c h i l d p r o o f c o n ta in e r , r e m o v in g to p o f flip /f lo p c o n ta in e r, re m o v in g to p o f f a s c r e w /to p c o n ta in e r, r e m o v in g to p o f f o f b lis te r p a c k , r e a d in g a la b e l, c o o k in g s im u la tio n , s w e e p in g s im u la tio n , r e a c h in g s im u la tio n , w a lk in g 12 fe e t. ■u -4 Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le 11-3 ( c o n t.) . S u m m a r y o f te n r e s e a r c h s tu d ie s b y d a ta s o u rc e , m e a s u r e s , m e th o d , a n d fin d in g s S tu d y D a ta S o u rc e M e a su r e s M e th o d F in d in g s C r e s s e t a l . , 1995 4 1 7 in d iv id u a ls r e s id in g in th e c o m m u n ity a n d 2 0 0 n u r s in g h o m e r e s id e n ts 6 2 - 9 8 y e a r s o f a g e . C r o s s - S e c tio n a l S e lf - R e p o r te d (S R ): 4 s u b - s c a le s fro m th e S ic k n e s s I m p a c t P r o file (S IP ) : b o d y c a r e a n d m o v e m e n t s c o r e , a m b u la tio n s c o re , m o b ility s c o re , p h y s ic a l d im e n s io n s c o re ( B e r g n e r e t a l., 1 9 8 1 ) Classified: S c o re s o n e a c h s u b - s c a le s ra n g e fr o m 0 to 10 0 w ith h ig h e r s c o r e s in d ic a tin g w o r s e fu n c tio n in g . P e r f o r m a n c e - B a s e d (P R ): 4 m e a s u re s : b a la n c e ( p a ra lle l, s e m i- ta n d e m , a n d ta n d e m s ta n c e s ) , 3 r e p e a te d c h a ir s ta n d s , g r ip s tr e n g th , g a it o f 6 m e te rs to 4 0 m e te r s ( n o tu rn a r o u n d ). Classified: B a la n c e (0 -5 s e c o n d s ) . C h a ir s ta n d s (s e c o n d s ), G r ip s tr e n g th (k g ), G a it s p e e d (m /s e c o n d s )._ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ C o r r e la tio n F o r b o th p o p u la tio n s m o d e ra te b e tw e e n s e lf - c o r r e la tio n s ( r a n g e d fro m r = - 0 .1 9 r e p o r te d a n d to r = - 0 .6 3 ) w e re fo u n d b e tw e e n p e r f o r m a n c e - s e lf - r e p o r te d a n d th e fo u r S I P b a s e d s c a le s , s u c h th a t in d iv id u a ls m e a s u r e s fo r r e p o r tin g p o o r e r fu n c tio n in g h a d th e tw o w o r s e p e r fo rm a n c e , p o p u la tio n s . O O CD ■ o - 5 o Q . Q . I ■ o T a b le H -3 ( c o n l.) . S u m m a r y o f te n r e s e a r c h s tu d ie s b y d a ta s o u r c e , m e a s u r e s , m e th o d , a n d f in d in g s ( / ) ' ( f \ S tu d y D a ta S o u r c e M e a su r e s M e th o d F in d in g s o ' R e u b e n 8 3 in d iv id u a ls (6 4 S e lf - R e p o r te d (S R ): C o r r e la tio n C o r r e la tio n w e r e s ig n if ic a n t a n d o e t a l., 1 9 9 5 to 9 2 y e a r s o f a g e ) 2 F u n c tio n a l S ta tu s Q u e s tio n n a ir e s (F S Q ) b e tw e e n s e lf - h ig h b e tw e e n : fro m a c o m m u n ity ( J e tte e t a l., 1 9 8 6 ) a n d th e M e d ic a l O u tc o m e s re p o r te d a n d F S Q /B A D L a n d P P T 0 .5 5 CD o u tr e a c h p r o g r a m . S u rv e y ( M O S S F - 3 6 ) ( W a r e a n d S h e rb o u m e , p e r f o r m a n c e - F S Q /IA D L a n d P P T 0 .4 5 O O 1 9 9 2 ): b a s e d M O S S F -3 6 a n d P P T 0 .2 6 " O * < C r o s s - S e c tio n a l • 3 ite m s fr o m F S Q /B A D L : m e a s u r e s I n d ic a tin g th a t b e tte r re p o r te d c q ' D if f ic u lty e a tin g /d r e s s in g /b a th in g fu n c tio n in g w a s a s s o c ia te d w ith l-H Q D if f ic u lty m o v in g in /o u t o f b e d o r c h a ir b e tte r p e r fo rm e d fu n c tio n in g . 3 tT \ D if f ic u lty w a lk in g in d o o rs KV —S • 4 ite m s fr o m F S Q /IA D L : T | C D if f ic u lty w a lk in g s e v e r a l b lo c k s D if f ic u lty w a lk in g o n e b lo c k o r c lim b in g CD o n e flig h t o f s ta irs ~ s CD - Q D if f ic u lty w o r k in g a r o u n d th e h o u s e —i O D if f ic u lty d o in g e r r a n d s s u c h a s g r o c e r y Q . C s h o p p in g a • 10 Ite m s fro m M O S S F -3 6 : o D o e s H e a lth lim it p a r tic ip a tio n 3. ■o c \ Classified: u • B o th F S Q S c r e e n s s c a le d as: o ; l-H A b le /n o d if f ic u lty , CD Q . A b le /s o m e d if f ic u lty , A b le /m u c h d iff ic u lty . D id n o t d o b e c a u s e o f h e a lth , o r u C l-H D id n o t d o f o r o t h e r re a s o n s . S c o r e s ra n g e d T 3 CD fro m 0 in d ic a tin g w o r s e f u n c tio n in g to 10 0 3 in d ic a tin g b e s t fu n c tio n . w ' • M O S S F - 3 6 s c a le d as: i f ) o ' L im ite d a lo t. 3 L im ite d a little , o r N o t lim ita tio n s "O Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le 11-3 ( c o n t.) . S u m m a r y o f te n re s e a rc h s tu d ie s b y d a ta s o u rc e , m e a s u r e s , m e th o d , a n d fin d in g s S tu d y D a ta S o u rc e M e a su r e s M e th o d F in d in g s R e u b e n e t a l . , 1 9 9 5 ( c o n tin u e d ) P e rfo rm a n c e -B a s e d (P B ): 7 ite m P h y s ic a l P e r f o r m a n c e - T e s t ( R e u b e n a n d S iu , 1 9 9 0 ): w ritin g a s e n te n c e s im u la te e a tin g e a tin g liftin g a b o o k a n d p u ttin g it o n a s h e l f a b o v e p u ttin g o n a n d r e m o v in g a j a c k e t p ic k in g u p a p e n n y fr o m th e f lo o r tu rn in g 3 6 0 d e g re e s w a lk in g 5 0 fe e t Classified: E a c h ite m w a s s c o r e d o n a 5 p o in t s c a le (0 - 4 ) w ith 0 = u n a b le to d o a n d 4 = q u ic k e s t tim e o r b e s t fu n c tio n . H ig h e r o v e ra ll s c o r e s in d ic a te b e tte r fu n c tio n a n d th e to ta l s c o r e r a n g e s fro m 0 to 2 8 . Notes: 3 10 re p o r te d ite m s fr o m th e M O S S F -3 6 u s e d b y R e u b e n e t a l., (1 9 9 5 ) : r u n n in g /lif tin g h e a v y o b je c ts /p a r tic ip a tin g in s tr e n u o u s s p o rts , m o v in g a ta b le /p u s h in g a v a c u u m c le a n e r /b o w lin g /p la y in g g o lf , lif tin g /c a r r y in g g r o c e r ie s , c lim b in g s e v e r a l flig h ts o f s ta irs , c lim b in g o n e flig h t o f s ta irs , b e n d in g /k n e e lin g /s to o p in g , w a lk in g m o r e th a n a m ile , w a lk in g s e v e r a l b lo c k , w a lk in g o n e b lo c k , b a th in g /d r e s s in g . V / i o Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le 11-3 ( c o m .) . S u m m a r y o f te n r e s e a r c h s tu d ie s b y d a ta s o u rc e , m e a s u re s , m e th o d , a n d f in d in g s S tu d y D a ta S o u r c e M e a su r e s M e th o d F in d in g s K e m p e n , V a n H e u v e le n e t a l., 1 9 9 6 6 2 4 in d iv id u a ls 5 7 a n d o ld e r iVom th e G ro n in g e n L o n g itu d in a l A g in g S tu d y (G L A S ) C r o s s -S e c tio n a l S e lf - R e p o r te d (S R ): 3 Ite m s : 1) M o to r fu n c tio n in g fro m th e O E C D in d ic a to r “fr e e d o m o f m o v e m e n t’ th a t c o n s is ts o f th re e ite m s (c a r r y in g 5 k g , p ic k in g u p s o m e th in g fr o m th e flo o r, a n d w a lk in g 4 0 0 m e te rs w ith o u t s to p p in g ) 2 ) H e a r in g w a s a s s e s s e d b y te n q u e s tio n s fr o m th e H e a r in g H a n d ic a p a n d D is a b ility In v e n to r y 3 ) V is io n w a s a s s e s s e d b y e ig h t q u e s tio n s (e .g ., c a n y o u re a d n u m b e s in a p h o n e b o o k , c a n y o u s e e a p p ro a c h in g b ic y c le s a n d c a r s , e tc .) Classified: F o r a ll 3 ite m s , c la s s if ie d a s : a lm o s t a lw a y s , o f te n , s o m e tim e s , a n d a lm o s t n e v e r. P e r f o r m a n c e - B a s e d ( P B ) a n d th e ir C la s s if ic a tio n : 5 Ite m s fo r m o to r fu n c tio n in g : I ) W a lk in g e n d u r a n c e w a s a s s e s s e d o n a r e c ta n g u la r c o u r s e w h e r e w a lk in g s p e e d w a s in c r e a s e d fr o m 4 k m /h to 7 k m /h to I k m /h e v e ry 3 m in u te s . S c o re re f le c ts to ta l d is ta n c e w a lk e d . F le x ib ility o f th e h ip a n d s p in e w e re a s s e s s e d b y th e s it a n d re a c h te s t: P a rtic ip a n ts sit o n th e flo o r, le g s o u ts tr e tc h e d a n d b e n d fo rw a r d to p u s h a s lid e a s fa r a s p o s s ib le w ith th e ir fin g e rtip s . S h ift le n g th o f s lid e re c o r d e d ._ _ _ _ _ _ M u ltip le r e g r e s s io n a n a ly s e s to e x a m in e th e e x te n t o f v a r ia n c e e x p la in e d in s e lf - r e p o r te d p h y s ic a l fu n c tio n in g b y p e r f o r m a n c e - b a s e d le v e ls. T h e s tr e n g th o f th e a s s o c ia tio n b e tw e e n s e lf -r e p o r t a n d p e r fo rm a n c e -b a s e d m e a s u re s o f p h y s ic a l fu n c tio n in g is m o d e ra te . W ith in th e d o m a in s o f m o to r fu n c tio n in g (3 3 .1 % , p < 0 .0 0 l ) a n d h e a r in g (4 1 .4 % , p < 0 .0 0 1) m u c h s tr o n g e r a s s o c ia tio n e x is t th a n w ith th e d o m a in o f v is io n (1 2 .5 % , p < 0 .0 0 1 ) . Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le 11-3 ( c o n t.) . S u m m a r y o f te n re s e a rc h s tu d ie s b y d a ta s o u rc e , m e a s u r e s , m e th o d , a n d fin d in g s S tu d y D a ta S o u r c e M e a su r e s M e th o d F in d in g s K e m p e n , V a n H e u v e le n e t a l., 1 9 9 6 ( c o n t.) 2 ) F le x ib ility o f th e s h o u ld e r s w a s a s s e s s e d b y h o ld in g b o th h a n d le s o f a c o r d a n d p a s s th e c o r d fr o m th e fro n t o f th e b o d y o v e r th e h e a d to b e h in d th e b o d y , k e e p in g th e a r m s s tr a ig h t a n d a s c lo s e a s p o s s ib le . T h e s h if t o f th e h a n d le w ith th e p e r s o n ’s a r m le n g th w a s u s e d to d e te r m in e th e s c o re . 3 ) B a la n c e a s s e s s e d b y s ta n d in g o n a p la tfo r m th a t tilte d fro n t s id e to s id e . D u r in g a 3 0 s e c o n d in te rv a l th e to ta l tim e th a t b a la n c e w a s m a in ta in e d ( p la tf o r m d id n o t to u c h th e f lo o r ) w a s s c o re d . 4 ) G r ip S tr e n g th w a s a s s e s s e d b y a d y n a m o m e te r H e a rin g . H e a r in g w a s a s s e s s e d b y w ith a p u r e to n e a ir c o n d u c tio n a u d io m e tr y . T h e m e a n lo ss in d e c ib e ls a b o v e f o u r fr e q u e n c ie s (0 .5 , 1, 2 a n d 4 k H z w a s c o m p u te d ) . V is io n : V is io n w a s a s s e s s e d b y v is u a l a c u ity , c o n tr a s t s e n s itiv ity , d is a b ility g la r e , te x t r e a d i n g , a n d fo v e a l lig h t s e n s itiv ity . T h e r e s u lts o f th e s e te s ts w e r e c o n v e r te d to s ta n d a r d iz e d u n its w ith a s c o r e ra n g e fro m O to 115. C / i IO Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le 11-3 (c o n t.) . S u m m a r y o f te n r e s e a r c h s tu d ie s b y d a ta s o u rc e , m e a s u r e s , m e th o d , a n d fin d in g s S t u d y D a t a S o u r c e M e a s u r e s M e t h o d F i n d i n g s K e m p e n , 7 5 3 in d iv id u a ls S e lf-R e D o rte d (S R ): M u ltip le T h e s tr e n g th o f th e a s s o c ia tio n S te v e rin k 5 7 a n d o l d e r a t S u m m a r y s c o r e a c r o s s a n 11 ite m A D L s c a le r e g r e s s io n b e tw e e n s e lf -r e p o r t a n d e t a l., 1 9 9 6 b a s e lin e ( 1 9 9 3 ) Classified: a n a ly s e s to p e r fo rm a n c e -b a s e d m e a s u re s o f fro m th e S c o re r a n g e s fro m 11 (n o lim ita tio n s ) to e x a m in e th e p h y s ic a l f u n c tio n in g is m o d e ra te . G r o n in g e n 4 4 ( m a x im u m lim ita tio n s ) e x te n t o f L o n g itu d in a l v a r ia n c e F o r th e A D L s u m m a ry s c o re w h e r e A g in g S tu d y P e rfo rm a n c e -B a s e d (P B ): e x p la in e d in a ll th r e e p e rfo rm a n c e -b a s e d (G L A S ) S u m m a r y s c o re o f 3 ite m s : s e lf -r e p o r te d m e a s u r e s a r e in c lu d e d in th e 1) P u ttin g o n /ta k in g o f f a j a c k e t p h y s ic a l r e g r e s s io n e q u a tio n , 3 8 .8 % C r o s s - S e c tio n a l 2 ) W a lk in g 6 m e te rs ( in c lu d in g a 18 0 d e g r e e f u n c tio n in g b y ( p < 0 .0 l ) o f th e v a ria n c e in re p o r te d tu rn a 3 m e te rs ) p e r f o r m a n c e - A D L is e x p la in e d b y p e r fo rm a n c e . 3 ) 5 r e p e a te d c h a ir s ta n d s b a s e d le v e ls. Classified: T h e a s s o c ia tio n fo r th e in d iv id u a l T h e s c o r e fo r e a c h ite m is in n u m b e r o f m e a s u r e s w a s s m a lle r. F o r th re e s e c o n d s , w ith h ig h e r s c o r e s r e f le c tin g p o o r e r in d iv id u a l A D L ite m s (d re s s in g , p e r fo rm a n c e . P rin c ip le c o m p o n e n t a n a ly s is w a s g e ttin g a r o u n d th e h o u s e , a n d th e n u s e d to d e v e lo p a s in g le s u m m a r y m e a s u r e s ta n d in g u p fro m s ittin g in a c h a ir ) a c r o s s th e s e th re e ite m s . th e a p p r o p r ia te s in g le item p e r f o r m a n c e m e a s u re w a s in c lu d e d . H e re , th e v a ria n c e in th e r e p o r te d a b ility to d r e s s (1 2 .5 % , p < 0 .0 1), g e ttin g a r o u n d th e h o u s e (1 5 .3 % , p < 0 .0 1 ),a n d s ta n d in g u p fro m s ittin g in a c h a ir (1 3 .5 % , p < 0 .0 1) a r e e x p la in e d b y le ss th a n 1 5 % b y th e p e r fo rm a n c e -b a s e d m e a s u re s . L /i u > Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le 11-3 ( c o n t.) . S u m m a r y o f te n r e s e a r c h s tu d ie s b y d a la s o u rc e , m e a s u re s , m e th o d , a n d f in d in g s S t u d y D a t a S o u r c e M e a s u r e s M e t h o d F i n d i n g s M e n d e s d e L e o n e t a l., 1 9 9 6 1 ,103 c o m m u n ity re s id e n ts 7 2 a n d o ld e r L o n g itu d in a l S e lf - R e p o r te d f S R ): S u m m a r y s c o r e o f A D L b a s e d o n s ix ite m s ( b a th in g , d r e s s in g , e a tin g , to ile tin g , w a lk in g a c r o s s a r o o m , a n d tr a n s f e r r in g f r o m b e d to c h a ir). Classified: I n d iv id u a l ite m s w e r e first d ic h o to m iz e d to in d ic a te a b ility t o p e r fo rm th e ta s k w ith o u t h u m a n h e lp ( I ) a n d th e n e e d fo r h u m a n h e lp (0 ). T h e s e ite m s w e r e th e n s u m m e d to c r e a te a s c a le r a n g in g fr o m 0 to 6 w ith h ig h e r s c o r e s m e a n in g b e tte r f u n c tio n in g . P e r f o r m a n c e - B a s e d IP B ): A s u m m a r y s c o r e b a s e d o n s ix ite m s w a s d e v e lo p e d . T h e fiv e ite m s in c lu d e d w e re : F o o t ta p p in g , 3 r e p e a te d c h a ir s ta n d s , tu r n in g in a c irc le , a 2 0 fo o t w a lk , a n d b e n d in g o v e r to p ic k u p a p e n c il f r o m th e n flo o r, a n d s in g in g o n e ’s n a m e . Classified: T h e tim e to c o m p le le e a c h ta s k w a s re c o r d e d . A ll ta s k s w e r e h a lte d a t 3 0 s e c o n d s e x c e p t fo r w a lk in g w h ic h w a s h a lte d a t 6 0 s e c o n d s . T h e s c o re fo r w a lk in g w a s d iv id e d b y tw o to g iv e it a n e q u a l w e ig h t. A s u m m a ry s c o r e w a s c o m p u te d b y s u b tr a c tin g e a c h te s t s c o re fro m 3 0 s o th a t h ig h e r s c o r e s w o u ld r e p re s e n t b e tte r p e r fo rm a n c e a n d th e n a v e r a g in g th e s c o r e s o f th e s ix te s ts . T h is y ie ld e d a r a n g e o f 0 to 30._ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ M u ltiv a r ia te H ig h e r b a s e lin e le v e ls o f p h y s ic a l R e g r e s s io n p e r fo rm a n c e w a s a s s o c ia te d w ith h ig h e r le v e ls o f A D L fu n c tio n in g a t fo llo w -u p (b = 0 .0 2 9 , p < 0 .0 0 l ) , a f te r c o n tr o llin g fo r b a s e lin e A D L . L/i -fc. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le 11-3 (c o n t.) . S u m m a r y o f te n r e s e a r c h s tu d ie s b y d a ta s o u r c e , m e a s u r e s , m e th o d , a n d fin d in g s S tu d y D a ta S o u r c e M e a su r e s M e th o d F in d in g s M e rrill 1 ,4 5 8 S e lf - R e p o r te d (S R ): C r o s s F o u n d h ig h le v e ls o f a g r e e m e n t fo r e t a l . , 1 9 97 c o m m u n ity I n d iv id u a l M e a s u re s ta b u la tio n s a ll th r e e c o m p a ris o n s fo r m e n a n d r e s id e n ts 6 5 a n d A D L : w a lk in g w o m e n : o l d e r fro m th e N a g i: s to o p in g /c r o u c h in g /k n c e lin g a n d W a lk in g /w a lk in g E s ta b lis h e d re a c h in g /e x te n d in g a r m s o v e r h e a d . ( 8 9 .2 % m e n , 8 7 .5 % w o m e n ) P o p u la tio n s fo r Classified: u n a b le /a b le S to o p in g /C h a ir S ta n d th e ( 8 5 .5 % m e n , 6 9 .8 % w o m e n ) E p id e m io lo g ic a l P e rfo rm a n c e -B a s e d (P B ): R e a c h /R o ta te s h o u ld e r s S tu d y o f th e A b ility to w a lk 8 fe e t a n d b a c k , a b ility to ( 9 0 .7 % m e n , 8 0 .0 % w o m e n ). E ld e r ly p e r f o r m 5 c o n s e c u tiv e ris e s fro m c h a ir , a b ility to ( E P E S E ) , N e w r o ta te s h o u ld e r s F o r a ll th re e c o m p a ris o n s , o v e r- H a v e n ( C T ) site . Classified: u n a b le /a b le r e p o r tin g o f d is a b ility w a s m o re c o m m o n th a n u n d e r-re p o rtin g . C r o s s - S e c tio n a l M o r e w o m e n o v e r- re p o rt th a n m e n . Ul C / i Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le 11-3 ( c o n t ). S u m m a r y o f te n r e s e a r c h s tu d ie s b y d a ta s o u rc e , m e a s u re s , m e th o d , a n d fin d in g s S tu d y D a ta S o u r c e M e a su r e s M e th o d F in d in g s F e r r e r 6 2 6 in d iv id u a ls S e lf - R e p o r te d (S R > : C r o s s - e t a l . , 1 9 9 9 6 5 a n d o ld e r 2 Ite m s : w a lk in g a n d s ta n d in g u p /s ittin g d o w n ta b u la tio n s r e s id in g in fro m a c h a ir B a r c e lo n a , Classified: S p a in . A b le /u n a b le C r o s s - S e c tio n a l P e r f o r m a n c e - B a s e d (P B >: 2 Ite m s : 4 m e te r w a lk in g s p e e d le s t a n d 5 r e p e a te d c h a ir s ta n d s . Classified: R e p o r te d n u m b e r o f s e c o n d s to c o m p le te ta s k s w e r e re c o r d e d . T h e n a d ic h o to m o u s v a r ia b le w a s c r e a te d : a b le /u n a b le . K e m p e n , 7 5 3 in d iv id u a ls S e lf - R e p o r te d (S R ): C o r r e la tio n S u lliv a n e t 5 7 a n d o ld e r a t S u m m a r y s c o r e a c r o s s a n 11 ite m A D L s c a le a l., 1 9 9 9 b a s e lin e ( 1 9 9 3 ) Classified: fro m th e S c o r e r a n g e s f r o m 11 ( n o lim ita tio n s ) to G ro n in g e n 4 4 ( m a x im u m lim ita tio n s ) L o n g itu d in a l A g in g S tu d y P e r f o r m a n c e - B a s e d (P B ): ( G L A S ) S u m m a r y s c o r e o f 3 ite m s : I ) P u ttin g o n /ta k in g o f f a ja c k e t L o n g itu d in a l 2 ) W a lk in g 6 m e te rs ( in c lu d in g a 18 0 d e g re e (1 9 9 3 to 1 9 9 5 ) tu r n a 3 m e te rs ) 3 ) 5 r e p e a te d c h a ir s ta n d s Classified: T h e s c o r e fo r e a c h ite m is in n u m b e r o f s e c o n d s , w ith h ig h e r s c o r e s r e f le c tin g p o o r e r p e r f o r m a n c e . P r in c ip le c o m p o n e n t a n a ly s is w a s th e n u s e d to d e v e lo p a s in g le s u m m a ry m e a s u re a c r o s s th e s e th r e e ite m s . W a lk in g : 7 8 % a g re e m e n t S ta n d in g U p : 8 6 % a g re e m e n t R e p o rte d d is a b ility w a s lo w : 3 1 % fo r w a lk in g a n d 19 % fo r s ta n d in g u p M o re w o m e n o v e r- re p o rt o n th e s e ite m s. C o r re la tio n b e tw e e n c h a n g e in s e lf -r e p o r te d a n d c h a n g e in p e r fo rm a n c e -b a s e d p h y s ic a l fu n c tio n in g w a s m o d e ra te (r="0.28). C/t Os 93.% for walking to 96.4% for eating). These reported and performed measures variables were categorically classified, see Table II-2. They found that differences occurred more frequently for walking and stair climbing (6.5%) and least frequently for feeding (3.1%). In addition, when a difference occurred, reported disability was greater than functional limitations for 89% of the comparisons. Myers et al. (1993) found that discrepancies between 15 reported and observed instrumental activities o f daily living (IADL) were more common for items o f lower body mobility (walking-55%), than for upper body function, (e.g., using the telephone-84 %). For clarification. IADLs are consider tasks for which higher levels o f functional ability are required (e.g.. shopping, meal preparation, house-keeping, managing money, etc.). Agreement was high for three other comparative IADL measures: telling time (82%), cooking light meals (90 %). and cooking full meals (80 %). When a discrepancy occurred for these items, individuals more commonly underestimated their ability (walking, telling time, cooking light and heavy meals) although they did overestimate their ability to use the telephone directory. Kempen. Van Heuvelen et al. (1996) showed that among a community based sample of persons 57 and older, a moderate association exists between self-report and performance based measures. Within the domain of motor functioning (percent variance explained in reported functioning by performed functioning was 33.1%. p<0.001) and hearing (percent variance explained in reported functioning by performed functioning was 41.4%, p<0.001) a much stronger association existed than within the domain o f vision (percent variance explained in reported functioning by performed functioning was 12.5%. pO.OOl). Merrill et al. (1997) compared the performance o f three tasks to comparable self-reported items: I) reported ability to walk across a small room and the 5 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. observed ability to walk eight feet and back; 2) reported difficulty to stoop, crouch, kneel, and the observed ability to rise from a chair five times; and 3) reported the difficulty to reach or extend one's arms and observed ability to rotate one's shoulders. Although performance-based measures were initially in a metric o f seconds, they were re-classified into a metric for direct comparison with the reported measure (i.e.. unable/able). Despite the fact that these analyses were stratified by gender, all three comparisons had agreement levels greater than 80%, except that specific to lower body strength (stooping/repeated chair stands) for women (69.8%). Furthermore, for all three comparisons, over-reporting of disability was more common than under-reporting, especially among women. More recently. Ferrer et al. (1999) compared measures o f gait (i.e., the reported difficulty to walk across a room with the observed ability to walk four meters) and lower body strength (i.e.. the reported difficulty to stand up and sit down from a chair with the observed ability to rise five times from a chair). After the performance-based measures were re-classified for comparison with the reported measures (from a metric o f seconds to a dichotomous variable unable/able), agreement between these measures were relatively high: 86% for lower body strength and 78% for gait ability. Furthermore, they found that women more frequently over-report on these items than men. Two studies have examined the relationship between reported and performed measures o f physical functioning using correlation coefficients (Cress et al., 1995; Reuben et al.. 1995). Cress et al. (1995) compared four continuous sub-scales from the Sickness Impact Profile (SIP) (e.g., body care and movement, ambulation, mobility, and physical dimension) to four performance-based measures (e.g., balance, three 5 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. repeated chair stands, grip strength, and gait speed). These performance-based measures also had continuous scores, see Table II-2. They found moderate to high correlations between these measures such that individuals reporting poorer functioning had worse performance (r = -0.19 to r = -0.63). Reuben et al. (1995) compared three reported questionnaires (as assessed by the Medical Outcome Survey, MOS SF-36 and the Functional Status Questionnaire. FSQ) to the seven item Physical Performance Test (PPT). They found high correlations between reported and observed measures indicating that better reported functioning was associated with better performed functioning (FSQ/BADL with PPT r =0.55. p<0.05; FSQ/IADL with PPT r =0.45. p<0.05; and MOS SF-36 with PPT r =0.26. p<0.001). Among hospitalized patients, Elam et al. (1991). Sager et al. (1992), and Dorevitch et al. (1992) identified rates o f agreement on individual and summary ADL items o f comparison. Elam et al. (1991) found that rates o f agreement for five individual ADL items (e.g.. walking, telephone use, transferring, eating, and dressing) varied. Rates o f agreement ranged from 50.0% for dressing to 90.4% for walking. For these comparisons, reported and performed measures were classified into categorical variables (dependent-can’t do or need assistance, or independent-able to do without assistance from other persons). Unfortunately, they did not examine overall rates o f agreement and when a disagreement occurred, the direction o f the discrepancy is not clear. Sager et al. (1992) also found varying rates of agreement for five individual (i.e.. eating, "toileting.” transferring, dressing, and bathing) and summary ADL measure, see Table II-2. For their study, both reported and performed items were classified categorically (e.g., independent, need help, unable to do). For individual items, rates of agreement ranged from 59 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 63% for bathing to 94% for eating. Items that may require lower body function had lower rates o f agreement (e.g.. dressing-64%; bathing-63%) than those that encompass upper body function (eating-94%). Overall, rates o f agreement for these five ADL items was 78%. Unfortunately, from their study, the direction o f the discrepancy is unclear. Dorevitch et al. (1992) compared 13 reported and performed items of ADLs. These items were classified as independent (performed alone) or dependent (needed assistance or unable to perform). Agreement between these reported and performed items ranged from 67% for washing to 100% for drinking, see Table II-2. Similar to the study by Sager et al. (1992), in general, high rates of agreement were found for items that required upper body function (e.g.. drinking-100%; eating-100%; dressing upper body-94%; grooming 94%. chair and toilet transfers-94%) and slightly lower rates o f agreement were found for measures requiring lower body function (e.g.. walking 50 meters-87%; walking up/down a flight of stairs-77%). Overall, rates o f agreement between these 13 reported and performed ADL items was 88.9% (6.1% over-reported and 5.0% under-reported). Researchers using data from population samples have also found moderate associations between reported and performed measures o f physical functioning. Jette and Branch (1985) used the third wave o f the Massachusetts Health Care Panel Study o f non- institutionalized elderly. They found a modest association between summary performance o f 10 gross body movements and six basic ADLs (b=0.30, p<0.001). Similarly. Kempen, Steverink et al. (1996) found that individuals 57 and older, from the Groningen Longitudinal Aging Study (GLAS), had a moderate association between a summary measure o f reported ADLs (as assessed by 11 items) and a summary score of three performance-based measures (putting on/taking off a jacket, walking six 60 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. meters, and five repeated chair stands) (variance explained in self-reported ADL by performance was 38.8%. p<0.01). As mentioned three studies investigating the relationship between self-reported and performance-based measures had longitudinal components (Myers et al., 1993; Mendes de Leon et al.. 1996; and Kempen. Sullivan et al., 1999). The study by Myers et al. (1993) found that among a community based sample o f persons 60 to 92 years o f age. ten o f the fifteen matched IADL items, concordance declined (less than 80% agreement), especially for walking and using the telephone directory (less than 60% agreement). Mendes de Leon et al. (1996) found that among community residents 72 and older, a moderate association exists between summary baseline physical performance (as assessed by five tasks: foot tapping, three chair stands, turning in a circle, twenty foot walk, bending over to pick up a pencil, and signing one's name) and summary reported ADL. Individuals with higher levels of physical performance at baseline had higher levels of ADL functioning at follow-up. after controlling for baseline ADL ( b = 0.029, p<0.00l). In a more recent study. Kempen. Sullivan et al., (1999) used data from the Groningen Longitudinal Aging Study (GLAS) to examine the strength of the longitudinal association between reported (summary score across 11 ADLs) and performed (summary score across three items; putting on, taking off jacket, walking 6 meters— including 180 degree turn at 3 meters, and standing up/sitting down 5 times from a kitchen chair without using one's arms) measures o f physical functioning. They found a moderate association between changes in self-reported physical functioning and performance-based physical functioning (r = 0.28). 61 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. F. Sources o f Discrepancies in the Cross-Section The moderate association found between levels o f self-reported and performance- based functioning suggests that the discordance between these measures may be due to other factors. Although the author has chosen the theoretical model proposed by Nagi because o f its clearer definitions o f individual components, the author agrees with the suggestion by Verbugge and Jette (1994) that risk factors, social environments and psychological characteristics can influence the progression toward greater dysfunction. Consequently, consideration of how other factors may influence the discrepancies between self-reported and performance-based measures of physical functioning is made. Studies examining the roles o f demographic, health, and psychological characteristics are reviewed here. 1. Demographic Characteristics Age. Studies have shown that age predicts physical decrement (Guralnik et al.. 1993; Boult et al.. 1994). Among a longitudinal sample of persons 65 and older from the Established Populations for Epidemiologic Studies. Guralnik et al. (1993) found persons 85 and older were associated with significantly lower rates o f maintaining mobility (on reported measures o f walking up and down stairs and walking half a mile) compared to those in the middle (75 to 84 years o f age) and low (65 to 74 years o f age) age categories. In addition, based on data from the Longitudinal Study of Aging (LSOA), Boult et al. (1994) found that persons in older age categories (75 to 79; 80 to 84; 85 and older) significantly predicted becoming decrement in functioning (as assessed by reported ADL measures) compared to those 70 to 74 years o f age. Although age is a significant 62 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. predictor o f decline in physical functioning, it may also predict physical improvement. Crimmins and Saito (1993) found that among a community residing persons 70 and older, physical functioning may improve among older individuals. Jette and Branch (1985) showed that age is related to both self-reported activities o f daily living (ADL) and performance o f gross body movements, such that when performance levels o f functioning are controlled, older persons reported lower levels o f functioning. Among community residents o f persons 63 to 94 years o f age, Kelly-Hayes et al. (1992) showed that individuals 75 and older over-report (i.e.. overestimate with higher reported levels o f functional limitation than performance levels) their dysfunction walking (as measured by Katz and Barthel Index) compared to younger individuals. In addition. Kempen. Van Heuvelen et al.. (1996) showed that among a community based sample o f persons 57 and older, older individuals, over-report on a summary measure of motor functioning (e.g., carrying 5 kg, picking up something from the floor, and walking 400 meters) and on individual measures of hearing and vision. In Spain, however. Ferrer et al., (1999) showed that individuals 75 and older under-report (i.e.. underestimate with lower reported levels o f functional limitation than performance levels) their dysfunction walking (as measured by ADL) compared to younger residents (72 to 74 years o f age). Thus, although age is related to both reported and performed measures o f physical functioning, the direction of observed discrepancies between these measurement types remains unclear. Gender. Although the incidence o f physical dysfunction is the same for men and women, the prevalence of physical dysfunction is higher for women than for men (Manton, 1988; Guralnik and Kaplan, 1989; Strawbridge et al., 1992; Kaplan et 63 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. al., 1993). In addition, although women appear more likely than men to decline on Nagi measures of physical activity and less likely on measures specific to ADL and IADL ability; Crimmins and Saito (1993) found that the recovery o f ability does not differ by gender. These gender differences may result from differences in health practices and health care utilization. Discrepancies in the direction o f inaccurate reporting has been shown with respect to gender. Jette and Branch (1985) found that gender remains significantly associated with reported measures o f ADL functioning when performance-based measures of gross body movements are controlled. Among community residents o f older persons in the United States (Kelly-Hayes et al.. 1992; and Merrill et al., 1997) and Spain (Ferrer et al.. 1999). studies have shown that, compared to men. women inaccurately over-report their dysfunction walking (Kelly-Hayes et al., 1992, Ferrer et al.. 1999). In addition, the study by Merrill et al. (1997) shows that on summary measures o f dysfunction on ADL (Katz et al.. 1963). range o f motion (Nagi, 1976) and gross mobility (Rosow and Breslau. 1966) items, women also over-report dysfunction, compared to men. Furthermore. Kempen. Van Heuvelen, et al. (1996) show that among a community population o f persons 57 and older, women over-report their level o f dysfunction on a summary measure o f reported motor function (assessed by the reported ability to carry 5 kg. pick up something from the floor, and walk 400 meters). The direction of the discrepancy based on gender, however, does not seem consistent. For instance, although these studies suggest that women over- report dysfunction walking, Ferrer et al. (1999), also demonstrated that women under report dysfunction. Specifically, on the reported ability to "stand up and sit down from a chair” women under-report significantly less "difficulty” than men. These mixed 6 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. results thereby suggest that the inaccurate reporting o f dysfunction may reflect the higher rates o f dysfunction among women instead of the differences in reporting by gender. Economic Status. Economic status has also been shown to influence physical functioning. Those with lower family income have poorer physical functioning on a summary score or reported measures (e.g.. BADLs, climbing stairs, walking half a mile, pushing/pulling large objects, stooping/kneeling, crouching, lifting/carrying ten pounds, gardening, meal preparation, heavy housework, shopping, and getting places) (Guralnik and Kaplan. 1989). In addition, among a longitudinal sample of persons 65 and older from the Established Populations for Epidemiologic Studies, Guralnik et al. (1993) found that persons with lower (< $5,000 per year) and middle ($5,000 to $9,999 per year) income levels were associated with significantly lower rates of maintaining mobility (on reported measures o f walking up and down stairs and walking half a mile) than persons with high (> $10,000 per year) income levels. In their study, persons with low (< 8 years) and medium (9 to 12 years) levels o f education were also associated with lower rates o f maintaining mobility compared to those with more than 12 years o f education. Research has also shown that individuals with higher education levels are less likely to lose ability on measures o f physical functioning as assessed on four reported measures (e.g., walking one quarter o f a mile, stooping, crouching, or kneeling, lifting ten pounds, and walking up ten steps without resting) (Harris et al.. 1989). Proxies for economic status such as education and social class may explain discrepancies between reported and performed measures of physical functioning. With respect to education, Guralnik, Branch et al. (1989) stated that performance based 6 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. measures are less sensitive to education levels than reported measures. Supporting Guralnik's position. Rozzini et al. (1993) found that although a modest negative correlation exists between higher education levels and reported functional limitation on a summary ADL scale; the relationship between education and summary measures (based on the modified Physical Performance Test, PPT) is not significant. Furthermore, Kempen. Van Heuvelen, et al. (1996) found that among persons 57 and older, those with higher levels o f education reported lower levels on a summary measures o f motor functioning (e.g., carrying 5 kg. picking up something from the floor, and walking 400 meters) compared to a summary measure of performance (e.g., gait, flexibility o f the hip and spine, flexibility o f the shoulder, balance, and grip strength). When comparing individual items. Ferrer et al. (1999) showed that among community residents 72 and older, individuals with higher levels o f education tend to under-report their dysfunction to walk (as assessed by ADL). Significant associations are also found between social class and both reported and performed measures o f physical functioning. Among a nationally representative sample o f Swedish citizens 77 and older, a significant association between social class and both reported and performed measures o f physical functioning was demonstrated by Parker and colleagues (1994), such that former white-collar workers have higher levels of functioning than blue collar workers. Together, these findings suggest that economic status can influence both reported and performed measures leading to a discordance between measurement type. Furthermore, the literature suggests that higher levels of education reduces the discrepancy between reported an performed measures o f physical functioning. 66 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2. Health Characteristics Reported Health Status. The studies by Kelly-Hayes et al. (1992) and Ferrer et al. (1999) also showed that the discordance between reported and performed measures o f physical functioning can result from lower reported assessments o f health. Among communities of older persons these researchers found that individuals who more negatively rate their health tend to over-report their dysfunction walking (as measured by ADL). Rozzini et al. (1993). however, showed that among a community o f Northern Italian elders 70 and older, the number o f diseases participants reported was not associated with either dysfunction on summary reported (ADL) or performed measures (PPT). Thus, the role o f health characteristics on the discrepancies between reported and performed measures o f physical functioning remains unclear. Research does, however, show that among community dwelling older persons, a larger proportion of people with activity limitations have difficulty with physical tasks o f daily life and that the proportion o f people with activity limitations is increased among those who report having chronic diseases (Harris et al.. 1989: Lammi et al.. 1989). Furthermore, studies suggest that specific disease conditions influence decrement in physical functioning (Satarino et al., 1990; W olf et al., 1992; Guralnik et al., 1993; Kurtz et al., 1993; Ettinger et al.. 1994; Mulrow et al., 1994; Ormel et al.. 1998; Kurtz et al.. 1999). Presented here is a review o f these studies investigating the effect o f specific diseases on physical functioning. Several studies have compared the role o f various disease conditions and injuries on physical functioning (Guralnik et al., 1993; Ettinger et al., 1994; Mulrow et al.. 1994; Ormel et al.. 1998). In 1993, Guralnik et al. used longitudinal data from the 6 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Established Populations for Epidemiologic Studies o f the Elderly from 1981 to 1987 to assess the role o f chronic conditions in maintaining mobility among persons 65 and older. They found that after adjusting for demographic factors, reports o f previous heart attack, stroke, high blood pressure, diabetes, dyspnea, and exertional pain were associated with significant risks for mobility loss [as assessed by two items from the self-reported Rosow and Breslau scale (1966): walking up and down stairs and walking half a mile]. In 1994. Ettinger et al. used data from community dwelling persons 65 and older to determine the major conditions reported to cause difficulty in 17 physical tasks (e.g.. walking half a mile, walking around the house, getting out of a bed or chair, walking up ten steps, doing heavy housework, shopping, preparing meals, paying bills, using the telephone, eating, dressing, bathing, using the toilet, lifting or carrying 10 pounds, reaching out with your arms, and gripping with your hands). They found that arthritis and other musculoskeletal diseases were the primary cause of difficulty in performing physical tasks (49.0%). followed by heart disease (13.7%). injury (12.0%), lung disease (6.0%), and stroke (2.9%). Mulrow et al. (1994) used data obtained from older nursing home residents (greater than 60 years of age) without severe cognitive impairment to determine the associations between disease and performed (as assessed on an ADL scale) and reported (as assessed by the Sickness Impact Profile, SIP) functioning. They found that although the 24 most frequent disease categories did not explain a significant amount o f the SIP. these diseases explained a significant amount o f ADL (r2 = 0.25. p=0.001). The 24 diseases included: infections, hypertension, cerebrovascular disease, arthritis, dementia, depression, injuries/fractures, heart failure, ischemic heart attach, diabetes, gastrointestinal, skin decubiti, arrhythmias, anemia, bladder/prostate dysfunction. 68 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. osteoporosis, peripheral vascular disease, obstructive lung disease, renal disease, surgeries, seizures, thyroid disease, cancer, and Parkinson’s disease. Ormel et al. (1998) used data from the Groningen Longitudinal Study, comprising o f persons 57 and older, to show that specific disease conditions influence decrement on two measures o f physical functioning. Persons with diabetes, high blood pressure, stroke and heart disease had poorer physical functioning on the Medical Outcome Survey Short-Form (SF-20). than persons without these conditions. In addition, persons with heart disease, stroke, and diabetes had poorer physical functioning (as assessed by the 18 item Groningen Activity Restriction Scale which measures ADL/IADLs) than those without these conditions. The scientific literature also provides evidence that specific disease conditions (e.g., stroke, cancer) are associated with physical dysfunction (Satarino et al.. 1990; Wolf et al., 1992; Kurtz et al.. 1994; Duncan et al., 1997; Kurtz et al., 1999). Specific to the condition of stroke. Duncan et al. (1997) found that among a community residing sample of persons 65 and older, patients with mild stroke are significantly more impaired in physical functioning (as assessed by the MOS-36) than those with transient ischemic attacks. In addition, the Framingham Study found that stroke survivors had significantly greater physical dysfunction (as assessed by reported ADL) than matched control subjects (W olf et al.. 1992). Cancer is also associated with physical difficulties. In a case controlled study of difficulties in physical functioning among middle-aged and elderly women with breast cancer. Satarino et al. (1990) found that patients experienced greater difficulties than controls in completing tasks requiring upper body strength. In their study, physical functioning was assessed through reported measures on ten physical activities (e.g., pushing heavy objects, stooping, lifting items under 10 pounds, 6 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lifting items over 10 pounds, reaching arms above and below shoulders, writing or handling small objects, standing in place for 15 minutes or longer, sitting for long periods (1 hour), walking alone, walking up and down a flight of stairs, and walking half a mile without help). In a longitudinal study of cancer patients. Kurtz et al. (1993 ) found that loss o f physical functioning [as assessed by the Medical Outcomes Study (MOS 36) Item Short Form Health Survey (SF-36) that measures the degree o f limitation in activities such as lifting or carrying groceries, bending, kneeling, or stooping, walking one block, bathing, dressing, etc.] was associated primarily with a composite measure o f symptoms and to a lesser degree with age. Decrement in physical functioning varied significantly according to cancer site, with higher levels for patients with lung cancer and lower levels for patients with breast or colorectal/gastrointestinal cancers. This finding has been supported by a more recent study. In 1999, Kurtz et al. found that among a sample of women 65 and older, lung cancer patients reported greater losses in physical functioning (as assessed by the MOS 36 and SF-36) than either breast or colon cancer patients. In addition, age, comorbidity, symptom severity were significant predictors o f physical dysfunction. No one has examined the role o f specific disease conditions or events such as bone fractures on the discrepancy between reported and performed measures o f physical functioning. This study will examine the role of specific disease conditions (cancer, diabetes, high blood pressure, myocardial infarction, and stroke) and events (hip fractures and non-hip fractures) on the observed discrepancies between reported and performed measures o f physical functioning. 7 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cognitive Functioning. The lack of concordance between reported and performed measures o f physical functioning can also precipitate from problems o f memory or judgment (Branch and Myers, 1987). Furthermore, Guralnik. Branch et al. (1989) suggested that performance-based measures are less sensitive to poor cognitive functioning. Researchers provided support for these positions. Among hospitalized persons 70 and older. Sager et al. (1992) found that individuals who scored 24 or lower on the Mini Mental Status Exam. MMSE (i.e., cognitively impaired) have higher rates of disagreement across 5 ADLs (35%) compared to those scoring higher than 24 on the MMSE (18%). The direction o f the disagreement is unclear. The study by Kelly-Hayes et al. (1992). however, showed that the higher rates o f disagreement among cognitively impaired older persons was in the direction of over-reporting functional limitation. Kelly-Hayes et al. (1992) found that individuals who scored 23 or lower on the MMSE (i.e.. impaired) over-reported dysfunction walking (as measured by Katz and Barthel Index) compared to those who scored higher than 23 on the MMSE. Among individuals with higher cognitive functioning, however, the direction o f the discrepancy is reversed. Kempen. Van Heuvelen et al. (1996) and Kempen. Steverink et al. (1996) showed that among individuals 57 and older, from the Groningen Longitudinal Aging Study (GLAS), those with higher cognitive ability under-report on measures o f physical dysfunction. In the first study. Kempen Van Heuvelen et al. (1996) found that individuals with higher verbal intelligence (as assessed by the National Adult Reading Test) under-report dysfunction on a summary measure of motor functioning (e.g., carrying 5 kg. picking up something from the floor, and walking 400 meters) and on an individual measure of vision. In addition, persons with higher levels o f memory (assessed by a verbal 71 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. learning test) under-report dysfunction on vision, as well. In the second study. Kempen. Steverink et al. (1996) found that individuals with higher levels of memory and verbal intelligence under-reported dysfunction on the ADL item o f “getting around the house."’ In addition, persons with higher levels o f memory also under-reported dysfunction dressing. Evidence o f positive associations between cognitive functioning and both reported (Barberger-Gateau. Commenges. et al.. 1992; Rozzini et al.. 1993; Cress et al.. 1995) and performed (Rozzini et al., 1993; Cress et al.. 1995) measures o f physical functioning also exists. Barberger-Gateau, Commenges, and colleagues (1992) used data from the Personnes Agees QUID (PAQUID). an epidemiologic project o f community residents 65 and older residing in Southewestern France, to identify which instrumental activities o f daily living (IADL) items are more related to cognitive impairment (as assessed by the MMSE). Independent of age. sex. and education, individuals who were partly dependent on four IADL items (i.e.. telephone use. use of means of transportation, responsibility for medication intake, and handling finances) were positively associated with cognitive impairment (defined as an MMSE score less than 24). compared to those who are fully independent. Also relying on the MMSE, Rozzini et al. (1993) showed that cognitive status is positively associated with reported (ADL and IADL) and performed (PPT) summary measures o f dysfunction independent of demographic characteristics and indicators o f health. Among nursing home residents. Cress et al. (1995) found that MMSE scores are independently associates on three o f four sub-scales o f the Sickness Impact Profile (SIP): body care and movement, ambulation, and a summary scale o f physical dimensions. In addition, they found that MMSE scores were positively 7 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. correlated with poor performance on gait speed (participants were asked to walk a self selected speed from point to point with no turn around). Together, these studies suggest that cognitive functioning seems to influence reported and performance based measures of physical functioning. In addition, they suggest that whereas individuals with impaired cognitive functioning tend to over-report their level o f functional limitation, those with higher cognitive functioning under-report their level o f physical dysfunction. Thus, higher levels o f cognitive functioning seem to reduce the discrepancies between reported and performed measures of physical functioning. Depression. Several studies have shown that older persons with depressive symptoms reported higher levels o f physical dysfunction than those without these symptoms while controlling for levels o f performance-based functioning (Cress et al.. 1995; Kempen, Van Heuvelen et al.. 1996; Kempen. Steverink. et al.. 1996). Among a community population of nursing home residents 62 to 98. depressive symptomatology independently predicted poorer reported functioning on four sub-scales o f the Sickness Impact Profile (SIP) (i.e.. body care and movement, mobility, ambulation, and physical dimensions) after controlling for physical performance (Cress et al., 1995). Examining three domains o f physical functioning (i.e.. motor functioning, hearing and vision). Kempen. Van Heuvelen et al. (1996) found that among a community based sample o f persons 57 and older, discrepancies between reported and performance based measures are higher for individuals with depressive symptoms (as assessed by the Hospital Anxiety and Depression Scale) across all three domains o f physical functioning. Specifically, for two domains o f functioning (motor functioning and hearing) older persons with 73 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. higher levels o f depression report greater levels o f dysfunction compared to performance. In addition. Kempen. Steverink et al. (1996) found that among these same individuals, those with high levels o f depression over-reported their dysfunction on a summary ADL measure o f 11 items and on the individual ADL item of "getting around the house.” Furthermore, although some researchers have documented positive associations between depressive symptoms and reported functioning (Griffiths et al. 1987: Kempen and Suurmeijer, 1991: Rozzini et al., 1993). others have failed to identify an association between depressive symptomatology and performed functioning (Rozzini et al., 1993: Kempen, Steverink et al., 1996). Furthermore, the study by Sager et al. (1992) showed that depressed hospitalized persons 70 and older tended to have lower rates o f agreement between reported and performed measures across 5 ADLs. although these differences were not significant. Together, these results suggest that depressive symptoms are more related to reported than performance measures o f physical functioning and that depressed persons over-report their level of dysfunction, increasing the observed discrepancy between reported and performed measures. 3. Psychological Attributes Self-Efficacv and Mastery. Self-efficacy and mastery have also been found to influence the discordance between reported and performed measures o f physical functioning. Self-efficacy is defined as the degree o f confidence that individuals have in their ability to perform specific activities successfully (Bandura. 1986). According to Bandura's social cognitive theory (Bandura 1977 and 1986), perceived self-efficacy influences the likelihood that a given behavior will be undertaken. Related to 74 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. self-efficacy is the concept o f mastery. Pearlin and Schooler (1978) first defined mastery as a construct, similar to control, protecting individuals from life strains and stressors. Roberts et al. (1994) and Schoenfeld et al. (1994) stress the importance o f a sense of mastery or personal control for good functioning in very late life. Researchers have explored the role o f these psychological resources in explaining the discrepancies between reported and performance measures of physical functioning. Cross-sectional studies show an association between self-efficacy and physical functioning (Berkman et al., 1993; Tinetti et al.. 1994; Kempen, Steverink et al.. 1996) and between perceived control and physical functioning in older persons (Kempen. Steverink et al.. 1996; Kempen, Van Heuvelen. et al.. 1996). Among a community sample of high functioning Americans 70 and older. Berkman et al. (1993) examined the psychological characteristics o f persons at different levels o f physical and cognitive functioning. In their study individuals were categorized as "low.*'' "medium," and “high" functioning based cognitive and physical functioning (both reported and performed) measures. They found that compared to individuals in the "low" or "medium" functioning groups, "high" functioning individuals have higher levels o f self-efficacy (assessed by the scale developed by Rodin and McAvay. 1993) and mastery (assessed by Pearlin and Schooler, 1978). Among a probability sample o f community residents 72 and older. Tinetti et al., (1994) found a positive association between self-efficacy (measured on the Falls Efficacy Scale) and ADL/1ADL dysfunction, while controlling for physical activity. Furthermore, a couple o f studies by Kempen and her colleagues (Kempen, Steverink et al. 1996 and Kempen, Heuvelen et al.. 1996) show that these psychological resources also influence the discordance between reported and performed 75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. measures o f physical functioning. In the first study. Kempen, Steverink, et al. (1996) demonstrated that individuals with higher levels o f mastery tend to under-report on individual and summary measures o f ADL dysfunction: dressing, "getting around the house. ” and across a summary o f 11 ADL items. Furthermore, they found that individuals with higher levels o f self-efficacy under-report dysfunction on the summary measure o f 11 ADL items and on 3 individual ADL items: dressing, getting around the house, and standing up form sitting in a chair. In their second study. Kempen. Van Heuvelen et al.. (1996) found that controlling for levels o f performance-based functioning, older persons with higher levels of mastery (as assessed by the Pearlin Mastery Scale) report higher levels of physical functioning (specific to motor and hearing) compared to performance. In summary, these research findings suggest that higher levels o f mastery and self-efficacy reduce discrepancies between reported and performed measures of physical functioning. G. Sources o f Discrepancies Over Time Most o f the research on the associations between self-reported and performance- based physical functioning have been in the cross-section. More recently, however, a single study by Kempen. Sullivan, et al. (1999) has provided information on the role o f depressive symptomatology on the change in reported and performed measures of physical functioning. In addition, some researchers have provided information on how- psychological attributes influence change in reported and performed measures of physical functioning. Findings from these studies are detailed here. 7 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1. Depression Kempen, Sullivan et al. (1999) have examined the effect o f depressive symptoms on the congruence o f change in reported and performed measures o f ADL. They used data from a sub-sample o f the Groningen Longitudinal Aging Study (GLAS) to examine whether I) change in summary reported physical functioning (from an 11 item ADL subscale) is predicted by initial levels of and change in depressive symptoms (assessed by the Hospital Anxiety and Depression Scale-HADS); and 2) congruence in measured change between summary reported and summary performed measures o f 3 tests (putting on. taking off jacket, walking 6 meters— including a 180 degree turn at 3 meters, and standing up/sitting down 5 times from a kitchen chair, without using one's arms) of physical functioning is related to initial levels and change in depression symptoms. A significantly stronger association was found between change in depressive symptoms and change in reported functioning (r=0.26) than between change in depressive symptoms and change in performance (r=0.13). Furthermore, controlling for age. gender, baseline number o f chronic conditions and reported functioning, they showed that although baseline levels o f depressive symptoms are not predictive o f change in reported measures o f physical functioning, change in depressive symptoms are predictive. Thus, the strength o f the association between change in reported and performed physical functioning depends on the change in depressive symptoms. Finally, they show that higher congruence is observed between change in reported and performed measures of physical functioning for those with increased depressive symptoms. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2. Psychological Attributes Some researchers suggest that psychological levels of mastery and self-efficacy act to slow the decline in reported measures o f physical functioning (Mendes de Leon et al., 1996; Femia et al.. 1997; Kempen et al.. 1999). Seeman et al.. (1995). however, found that these attributes did not influence summary performance measures o f physical functioning. Thus, although psychological attributes seem to slow the progression or onset o f reported dysfunction, they do not seem to influence decrement on performance- based measures o f physical functioning. Femia et al. (1997) used data from a longitudinal panel study in South Central Sweden to examine the role o f psychological variables in the stability or decline in summary ADL functioning. A negative association was found between mastery (as assessed on the Pearlin Mastery Scale) and subsequent levels of ADL dysfunction. Furthermore, among a community o f residents 72 and older. Mendes de Leon et al. (1996) showed that among individuals who decline on a summary measures o f physical performance, individuals with higher levels o f self-efficacy (as measured by the Falls- Efficacy Scale, at baseline) had better reported functioning on measures o f summary ADL over time, compared to those with lower levels o f self-efficacy. More recently. Kempen, VanSonderen et al., (1999) demonstrated that independent of age and gender, community residents 57 and older who had higher levels o f mastery and self-efficacy did not significantly increase on an 18 item scale o f ADL and IADL dysfunction. In contrast to these findings, Seeman and colleagues (1995). provide evidence that self-efficacy (assessed by the scale developed by Rodin and McAvay, 1993) and mastery (assessed by Pearlin and Schooler, 1978) do not influence change over two and a 7 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. half years on a summary performance-based measure o f physical functioning. Using data from the MacArthur Research Network on Successful Aging Community Study, a longitudinal cohort study o f high functioning Americans 70 and older, these psychological characteristics were not predictive of improvement (or decrement) in physical functioning based on a summary physical performance measure o f five items (i.e., based on balance, gait, 5 repeated chair stands, foot taps, and signing one's name). These findings together suggest that although performance-based measures are not influenced by psychological characteristics, individuals with higher levels o f self-efficacy and mastery can slow the progression o f decrement on reported measures o f physical functioning and reduce discrepancies between change in reported and performed measures of physical functioning. H. Other Potential Risk Factors: Behavioral Characteristics Although it remains unclear whether health behaviors influence the discordance between self-reported and performance-based measures, as some o f these behaviors have been shown to influence various measures o f physical functioning, they are also worthy o f consideration. The three health behaviors discussed here are 1) body composition. 2) smoking, and 3) alcohol consumption. Body Composition. Research suggests that with aging, changes in body composition (i.e., body fat) (Baumgartner et al., 1995; Mazariegos et al., 1994) place individuals at increased risk of dysfunction. A high body mass index (BMI) is an indicator o f obesity. This index is commonly referred to as the weight in kilograms divided by the height in meters squared. High BMI may influence the risk for 79 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dysfunction in several ways. First it may increase the wear and tear on various joints or reduce the flexibility o f movement (Launer et al.. 1994). It may also serve as a proxy for inactivity and disuse (Rissanen et al.. 1991) which can lead to a reduced capacity in neurological and musculoskeletal systems and functional decline (Buchner and Wagner. 1992). High BMI may also increase the risk of specific diseases that may influence the subsequent development o f dysfunction. Researchers found that a high body mass index is associated with an increased risk o f functional limitation in older men and women (LaCroix et al.. 1993; Launer et al.. 1994; Nelson et al.. 1994; Clark et al.. 1998; Visser. Langlois et al. 1998). Data from the Established Populations for Epidemiologic Studies o f the Elderly (EPESE) showed that among both men and women 65 and older, individuals in the highest BMI quintile are associated with an increased risk oflosing mobility compared to those with moderate BMI levels (i.e.. a BMI score falling within the middle three quintiles) (LaCroix et al.. 1993). In their study, mobility is measured by the reported ability to walk half a mile and walk up/down a flight o f stairs without help. In 1994. Launer and colleagues showed that based on data from the National Health and Nutrition Examination Survey-1 (NHANES- 1), women 60-74 with high past (from 1971 to 1975) BMI (greater than 28.1) have a twofold increase in the risk for functional limitation compared to women with low past BMI (< 23.8). In this study, physical dysfunction is defined as reports o f any difficulty in executing at least one of the following activities: walking a quarter o f a mile, walking across a room, climbing two steps, doing heavy chores, carrying a full bag o f groceries, running errands, bending to the floor, or transferring from a car. bed. bath, chair or toilet. 80 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Nelson and colleagues (1994) found that among community residing women 65 and older, higher BMI is positively associated with performance measures o f strength (grip, triceps, knee extension, and hip abduction). Furthermore, they found that higher BMI is negatively associated with the performance measures of foot tapping, walking speed, stepping up on a 23 cm high step (without the use o f arms) and a summary measure of physical and IADL dysfunction (e.g.. walking, climbing/descending stairs, preparing meals, performing housework, and shopping). More recently. Clark et al. (1998) used information form the Health and Retirement Survey (HRS) to identify whether BMI has a direct or indirect effects on the onset o f mobility difficulty. Among persons 51-61, difficulty with the following reported mobility items were examined: I ) walking one block. 2) walking several blocks, and 3) climbing one flight of stairs without resting. They found that BMI has one o f the strongest direct effects on the onset of lower body difficulty. Individuals at or above 140% o f their ideal body mass are 2.1 times as likely to experience difficulty onset as those with a moderate BMI level (between 91 and 139 percent o f their ideal b o d y mass). These studies have all relied on indirect (BMI) measures o f body composition. In a recent study by Visser. Langlois et al. (1998) body composition was directly measured through the use o f bioelectrical impedance to examine whether high fat mass is associated with an increased risk o f dysfunction (a summary measure o f the reported difficulty in walking half a mile and walking up 10 steps). Among participants o f the Cardiovascular Health Study (65 and older), they found that both men and women in the highest quintile o f fat mass at baseline were more likely to deteriorate on their summary 81 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. measure o f physical functioning after three years than were those in the lowest quintile. Smoking. Cigarette smoking has also been shown to predict mobility limitations (LaCroix. et al., 1993; Nelson et al.. 1994; and Clark et al.. 1998). As mentioned earlier. La Croix et al. (1993) used data from the EPESE to show that current smoking is predictive o f the onset of mobility limitations for both men and women. Furthermore. Nelson and colleagues show that after controlling for age, history of stroke, BMI, physical activity, and current and past alcohol use. women 65 and older who currently smoke have poorer functioning on the following performance based measures compared to those who have never smoked: triceps strength, knee extension, hip abduction, stepping up a stair 23cm high, foot tapping, walking 6 meters, five repeated chair stands, tandem stand, and tandem walk. For example, smokers had a 0.30 kg lower triceps strength and a 0.69 kg lower knee extension strength compared with never smokers. Current smokers also have worse reported physical functioning (ADL) scores that those who never smoked. More recently, data from the HRS suggests that smoking has a direct effect on the onset of mobility limitations such that compared to those who have never smoked, current smokers are 1.7 times as likely to experience difficulty onset: 1) walking one block, 2) walking several blocks, and 3) climbing one flight o f stairs without resting (Clark et al.. 1998). Alcohol Consumption. Researchers have also investigated the effect o f alcohol intake on change in physical functioning. Guralnik and Kaplan (1989) showed that individuals, among a representative sample o f community residents (65 to 89 at follow- up), individuals who consume moderate amounts of alcohol at baseline (1-60 drinks per month), are 2.5 times more likely to be classified as “high” functioning, compared 82 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to those who consume high amounts of alcohol (> 60 drinks per month). Here physical functioning was defined as a summary measure of reported basic ADL, IADL. and other physically oriented tasks (i.e.. difficulty with household tasks, exercise, and recreational activities). Nelson and colleagues (1994) showed that among women 65 and older, nondrinkers (including former and never drinkers) have poorer reported (on summary ADL measure) and performed (i.e.. on grip strength, triceps strength, knee extension, hip abduction, stepping up. foot tapping, walking 6 meters. 5 repeated chair stands, tandem stand, and tandem walk) physical functioning compared to current moderate drinkers after controlling for age. BMI, history of stroke, physical activity, and current and past smoking status. Furthermore. Pinsky et al. (1989) found that among the participants of the Framingham Study (35-68 years o f age), alcohol intake (ounces/month) was related to good functioning (based on the Cumulative Disability Index based on 18 reported measures o f physical functioning - 4 ADL items, 2 from Branch et al. (1984). 3 Rosow and Breslau items, and 9 Nagi items) in men after controlling for age. cigarette smoking, education, heart size, hematorcrit. systolic blood pressure, and ventricular rate. Together, these studies show that health behaviors (e.g.. body composition, smoking, and alcohol consumption) influence reported and performed measures of physical functioning. How these factors influence discrepancies between reported and performed measures o f physical functioning, however, has yet to be explored. I. The Relationship Between Cognitive Impairment and Physical Functioning A third aim o f this investigation is to examine the relationship between cognitive impairment and physical functioning. Two o f the most feared states o f the elderly 83 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. are cognitive and physical dysfunction because they can lead to both social and physical dependency. Research on the relationship between cognitive and physical performance, however, is scarce. The causes of cognitive impairment in the elderly include dementia, delirium, toxic effects of medication, trauma, and psychiatric illness (Barberger-Gateau and Fabrigoule et al.. 1997). Cognitive impairment and related diseases are generally associated with aging (Brody and Scheider. 1995). although a meta-analysis of epidemiological studies of dementia showed that senile dementia should be considered as an ’'age-related' disorder occurring within a specific age range (Ritchie, 1995). Studies show that both the prevalence and incidence of dementia (including that caused by Alzheimer’s disease) increase exponentially with age (Letenneur et al.. 1994. and Letenneuret al.. 1993). Although physical dysfunction shows a similar pattern o f increase with age. it also seems that physical dysfunction is malleable. Data from the Established Populations for Epidemiologic Studies of the Elderly (EPESE) indicate that the proportion o f community-dwellers aged 65 and over reporting a need for assistance in performing basic activities o f daily living (ADL) generally increases with age (Foley et al., 1986). In addition, others have estimated that 5.0% to 8.1% of the total non-institutionalized elderly population in the United States aged 65 and over require help o f another person for at least one o f five ADLs (Wiener et al., 1990). The gain in physical dysfunction associated with age does not, however, seem irreversible. Findings from Crimmins and Saito (1993) suggest that a significant proportion of older persons with greater functional 84 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. limitations report improved functioning over time (Crimmins and Saito, 1993). This information suggests a certain plasticity to functional limitations. A body o f literature suggests that cognitive functioning may influence physical functioning and subsequent disability. Studies examining the validation o f instruments to assess physical functioning o f older persons include data on the correlation between scales o f physical functioning and mental status. Lawton and Brody's study (1969) found that the six ADL items o f the Physical Self-Maintenance Scale (PSMS) and the instrumental activities o f daily living (IADL) items had significant, moderate correlations with the Mental Status Questionnaire (MSQ). Fillenbaum (1985) developed a Guttman scale o f five IADL items that can be performed unaided. This Guttman scale was correlated with mental health in subjects 65 and older, as well as those 75 and older. In another study, MSQ was moderately correlated with activity limitations in ADL (Teresi et al., 1984). Validation studies using performance tests also show significant correlation with cognitive ability. Among geriatric psychiatric patients (Kuriansky and Gurland. 1976). performance tests requiring patients to demonstrate ability in selected ADLs were correlated with mental status. A Functional Activities Questionnaire determining the level o f performance in 10 activities predicted cognitive scores in subjects 61 to 91 years o f age (Pfeffer et al., 1982). Although the findings from these studies may simply represent the fact that items describing physical dysfunction were chosen to reflect functional limitations associated with cognitive dysfunction, studies using validated instruments confirm the association between physical and cognitive functioning. Among geriatric female inpatients, mental tests (using the MSQ) were correlated with an occupational therapist’s assessment 85 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o f seven ADL (Wilson et al.. 1973). This study showed that the intellectual dysfunction was the main impediment to self-care. In the EPESE study, a strong correlation between disability (assessed by a modified Katz scale) and tests o f cognitive functioning (evaluating memory, attention and orientation) persisted in multivariate models after the adjustment for sociodemographic and physical factors (Scherr et al.. 1988). Among a community o f French elderly residents, a highly significant correlation between cognitive functioning (assessed by the MMS) and dependency for the ADL. IADL, Rosow-Breslau, and mobility scales is shown (Barberger-Gateau et al., 1992). This association was independent o f demographic (age. sex, education, place o f residence), medical (dysponea. visual and hearing impairments), and depressive symptomatology (assessed by the CES- D). Rozzini et al. (1993) found that poorer cognitive status (as measured by Mini Mental State Exam-MMSE) was independently associated with poorer scores on the summary Physical Performance Test (PPT). This finding was recently confirmed by Binder et al. (1999) who also used the PPT but relied on a series o f psychometric tests (the Weschler Associated Learning and 20 minute Delayed Recall. Verbal Fluency test. Trailmaking A and B tests, and the Cancellation Random Figures test). In the cross-section, other studies provide less strong evidence o f a relationship between physical functioning and cognitive functioning measures. The Short Portable Mental Status Questionnaire (SPMSQ) was a poor predicator of self-care (assessed by three ADL) among nursing home patients (Winogiad, 1984). Among community residents 65 and older in the EPESE study, functional status assessed by ADL and mobility items did not show a significant relationship with cognitive functioning assessed by the SPMSQ; cognitive status correlated only with performance of IADL 86 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Galanos, 1994). Among demented individuals, the MMSE scores explained only a small part o f the variance in ADL and were poorer predictor of ADL for mildly demented patients than for severely demented patients (Reed et al., 1989). Despite these mixed findings in the cross-section, some longitudinal studies using reported measures and objective performance measures o f physical functioning suggest that cognitive functioning is related to physical functioning. Seeman et al.. (1994) found that individuals with better baseline cognitive performance (as measured in the MacArthur Study) were associated with improved summary physical performance score (based on five exercises— balance, walking a ten foot course at a fast pace, five repeated chair stands, foot-tapping, and signing one's name). Furthermore, studies have shown that decrement on summary measures o f reported ADL parallel decline on cognitive functioning (using the MMSE) (Cress et al., 1995; Rozzini et al.. 1993; Jagger, 1989). To the author’s knowledge, however, it remains unclear, whether cognitive performance might disproportionately influence more physically challenging or novel activities. Research suggests that performance on many information-processing tasks assessing cognitive or perceptual process is slowed with age (Cerella, 1985). Salthouse (1994) suggests that this slower processing speed explains lower scores on measures of working memory functioning with age. Based on the findings by Kaufman and colleagues (1989) that aging seems to decrease fluid intelligence (e.g., tasks requiring sensorimotor coordination, new learning, and speedy performance-Cattell, 1943) while preserving crystallized intelligence (e.g., tasks requiring language skills and the use o f established habits— Cattell, 1943); Albert and Kaplan (1980) suggest that decreased fluid intelligence is related to deficits in functions o f initiative, flexibly and planning. 87 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Consequently. Barberger-Gateau and Fabrigoule (1997) proposed a classification scheme for the physical performance o f older persons: a loss in efficiency for attentional demanding or novel tasks; and good performance on more automatic or over-learned skills. Only a single study by Griener et al. (1996) suggests that declines in cognitive functioning (using the MMSE) result in greater decrement on ADL. Using the longitudinal Nun Study. Sisters with low cognitive functioning (e.g.. MMSE scores 0 to 23) had an increased risk o f losing independence on more physically demanding ADLs relative to Sisters with high cognitive functioning (e.g.. MMSE scores o f 28 to 30): 9.6- feeding, 8.61-toileting. 6.2-dressing. 3.6 standing from a seated to an erect position. 2.7- walking, and 2.4-bathing. J. Risk Factors Specific to Cognitive Impairment and Physical Functioning The association between cognitive impairment and physical functioning can be explained by some confounding factors related to both. These factors include age. education, occupation, chronic conditions, depression, and health behaviors. Presented here is a summary o f existing research documenting the relationship o f these factors to cognitive impairment and physical functioning. Age. As physical functioning and cognitive dysfunction occur more frequently with age, their apparent association could be explained by the fact that they are frequently found to be associated in the oldest individuals. Education and Occupation. Other demographic factors that may influence the relationship between cognition and physical functioning are educational level and past occupation. In the Framingham Study, education was a significant independent 88 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. predictor o f functional status 21 years later in both men and women (Pinsky, 1987). In the PAQUID study (Personnes Agees QUID), education was associated with poorer performance on mental status tests in the elderly (Barberger-Gateau et al.. 1992). Furthermore, some past occupations may act as risk factors for developing certain chronic diseases associated with physical and cognitive dysfunction. In Great Britain, the prevalence o f physical dysfunction was found to be inversely related to social class, based on last occupation (Victor. 1989). The PAQUID study showed that farm workers, farm managers, domestic service employees and blue-collar workers had a higher risk of cognitive dysfunction after age 65, after controlling for educational level (Dartigues et al., 1992). Chronic Conditions. Illness is significantly more common in older populations (Office o f Human Development Services, 1981). Diabetes, and cardiovascular diseases seem to pose particular threats to the cognitive well-being of older adults. Diabetes mellitus is associated with cortical atrophy that may be related to mental performance deficits (Perros et al.. 1997). Research shows decrements involving attention, memory and repetition among young diabetics (Dey, 1997). Consequently, these changes are more likely to manifest themselves among older persons. Common among older persons, cardiovascular diseases may also explain the relationship between disability and cognitive dysfunction. At least 50% o f all elderly individuals are affected by hypertension (Lindholm, 1990), which is a risk factor for vascular dementia (Forette and Boiler, 1991). Even moderate hypertension on a sustained basis has been associated with intellectual disturbance (Guo et al., 1997; Palombo et al., 1997). Several longitudinal studies have identified hypertension as an independent predictor o f poor physical 8 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. functioning (Pinsky et al., 1985; Keil et al., 1989; Nikel and Chirikos. 1990), None o f these studies, however, included measures of cognitive performance in their analyses. Depression. Depression is frequently encountered in the elderly (Kay et al.. 1985; Livingston et al., 1990). Although in the EPESE study depression (assessed by the CES- D score) had no significant association with cognitive scores in multivariate models (Scherr et al., 1988), other research studies have found a significant correlation between depressive symptomatology and cognitive function (Kay et al.. 1985; Fuhrer et al.. 1992; Lichtenberg et al., 1995; Alexopoulous et al., 1996). Furthermore, others have found a correlation between depression and physical dysfunction (Winograd, 1984; Berkman et al.. 1986; Barberger-Gateau et al., 1992). Consequently, depression could be a major confounding factor in the relationship between physical functioning and cognition. Health Behaviors. The association between specific health behaviors and the relationship between cognitive and physical functioning remains controversial. Although smoking is related to cardiovascular diseases and subsequent physical dysfunction, the relationship between smoking and cognitive dysfunction is not clear. In many longitudinal studies, an association between smoking and increased levels o f physical dysfunction persists when adjusting for socio-demographic and medical factors, with some differences between men and women (Pinsky et al., 1987; Pinsky et al., 1985: Keil et al., 1989; Branch, 1985; Guralnik et al., 1989). Specific to cognitive dysfunction, however, Graves et al. (1991) found that tobacco consumption was associated with lower risk o f Alzheimer’s disease, Letenneur et al., (1994) found that the relationship between tobacco consumption and cognitive functioning disappeared after adjusting for 9 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. occupational categories. Individuals who belonged to occupations associated with poorer cognitive performances were less frequently smokers. Alcoholism represents one o f the most important health problems in the United States (West et al., 1984). Due to its bimodal age incidence pattern with a second peak occurring between the seventh and eighth decades o f life (Zimberg. 1987), it should be viewed as an age-related illness. Chronic and /or heavy ethanol use is classically associated with several cognitive disorders, most notably Korsakov's syndrome, alcohol hallucinosis, and hepatic encephalopathy (Keefover. 1998). The relationship o f alcohol consumption to both cognitive and physical functioning among the elderly, suggests an inverted U-curve. For instance, elderly subjects with moderate alcohol intake have higher physical functioning on ADL and IADL scales at follow-up 20 years later than those who abstained (Guralnik et al.. 1989b) while elderly men with histories o f heavy drinking experienced higher mortality rates and poorer ability to performance ADL and IADL compared to those who abstain (Colsher and Wallace. 1990). Furthermore. Cosher and Wallace (1990) found that heavy drinkers were associated with lower scores on a mental status examination while in the PAQUID study, moderate wine consumption was associated with a lower risk for cognitive deficits (compared to those who abstained) (Letenneur et al.. 1993). Research also shows that nutrition may influence physical and cognitive functioning (Nutritional deficiencies, such as from vitamin B12, are involved in several forms o f cognitive dysfunction or dementia and have been suspected risk factors for Alzheimer's disease (Abalan, 1984). As a measure o f nutritional status, the relative weight o f an individual has been found to be associated with physical functioning. 91 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Research suggests that with aging, changes in body composition (i.e., body fat) (Baumgartner et al., 1995; Mazariegos et al., 1994) place individuals at an increased risk o f disability. High body mass index, an indicator o f obesity, was found to be associated with an increased risk o f functional limitation in older men and women (LaCroix et al.. 1997; L auneretal.. 1994; Nelson et al.. 1994; LaCroix et al., 1993; Clark et al.. 1988). In addition, direct measures o f body fat were positively associated with self-reported disability in men and women 72 to 95 years o f age, independent o f chronic disease and physical activity (Visser et al., 1998). K. Conclusion This chapter began by providing a brief history o f functional assessment. Since the mid-1800s. the growing number of federal programs serving the disabled population has fueled a growing demand o f functional assessment. Unfortunately, however, the lack of a concise and commonly accepted definition o f the term "functional assessment'* has led to a surge in the development o f both reported and performed measures o f physical functioning. As discussed there are advantages and disadvantages to using both o f these measures o f physical functioning. Although reported measure are less costly to implement, the accuracy o f these measures can be compromised if they are not clearly defined. In contrast, performance-based measures that rely on objective standards of measurement have stronger face validity than reported measures. Performed measures, however, are costly to implement and may be limited by their dependence upon the subject’s motivation to perform. The four theoretical models examined to better understand the path toward physical dysfunction included the: 1) International 9 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Classification o f Impairments, Disabilities, and Handicaps (ICIDH, WHO, 1980), 2) Nagi framework (Nagi. 1965, 1969. 1976. 1977), 3) Disablement Process (Verbugge and Jette. 1994), and 4) recently revised draft version o f the ICIDH (ICIDH-2). Nagi (1965, 1969. 1976. 1977) suggests that functional limitations are defined as the inability o f the organism as a whole to perform activities such as walking, climbing, lifting, and reasoning. After reviewing these four theoretical models, a decision was made to treat self-reported and performance-based measures are defined as "functional limitations.” In contrast to the ideas proposed by Nagi (1965, 1969. 1976. 1977) and Verbugge and Jette (1994). self-reported measures within the domain o f ADL are included in this definition of "functional limitation.” Existing studies (in the cross section and longitudinally) suggest that a modest association exists between reported and performed measures o f physical functioning and that various demographic, health, and psychological characteristics can influence their concordance. The direction o f the disagreement between reported and performed measures o f physical functioning, however, remains unclear. Furthermore, as the vast majority o f these studies were conducted in the cross-section, a better understanding of the association between reported and performed measures o f physical functioning over time is needed. Thus the following two questions are asked: 1) What is the relationship between self-reported and performance-based measures over time? Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2) What demographic, health (conditions/events, cognitive, depressive symptomatology), health behavior, and psychological characteristics influence the observed discrepancy between self-reported and performance-based measures over time? Furthermore, although some studies have used summary reported measures when comparing reported and performed measures o f physical functioning (Cress et al., 1995; Reuben et al.. 1995; Kemen Van Heuvelen et al.. 1996; Kempen. Steverink et al.. 1996; Kempen, Sullivan et al.. 1999). we have yet to clearly understand how specific factors influence discrepancies with different domains of reported functioning. The three domains referred to here are the ADL. gross mobility (Rosow and Breslau. 1966) and physical activities (Nagi. 1976). Consequently the author asks: 3) How do specific factors influence discrepancies between self-reported and performance-based measures within different domains o f reported physical functioning (in the cross section and over time)? The final purpose o f this investigation is to examine the relationship between cognitive impairment and physical functioning. This aspect o f the investigation will advance knowledge concerning the characteristics o f older people in the United States who are at greater risk of decrement on specific physical tasks. Based on the findings of previous researchers, the author asks: 4) How does change in cognitive performance relate to change on an array of physical exercises when baseline physical performance, cognitive performance, demographic, health, and behavioral characteristics are controlled for? 94 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Together, the information extended from asking these questions can be used by professional health care workers, researchers, and policy makers to more effectively assess the functional limitations o f older persons. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter III Description o f Data and Measures A. Introduction This chapter begins by describing the data and sample used for the present study. The data used in this study come from the MacArthur Research Network on Successful Aging Community Study. Detailed information on how constructs of interest are measured is then provided. The descriptive statistics specific to change in self-reported and performance-based measures o f physical functioning are also shown. Finally, the demographic, health, behavioral, and psychological characteristics of the sample are discussed. B. Description o f Data and Sample Used The data used in the present study come from the MacArthur Research Network on Successful Aging Community Study, a longitudinal, three-site cohort study o f high functioning, Americans aged 70 to 79 in 1988. The MacArthur Study followed these persons over seven years and re-interviewed them in 1991 and 1995. Initially, individuals were sampled from three community-based studies o f individuals aged 65 and older in Durham. NC; East Boston. MA; and New Haven, CT (part o f the National Institute on Aging’s Established Populations for Epidemiologic Studies o f the Elderly - EPESE). Age eligible respondents were screened on the basis o f six criteria to identity a cohort o f individuals who function at the top third o f their age group in terms o f physical and cognitive functioning. These six criteria were 1) scores o f 6 or more correct on 95 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the 9 item Short Portable Mental Status Questionnaire; 2) remembering 3 or more of 6 elements on a delayed recall of short story; 3) reports o f no dysfunction on a 7 item scale o f activities o f daily living (ADL) (Katz et al.. 1963); 4) reports o f no more than one dysfunction on 8 items measuring gross mobility and range of motion developed by Rosovv and Breslau (Rosow and Breslau, 1966) (e.g., walk half mile, climb stairs, do heavy housework) and Nagi (Nagi, 1976) (e.g.. push/pull heavy objects, lift 10 lbs., raise arms above shoulder, stoop/crouch/kneel, pick up small objects); 5) the ability to hold a semi-tandem balance for at least 10 seconds; and 6) the abiiity to stand from a seated position 5 times within 20 seconds. The 1189 individuals (90.6%) that met these criteria completed a 90 minute face to face interview covering detailed assessments o f physical performance, cognitive performance, health status, social and psychological characteristics between May 1988 and December 1989. Individuals from the original Durham. North Carolina sub-sample have been omitted from the present study since measures specific to their physical functioning were not collected in 1995; this reduces the analytic sample used here from 1189 to 722 individuals. There are several advantages to using the MacArthur Study. The first strength of the MacArthur Study is in its longitudinal design. The analysis of change in alternative measures o f physical functioning can provide more informative answers about the physical functioning o f older persons than studies o f cross-sectional design. One should be cautions about drawing conclusions from studies based on cross-sectional designs because o f their limited use in making causal inferences. The second strength o f the MacArthur Study is the availability o f repeated self-reported and performance- 96 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. based measures. The availability of this information allows for the examination o f the relationship between self-reported and performance-based measures o f physical functioning, over time. Another advantage to using the MacArthur Study is the availability o f information on the demographic, health, psychological and behavioral characteristics o f these older persons. From this information, it is possible to examine whether these factors account for discrepancies between the self-reported and performance-based measures. Finally, the availability of information on twelve performance-based measures o f physical functioning grants the opportunity to examine whether change in cognitive performance is related to decrement on more physically demanding or novel tasks. C. Measurement o f Constructs The constructs studied here include physical functioning, demographic characteristics, health (conditions and events, cognitive performance and depression) health behaviors, and psychological characteristics. First, the three self-reported and thirteen performance-based measures o f physical functioning are described. Subsequently, indicators o f demographic, health (conditions/events, cognitive performance and depression), health behaviors and psychological characteristics are defined. Physical Functioning. The MacArthur Study included three types o f self-reported measures o f dysfunction (Katz et al., 1963; Rosow and Breslau 1966; and Nagi, 1976) that are included in the present study. The Katz et al.. (1963) scale of activities o f daily living used in the MacArthur Study asked individuals if they were able to walk 9 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. across a small room, bathe, groom, dress, eat. transfer from bed to chair and use the toilet (going to the bathroom as well as cleaning and getting dressed afterwards). Each item is dichotomously scored to indicate the inability/need o f help (“ I”) or ability (“0”) to perform the activity. A continuous summary score is then developed ranging from zero to seven, to reflect the number of ADL items individuals were unable/needed help to perform. In the MacArthur Study, items from the Rosow and Breslau (1966) scale were geared toward measuring gross functional mobility. Specifically, three items asked respondents about their ability to perform heavy house work, walk up and down a flight of stairs, and walk half a mile without help. Each item is dichotomously scored: unable (“ 1”) or able 0*0"). A continuous summary score is then developed, with a range from zero to three, to describe the number o f items individuals self-reported the inability to perform. The Nagi (1976) items are generally seen as measures o f physical activity (Allaire et al., 1999). In the MacArthur Study, individuals were asked to estimate their level of difficulty (none, a little, some, a lot, or unable) pushing/pulling large objects, stooping/crouching/kneeling, carrying ten pounds or more, reaching or extending their arms above their shoulder, and writing/handling small objects. These items are dichotomously scored, as well, to indicate: unable (“ 1”) or able (“0”. responding none, a little, some, a lot). A continuous summary score, ranging from zero to five, to quantifies the number o f Nagi items individuals were unable to perform. The MacArthur Study also included thirteen performance-based measures o f physical functioning. Physical functioning o f the upper body was determined by 98 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. measures of grip strength and fine motor coordination. By employing a dynamometer, maximum hand grip strength (in kilograms) was documented in the dominant arm, based on the average o f the best three measures (Nevitt et al., 1989). Fine motor coordination was assessed by asking respondents to pick up a pencil and sign their name and timing the number o f seconds (up to thirty) required to perform this activity (Jebsen et al.. 1969). Assuming a constant name length, respondents with good coordination were able to complete this task in less time than those with poor coordination. Using data from the MacArthur Study, Seeman et al. (1994) reported a two month test, re-test reliability coefficient o f 0.91 for this signature test indicating good reliability. A large number o f performance-based tests measured lower body functioning (balance, gait, coordination, and strength) in the MacArthur Study. First, Nevitt et al. (1989) detailed four tests o f balance that were used: 1) tandem stand, eyes open (up to 10 seconds); 2) tandem stand, eyes closed (up to 10 seconds); 3) tandem walk (up to 10 steps); and 4) single leg stand (up to 10 seconds). In addition, a summary score for balance was developed by the original investigators o f the M acArthur Study (Seeman et al., 1994). This summary balance measure represents the sum o f the scores (in seconds) on the tandem stand (eyes open) and single leg stand, ranging from zero to twenty seconds. Higher scores for all five o f these measures represented better balance. Previous researchers o f the MacArthur Study reported good reliability o f the summary balance measure (two month test, re-test reliability coefficient of 0.61) (Seeman et al., 1994). Turning balance (360 degrees) was also assessed by asking participants to turn around in a circle (Tinetti, 1986) and timing the number o f seconds required. 9 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The performance-based measures o f gait in the MacArthur Study asked subjects to walk ten feet at a normal pace, turn and return (Nevitt et al.. 1989). They were then asked to complete the same walk at a fast pace. Measures were based on time (in seconds) to complete the course. Participants in the study were instructed to perform this activity without the use o f any aids (e.g., a cane), but were permitted to do so, if necessary. A two month test, re-test reliability coefficient o f 0.80 indicated good reliability for this performance test (Seeman et al., 1994). In order to examine coordination and leg strength, respondents were asked to complete two exercises: foot tapping and repeated chair stand, respectively. When seated, individuals were asked to alternate tapping a foot, between two. two inch circles placed one foot apart, as fast as possible (Nevitt et al., 1989). Up to ten taps were noted and the minimum time (up to thirty seconds) required to complete the exercise was documented. Leg strength was measured by the number of seconds individuals required to perform five chair stands without the use o f their arms (Nevitt et al.. 1989). Respondents were first asked to stand from a seated position without the use o f their arms. If respondents were able to complete this activity, they were asked to repeat it five consecutive times. Using the MacArthur Study. Seeman et al. (1994) reported a two month test, re-test reliability coefficient o f 0.80 for foot tapping and 0.73 for chair stands indicating that these measures have good reliability. A summary measure reflecting the performance on five distinct tasks has also been previously developed by Seeman et al. (1994) for the MacArthur Study. This summary measure reflects the performance on five distinct tasks: balance, gait, foot tapping, repeated chair stands, and picking up a pencil and signing one’s name. 100 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Seeman et al. (1994) first re-scaled each o f these five sub-scales to have a range from zero (worst performance) to one (best performance). These re-scaled items were weighted equally and summed so that the summary score had a range from zero to five. Here, this summary score was multiplied by -1 so that higher scores reflect poorer performance. As will be presented later in this chapter, because participants were initially selected based on their high physical abilities, significant change in most o f these performance-based measures is not seen until 1995. Thus, change in self-reported and performance-based measures is based on data from 1988 and 1995. Individual self- reported items are dichotomously scored: become unable ("1”) or remain able ("0”). Within each domain of self-reported functioning, the dichotomously scored items were summed to construct a continuous summary score to reflect the number o f ADL, gross mobility, and physical activities individuals are unable to perform. For the performance- based measures, change was defined as the difference between the 1995 and 1988 continuous score (in seconds or kilograms). Furthermore, change for the summary performance score was defined as the difference between the 1995 and 1988 summary scores multiplied by -1 so that higher change scores reflect decrement in physical performance. Demographic characteristics. Studies have shown that age, gender, and socioeconomic status influence physical functioning among older persons (Manton, 1988; Guralnik and Kaplan. 1989: Strawbridge et al.. 199"; Crimmins and Saito. 1993; Guralnik et al., 1993; Kaplan et al.. 1993; Boult et al., 1994). For a more detailed discussion o f these studies, the author directs the reader to Chapter II. In the 101 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. present analysis age is coded in years. In addition, dichotomous variables are used to indicate the following characteristics (reference group in parentheses): female (male), having less than 12 years o f education (at least 12 years o f education), having an annual income less than $10,000 (annual income o f at least $10,000). In this study, as the MacArthur Study is a sub-sample of the Established Populations for Epidemiologic Studies (EPESE) study, income and education were dichotomized based on the findings from Guralnik et al. (1993). As discussed in Chapter II. Guralnik et al. (1993) used a longitudinal sample from the EPESE to assess the role o f demographic factors in maintaining mobility in older persons. They found that persons with lower (< $5,000 per year) and middle ($5,000 to $9,999 per year) income levels were associated with significantly lower rates o f maintaining mobility (on reported measures o f walking up and down stairs and walking half a mile) than persons with high (> $10,000 per year) income levels. In addition, they found that persons with low (< 8 years) and medium (9 to 12 years) levels of education were also associated with lower rates o f maintaining mobility compared to those with more than 12 years of education. Health conditions and events. As detailed in Chapter II. studies have shown that specific disease conditions influence decrement in physical functioning (Satarino et al.. 1990; W olf et al.. 1992; Guralnik et al.. 1993; Kurtz et al.. 1993; Ettinger et al.. 1994; Mulrow et al.. 1994; Ormel et al., 1998; Kurtz et al., 1999). Consequently, five prevalent health conditions and 2 health events are examined: whether the respondent had ever had cancer (excluding minor skin cancers), diabetes, or high blood pressure (average systolic and diastolic measurements after three seated readings greater than 140/90 mm Hg or 102 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. taking blood pressure medication), myocardial infarction, or a stroke. The two health events considered are: “having fractured a hip” and “having fractured other bones” as dichotomous independent variables. Cognitive performance. In the MacArthur Study, the total cognitive score represented a summation of the performance on five sub-scales: I ) spatial memory, 2) similarity o f abstract concepts, 3) language. 4) delayed verbal memory, and 5) spatial orientation. Spatial memory was assessed by implementing the delayed Span Test, which requires the placement o f circular disks on a board in a specific order and then asking individuals to identify the new disk that has been added (Moss et al.. 1986). This test identifies individuals with neurological diseases. By scoring individuals from 0 to 17, this easily administered test focused on measuring spatial memory separate from verbal related ability. The Wechsler Adult Intelligence Scale -Revised (WAIS-R) was used to evaluate abstract concepts (Wechsler. 1981). Participants were asked to explain how different objects or concepts (e.g., work and play) were similar and scored from 0 to 16 on this task. The Boston Naming Test tests language (Kaplan et al.. 1983). Individuals were shown 18 drawings (e.g., broom, bench, tree, harp, etc.) and asked to name each object. Only items correctly named were scored within a time limit o f 10 seconds per picture. Relying on the Boston Naming Test (Kaplan et al., 1983), delayed verbal memory was tested. Individuals were asked to recall the images shown during the naming test. Participants were challenged by this task as they were not asked explicitly to remember these images during the naming test. Finally, spatial orientation was tested through measurement o f subject's ability to copy geometric figures including a diamond, a diamond in a square, and a three-dimensional cube (Rosen et al., 1984). 103 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Participants were asked to draw these figures and a single rater scored drawings using a standardized system. A total cognitive score, representing the sum of these five tests, was determined by the original investigators o f the MacArthur Study and ranged from zero to eighty-nine such that lower scores indicate poorer cognitive functioning (Inouye et al., 1993). At baseline, total cognitive score was normally distributed (see Figure III-1) with a mean score of 53.95 (standard deviation 9.54). Continuous scores o f cognitive functioning were used and change in cognitive functioning was defined as the difference between the 1995 and 1988 scores multiplied by -1 so that higher scores reflect declining cognitive functioning. Depression. The depression scale is adapted from the Hopkins Symptom Check List (Derogatis et al., 1974). Participants were assigned a score based on the sum of eleven items such that individuals with higher continuous scores (ranging from eleven to forty-four) have a higher depression score. Due to the skewed distribution of this scale at baseline (see Figure III-2). a dichotomous variable was created based on the highest tertile (> 15) to indicate levels o f depression (‘T ’ = high levels o f depression and "O'* = moderate/low levels). Health behaviors. Three health behaviors were also examined: body composition, cigarette smoking, and alcohol consumption. Body mass index (kg/m2) captured the relative weight of each individual. Based on tertiles, a categorical variable describes individuals with low (< 24.27), medium (24.28 to 27.76). and high (> 27.77) BMI levels. Two dummy variables were used to indicate categories low BMI and high BMI with category medium BMI serving as the referent group. Smoking behavior is 104 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o o o o o o o o o v O r r f N O O C v O T f < N (u) Xousnbsj J 105 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cognitive Score o o o o o o o o ^ < N O oc ^ < N (u) /(ousnbaj j 106 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Depression Score measured using two items that assessed the use and frequency o f cigarette smoking. A categorical variable describes individuals who currently smoke, have smoked in the past, and have never smoked. Two dummy variables were used to indicate categories current smoker and past smoker with category never smoked serving as the referent. Finally, alcohol consumption is measured by four items that assessed the type and quantity of alcohol consumed. Again two dummy variables were used to indicate: < 1 gram ethyl alcohol per month and > 1 gram ethyl alcohol per month with category none in the past year serving as the reference group. Psychological Characteristics. Two psychological resources were measured: self-efficacy and mastery. Self-efficacy and mastery were operationalized by the scale developed by Rodin and McAvay (1993) and Pearlin and Schooler (1978). respectively. Individuals can score between nine and thirty-six on the self-efficacy scale and between seven and twenty-eight on the scale for mastery. Both of these scales were normally distributed at baseline with higher scores indicating higher levels o f these resources. D. Descriptive Statistics o f the Sample First, Table III-1 provides detailed information on each o f the thirteen performance-based measures by year. This table details the mean, standard deviation, minimum and maximum values, and the percent o f participants who were ''unable” to perform the activity. If individuals were not able to perform the tandem stand with their eyes open for ten seconds, they were not asked to perform the more strenuous exercises o f tandem stand, eyes closed and tandem walk. Consequently, to avoid the risk o f injury, 107 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T a b le I l l - l . D e s c rip tiv e c h a ra c te ris tic s o f p e r fo rm a n c e -b a s e d m e a s u re s o f p h y s ic a l fu n c tio n in g fo r a ll 3 y e a rs (1 9 8 8 , 19 91, 19 95)_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ [ M e a n M in im u m M a x im u m P e r c e n t % (n ) (std . d ev .1 ) n V a lu e V a lu e “U n a b le ” 1 9 8 8 B a la n c e 1 4 .7 9 (5 .3 2 ) 71 2 1.30 2 0 .0 0 0 . 6 % (4 ) S in g le L e g S ta n d 6 .9 7 (3 .1 9 ) 7 0 6 1.00 10.00 1.5 % (1 1 ) T a n d e m S ta n d , E y e s O p e n 8 .0 3 (3 .0 6 ) 6 9 9 0 .5 0 10.00 2 . 4 % (1 7 ) T a n d e m S ta n d , E y e s C lo s e d 4 .6 5 (3 .1 2 ) 4 3 6 0 .6 0 10.00 1 .9 % (1 4 ) S te p s in T a n d e m W a lk 7 .0 9 (3 .0 9 ) 4 3 9 1.00 10.00 2 .4 % (1 1 ) G r ip S tre n g th (K g .) 2 9 .2 6 (9 .9 3 ) 7 1 2 1 0 .0 0 6 2 .0 0 0 . 1 % ( I ) C h a ir S ta n d s 1 1 .9 6 (2 .8 5 ) 7 2 2 5 .2 0 2 0 .4 0 0 . 0 % (0 ) H a n d S ig n a tu re 9 .0 9 (2 .5 9 ) 7 1 6 3 .9 0 2 2 .7 0 0 . 4 % (3 ) R ig h t/L e ft F o o t T a p 4 .2 0 ( 1 .1 6 ) 7 1 7 1.90 10.10 0 . 3 % (2 ) T u r n in g a C irc le 2 .8 4 (0 .8 6 ) 7 1 0 0 .9 0 8 .8 0 0 . 8 % (6 ) W a lk /N o rm a l P a c e 9 .6 5 (2 .1 8 ) 711 5 .1 0 2 1 .6 0 0 . 7 % (5 ) W a lk /F a s t P a c e 6 .9 8 ( 1 .5 4 ) 7 0 8 3 .8 0 12.90 0 . 8 % (6 ) S u m m a r y S c o re 2 .8 2 (0 .4 8 ) 7 1 4 3 .8 6 0 .4 9 -- 1 9 9 1 B a la n c e 14 .4 6 (5 .3 2 ) 5 3 2 0 .3 0 2 0 .0 0 9.1 % (5 3 ) S in g le L e g S ta n d 6 .7 1 (3 .2 6 ) 5 1 7 0 .8 0 10.00 1 1 .6 % (6 8 ) T a n d e m S ta n d , E y e s O p e n 7 .9 2 (3 .0 8 ) 5 3 8 0 .3 0 10.00 9 . 0 % (5 3 ) T a n d e m S ta n d , E y e s C lo s e d 3 .8 9 (2 .8 3 ) 4 6 7 0 .4 0 10.00 1 8 .6 % (1 0 7 ) S te p s in T a n d e m W a lk 6 .4 7 ( 3 .3 5 ) 4 6 9 1.00 10.00 1 8 .2 % (1 0 4 ) G rip S tre n g th (K g .) 2 8 .4 1 (9 .6 0 ) 5 9 4 8 .0 0 5 8 .0 0 0 .0 % (0 ) C h a ir S ta n d s 10.61 (2 .8 3 ) 5 5 2 4 .1 0 3 1 .9 0 5 . 6 % (3 3 ) H a n d S ig n a tu re 9 .0 9 (2 .7 4 ) 5 9 7 2 .3 0 2 3 .2 0 0 .0 % (0 ) R ig h t/L e ft F o o t T a p 4 .2 2 ( 1 .4 1 ) 5 9 0 2 .1 0 18 .00 0 . 8 % (5 ) T u r n in g a C irc le 3 .0 9 ( 1 .0 1 ) 5 7 9 0 .8 0 9 .7 0 1 .7 % (1 0 ) W a lk /N o rm a l P a c e 10.21 (2 .9 7 ) 581 5 .3 0 4 1 .0 0 1 .7 % (1 0 ) W a lk /F a s t P a c e 7 .3 5 ( 1 .8 8 ) 5 7 5 3 .7 0 19.40 2 . 0 % (1 2 ) S u m m a r y S c o re 2 .7 9 (0 .5 9 ) 56 3 0.31 4 .0 0 — 1 9 9 5 B a la n c e 1 1 .9 4 (5 .7 7 ) 3 6 9 0 .4 0 2 0 .0 0 1 6 .5 % (7 3 ) S in g le L e g S ta n d 5 .3 4 (3 .3 3 ) 361 0 .3 0 10.00 1 8 .5 % (8 2 ) T a n d e m S ta n d . E y e s O p e n 6 .7 3 (3 .5 2 ) 3 7 4 0 .4 0 10.00 1 6 .3 % (7 3 ) T a n d e m S ta n d , E y e s C lo s e d 3 .5 3 (2 .9 9 ) 3 3 2 0 .4 0 10.00 2 4 .4 % (1 0 7 ) S te p s in T a n d e m W a lk 5 .1 2 ( 3 .3 2 ) 3 4 4 1.00 11.00 2 2 .7 % (1 0 1 ) G r ip S tre n g th (K g .) 2 3 .2 9 (9 .8 9 ) 451 1.00 5 4 .0 0 0 .2 % (1 ) C h a ir S ta n d s 1 1 .2 8 ( 3 .4 4 ) 3 8 5 3 .4 0 3 3 .0 0 1 5 .0 % (6 8 ) H a n d S ig n a tu re 1 0 .5 6 (4 .3 5 ) 4 4 9 3 .4 0 3 0 .0 0 0 .9 % (4 ) R ig h t/L e ft F o o t T a p 5 .3 4 (2 .7 4 ) 4 4 2 2 .3 0 2 3 .9 0 3.1 % (1 4 ) T u r n in g a C irc le 3 .8 2 (2 .1 8 ) 4 2 3 1.40 2 7 .4 0 6 . 2 % (2 8 ) W a lk /N o rm a l P a c e 1 2 .5 0 (7 .5 1 ) 4 3 0 5 .9 0 9 0 .0 0 4 . 2 % ( 1 9 ) W a lk /F a s t P a c e 8 .6 2 (3 .1 9 ) 4 1 5 3 .6 0 3 2 .5 0 5 . 5 % (2 4 ) S u m m a r y S c o re 2 .8 5 (0 .3 6 ) 2 5 4 3 .8 2 2 .2 9 — Notes: U n le ss s p e c ifie d in ta b le , u n its fo r a ll p e rfo rm a n c e -b a s e d m e a s u re s a re in s e c o n d s . 1 S ta n d a rd d e v ia tio n . 108 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. there are fewer individuals with scores on these two performance exercises. The descriptive characteristics o f self-reported and performance-based (among those who were “able" to perform) measures o f physical functioning for those with data across all three sampling years are presented in Tables III-2 and I1I-3. Chi-square and t-test statistics were used to determine whether there were any differences among reported and performed measures from 1988 to 1991 and 1988 to 1995. Although the majority o f respondents continued to self-report the ability to perform all items on the three scales o f functioning in 1995 (81.8% on the Katz scale; 55.1% on the Rosow and Breslau scale; and 72.9% on the Nagi scale), significant decrement is observed in self-reported measures o f physical functioning, see Table III-2. For all the individual and summary measures of reported functioning, a greater number o f respondents reported decrement in 1995 as compared to earlier years, most commonly 1988. For most of these self-reported measures, significant decrement is seen as early as 1991. Upon examining the thirteen performance-based measures o f physical functioning, two broad sub-categories are observed: 1) measures for which decrement is based on a reduction in average value (e.g., balance, single leg stand, tandem stand/eyes open, tandem stand/eyes closed, tandem walk, and grip strength); and 2) measures for which decrement is based on an increase in average value (e.g., chair stands, hand signature, right/left foot tap, turning in a circle, walking at a fast pace and walking at a normal pace), see Table 1II-3. Individuals had, on average, significantly poorer upper Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T a b le H l-2 . D e s c rip tiv e c h a ra c te ris tic s o f s e lf-r e p o r te d m e a s u re s o f p h y s ic a l fu n c tio n in g 1 9 8 8 P e r c e n t ( % ) n 1 9 9 1 P e r c e n t ( % ) n P - v a lu e 1 1 9 9 5 P e r c e n t ( % ) n P - - v a lu e K a tz I te m s U n a b le /N e e d H elp : W a lk in g (n = 5 3 3 ) 0 .0 0 2 .4 13 0 .0 0 1 11.6 6 2 0 .0 0 1 B a th in g (n = 5 3 3 ) 0 .0 0 2.3 12 0 .0 0 1 13.1 7 0 0 .0 0 1 G ro o m in g (n = 5 3 2 ) 0 .0 0 0 .9 5 0 .0 2 5 3 .9 21 0 .0 0 1 D re s s in g (n = 5 3 3 ) 0 .0 0 1.3 7 0 .0 0 8 6 .8 3 6 0 .0 0 1 E a tin g (n = 5 3 3 ) 0 .0 0 0 .8 4 0 .0 4 5 2 .4 13 0 .0 0 1 T r a n s f e r rin g (n = 5 3 3 ) 0 .0 0 1.3 7 0 .0 0 8 5.1 2 7 0 . 0 0 1 T o ile tin g (n = 5 3 3 ) 0 .0 0 1.3 7 0 .0 0 8 6 .8 3 6 0 .0 0 1 S u m o f Ite m s (n = 5 3 2 ) 0 1 0 0 .0 5 3 2 96 .1 511 0 .0 0 0 8 1 .8 4 3 5 0 .0 0 0 1 0 .0 0 1.9 10 7.7 41 2 0 .0 0 0 .4 2 3.6 19 3 0 .0 0 0 .9 5 1.9 10 4 0 .0 0 0 .0 0 0 .9 5 5 0 .0 0 0 .2 1 1.3 7 6 0 .0 0 0 .2 1 0 .9 5 7 0 .0 0 0 .4 2 1.9 10 R o so w -B r e sia u I te m s U n a b le to: W o rk A ro u n d H o u se 5 .0 2 6 16.3 85 0 .0 0 1 3 8 .0 199 0 .0 0 1 (n = 5 2 3 ) W a lk U p /D o w n S ta irs (n = 5 2 8 ) 0 .2 1 4 .5 24 0 .0 0 1 12.5 6 6 0 .0 0 1 W a lk H a l f M ile 3.3 17 11.0 5 6 0 .0 0 1 2 9 .7 151 0 .0 0 1 (n = 5 0 9 ) S u m o f Ite m s (n = 5 3 2 ) 0 9 1 .4 4 8 6 7 8 .6 4 1 8 0 .0 0 0 55.1 2 9 3 0 .0 0 0 1 8.5 4 5 14.5 77 2 0 .3 108 2 0 .2 1 3 .6 19 13.3 71 3 0 .0 0 3 .4 18 11.3 6 0 N a g i Ite m s U n a b le to: P u s h /P u ll L a rg e O b je c ts (n = 5 2 0 ) 2 .0 11 7.5 3 9 0 .0 0 1 16.2 8 4 0 .0 0 1 S to o p /K n e e l (n = 5 2 9 ) 0 .9 5 3 .6 19 0 .0 0 4 11.0 58 0 .0 0 1 L ift/C a rry 10 lbs. 1.0 5 9.1 4 7 0 .0 0 1 19.5 100 0 .0 0 1 ( n = 5 14) R e a c h /E x te n d 0 .0 0 1.5 8 0 .0 0 5 5.1 2 7 0 .0 0 1 (n = 5 3 2 ) W r ite /H a n d le S m all 0 .0 0 0 .5 3 0 .0 8 3 2.3 12 0 .0 0 1 O b je c ts (n = 5 3 2 ) S u m o f Ite m s (n = 5 3 1 ) 0 9 6 .0 5 1 0 8 6 .4 4 5 9 0 .0 0 0 7 2 .9 3 8 7 0 .0 0 0 1 4 .0 21 7 .9 4 2 12.1 6 4 2 0 .0 0 3 .0 16 7 .7 41 3 0 .0 0 2.3 12 4.3 23 4 0 .0 0 0 .2 1 2 .3 12 5 0 .0 0 0 .2 1 0 .8 4 Notes: p - v a lu e fo r d iff e re n c e b e tw e e n 1 9 8 8 a n d 1991 m e a s u re s. : p - v a lu e fo r d iff e re n c e b e tw e e n 1991 a n d 1995 m e a s u re s. 110 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and lower body function in 1995 as compared to 1988. Scores in 1988 and 1991 more closely approximate each other for some o f these healthy, high functioning individuals. T a b le III-3 . D e s c rip tiv e c h a ra c te ris tic s o f p e rfo rm a n c e -b a s e d m e a s u re s o f p h y s ic a l fu n c tio n in g 1 9 8 8 1 9 9 1 1 9 9 5 I V I e a n M e a n p - , M e a n p * (S td . D e v .)1 (S td . D e v .) v a lu e * (S td . D e v .) v a lu e B a la n c e (n = 3 4 0 ) 1 5 .4 4 (5 .1 1 ) 1 5 .3 5 (4 .9 1 ) 0 .8 3 2 1 2 .1 8 (5 .7 1 ) 0 .0 0 0 S in g le L e g S ta n d (n = 3 2 8 ) 7 .6 9 (2 .9 5 ) 7.11 (3 .1 8 ) 0 .0 1 6 5 .4 9 (3 .3 3 ) 0 .0 0 0 T a n d e m S ta n d , 0 .0 0 0 E y e s O p e n (n = 3 4 3 ) 8 .0 2 (3 .0 3 ) 8 .4 0 (2 .7 8 ) 0 .0 9 0 6 .8 7 (3 .4 9 ) T a n d e m S ta n d , 0 .0 0 0 E y e s C lo s e d (n = 1 9 4 ) 5 .3 6 (4 .4 8 ) 4 .4 8 (3 .0 5 ) 0 .0 0 7 3 .9 8 ( 3 .1 5 ) S te p s in T a n d e m W a lk 8 .1 2 ( 2 .9 3 ) 7 .0 7 (3 .2 1 ) 0 .0 0 1 5 .6 3 (3 .3 3 ) 0 .0 0 0 (n = 2 0 2 ) G rip S tre n g th (K g .) (n = 4 3 0 ) 2 8 .4 3 (9 .9 8 ) 0 .6 3 0 2 3 .4 0 ( 9 .8 3 ) 0 .0 0 0 2 8 .7 7 (1 0 .2 5 ) C h a ir S ta n d s (n = 3 6 4 ) 11.92 (2 .8 6 ) 1 0 .52 (2 .5 7 ) 0 .0 0 0 11.23 (3 .2 8 ) 0 .0 0 3 H a n d S ig n a tu r e ( n = 4 3 1) 8 .8 4 (2 .2 7 ) 8 .8 2 (2 .4 5 ) 0 .8 9 3 10 .4 4 (4 .1 4 ) 0 .0 0 0 R ig h t./L e f t F o o t T a p (n = 4 2 3 ) 4 .2 0 ( 1 .1 8 ) 4 .1 7 ( 1 .4 3 ) 0 .7 1 9 5.31 (2 .6 6 ) 0 .0 0 0 T u r n in g a C irc le (n = 4 0 2 ) 2 .8 0 (0 .8 9 ) 3 .0 2 (0 .9 3 ) 0 .0 0 1 3.81 (2 .2 1 ) 0 .0 0 0 W a lk /N o r m a l P a c e (n = 4 1 0 ) 9 .4 9 ( 1 .9 6 ) 9 .8 9 (2 .3 2 ) 0 .0 0 1 1 2 .4 2 (7 .5 4 ) 0 .0 0 0 W a lk 'F a s t P a c e (n -3 9 3 ) 6 .8 2 ( 1 .4 9 ) 7.11 (1 .6 8 ) 0 .0 1 0 8 .5 2 ( 3 .1 3 ) 0 .0 0 0 S u m m a r y S c o re (n = 2 4 4 ) 3 .0 4 (0 .3 3 ) 3 .0 9 (0 .4 0 ) 0 .1 7 2 3 2 .8 5 (0 .3 6 ) 0 .0 0 0 Notes: U n le s s s p e c ifie d in ta b le , u n its fo r all p e rfo rm a n c e -b a s e d m e a s u re s a re in se c o n d s. ' S ta n d a rd d e v ia tio n . 2p - v a lu e fo r d iff e re n c e b e tw e e n 1988 a n d 1991 m e a s u re s. 3p - v a lu e fo r d iff e re n c e b e tw e e n 1991 a n d 1995 m e a s u re s. Table III-4 provides descriptive statistics for change (1988 to 1995) in these self- reported and performance-based measures. The greatest change in the summary reported measures (i.e., reported the ability to perform activity in 1988 and subsequently reported decrement in at least one o f the items on each scale) is observed along the Rosow-Breslau items (41.3%) followed by the Nagi (26.2%) and the Katz (18.2%) items. Respondents most frequently reported "new” disabilities working around the house (36.0%) and walking half a mile (29.1%). Over ten percent o f the sample also reported becoming unable to lift/carry ten pounds (18.9%), push/pull large objects (15.4%), bathe (13.0%), III Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T a b le III-4 . D e s c rip tiv e c h a ra c te ris tic s o f c h a n g e (1 9 8 8 to 1 9 9 5 ) in s e lf -r e p o r te d a n d p e r fo rm a n c e - b a s e d m e a s u re s o f p h y s ic a l fu n c tio n in g (n = 7 2 2 )_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ S e lf-R e p o r te d M e a su r e s P e r c e n t (% ) N u m b e r (n ) K a tz S te m s B e c o m e U n a b le /N e e d H elp : W a lk in g (% y e s ) 11 .6 6 2 B a th in g (% y e s ) 13.1 70 G ro o m in g (% y e s ) 3 .9 21 D re s s in g (% y e s ) 6 .8 36 E a tin g ( % y e s ) 2 .4 13 T ra n s f e r rin g B e d to C h a ir (% y e s ) 5.1 2 7 T o ile tin g {% y e s ) 6 .8 36 S u m o f Ite m s 0 8 1 .8 4 3 5 1 7.7 41 2 10.5 56 R o so w -B r e sla u Ite m s B e c o m e U n a b le to : W o rk A ro u n d H o u s e (% y e s ) 3 6 .0 183 W a lk U p /D o w n S ta irs (% y e s ) 12.9 6 9 W a lk H a lf M ile W ith o u t H e lp (% y e s ) 29 .1 149 S u m o f Ite m s 0 5 8 .7 2 9 0 1 19.0 94 2 2 2 .3 110 N a g i Ite m s B e c o m e U n a b le to: P u s h /P u ll L a rg e O b je c ts (% y e s ) 15.4 80 S to o p /K n e e ! (% y e s ) 10.9 58 L ift/C a rry 10 lbs. (% y es) 18.9 9 8 R e a c h /E x te n d (% y e s ) 5 .0 2 7 W rite /H a n d le S m a ll O b je c ts ( % y e s) 2 .2 12 S u m o f Ite m s 0 7 3 .8 381 1 12.2 63 2 14.0 72 P e r fo r m a n c e -B a se d M e a su r e s M e a n (S td . D ev .)1 N u m b e r (n ) B a la n c e -3 .4 1 ( 5 .7 5 ) 3 6 9 S in g le L e g S ta n d - 2 .1 8 ( 3 .6 4 ) 3 5 9 T a n d e m S ta n d , E y e s O p e n - 1 .2 7 ( 3 .9 9 ) 3 6 9 T a n d e m S ta n d , E y e s C lo s e d -1 .4 7 (4 .0 0 ) 2 1 4 S te p s in T a n d e m W a lk -2 .4 8 ( 3 .7 2 ) 2 2 3 G r ip S tre n g th (K g .) -5 .5 1 (6 .2 1 ) 4 4 8 C h a ir S ta n d -0 .6 3 (3 .4 7 ) 38 5 H a n d S ig n a tu r e 1 .7 0 ( 3 .6 4 ) 4 4 8 R ig h t/L e ft F o o t T a p 1 .1 5 ( 2 .5 3 ) 441 T u r n in g a C irc le 1 .0 0 (2 .1 3 ) 4 2 0 W a lk a t N o rm a l P a c e 2 .9 7 ( 7 .1 2 ) 4 2 8 W a lk a t F a s t P a c e 1 .7 8 ( 2 .7 8 ) 4 1 2 S u m m a r y S c o re 0 .5 5 ( 0 .7 0 ) 4 2 0 iV oto. U n le s s s p e c ifie d in ta b le , u n its fo r a il p e rfo rm a n c e -b a s e d m e a s u re s a re in s e c o n d s . 1 S ta n d a rd d e v ia tio n . 11 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. walk up/down a flight o f stairs (12.9%), walk across a room (11.7%) and stoop/kneel (10.9%). Average change scores for performance-based measures are also presented in Table III-4. On average, individuals declined on all performance-based measures, except repeated chair stands. As mentioned earlier, change in these measures is defined by either a reduction in average value or an increase in average value. Based on the reduction in average value, grip strength (5.51 Kg) and balance (3.41 seconds) showed the greatest decrement. Decrement based on an increase in average value was most strongly seen for walking at a normal pace (2.97 seconds) and walking at a fast pace (1.78 seconds). In 1988, respondents were, on average, seventy-four years o f age. predominantly female (52.8%). with high education (52.0% 12 years or more) and annual income levels (56.5% > $10,000). see Table III-5. In general, these community residents over seventy reported their health positively. Although 41.2% reported having had high blood pressure, less than 25% reported having had other health conditions or events: fractured a non-hip bone (23.0%). cancer (20.4%), diabetes (14.2%). myocardial infarction (13.1%). stroke (2.7%), and having fractured a hip (2.1%). These persons also had relatively good cognitive functioning with an average score o f 54.0 (range 0 to 89) and approximately 36.8% had depression scores o f 15 or higher. They also seem to practice positive health behaviors (43.9% have never smoked and 61.1% have never had any alcohol in the past year or drink moderately). Furthermore, moderately-high average Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. levels o f mastery (19.0, range 7 to 28) and self efficacy (26.3, range 9 to 36) were corded for these older Americans. Table III-5 compares the analyzed sample o f 722 participants to the excluded cohort o f 467 persons from Durham, North Carolina. Compared to the excluded sub sample. persons selected for these analyses have a significantly smaller proportion of women (52.8% versus 59.5%) and are more highly educated (52% have 12 or more years o f education compared to 37.7%). These two samples are similar on the remaining two demographic characteristics o f age and income level. Although the cohort from Durham seems healthier than the analyzed sample, they have a lower average score o f cognitive performance than the sample analyzed (50.9 versus 54.0). Specific to health conditions and events, participants excluded from these analyses less frequently reported having had cancer (14.4%) and myocardial infarction (8.9%), compared to the analyzed sample (20.4% and 13.1%. respectively). For all other remaining health characteristics, there were no significant differences between these samples. The sub-sample from Durham. North Carolina also practiced more positive health behaviors than the analyzed sample. For instance, o f the 467 individuals excluded from these analyses. 53.8% have never smoked (compared to 43.9% o f those included in the analyses) and 80.5% have not had any alcohol in the past year or drank moderately (compared to 61.1%). As cut-points for BMI were based on tertiles of the analyzed sample, it is not appropriate to compare the groups based on this categorization of BMI. Finally there were no significant differences between average levels o f psychological attributes (i.e., mastery and self-efficacy) between these two samples. 114 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T a b le IH -5. D e s c rip tiv e c h a ra c te ris tic s o f th o s e in c lu d e d a n d e x c lu d e d in th e stu d y V a r ia b le I n c lu d e d P e r c e n t < % ) n E x c lu d e d P e r c e n t ( % ) n p -v a lu e D e m o g r a p h ic C h a r a c te r istic s A g e (m e a n , s ta n d a rd d e v ia tio n ) 7 4 .2 (2 .6 ) 72 2 7 4 .4 ( 3 .0 ) 4 6 7 0 .2 0 0 G e n d e r F e m a le 5 2 .8 % 381 5 9 .5 % 2 7 8 0 .0 2 2 E d u c a tio n < 12 y e a rs 4 8 .0 % 3 4 6 6 2 .3 % 291 0 . 0 0 1 In c o m e < $ 1 0 ,0 0 0 4 3 .5 % 2 7 6 47 .1 % 2 1 0 0 .2 4 8 H e a lth C h a r a c te r istic s H e a lth C o n d itio n s /E v e n ts C a n c e r 2 0 .4 % 146 14.4 % 67 0 .0 0 8 D ia b e te s 1 4 .2 % 102 1 1 .4 % 53 0 .1 5 1 H ig h B lo o d P re ssu re 4 1 .2 % 2 9 5 46 .1 % 2 1 2 0 .0 9 9 M y o c a rd ia l In fa rc tio n 13.1 % 94 8 .9 % 4 2 0 .0 3 0 S tro k e 2 .7 % 19 2.1 % 10 0 .5 7 8 H ip F ra c tu re 2.1 % 15 1.3 % 6 0 .3 0 2 O th e r B o n e F ra c tu re s 2 3 .0 % 165 2 4 .8 % 116 0 .4 7 0 C o g n itiv e F u n c tio n in g 5 4 .0 (9 .5 ) 721 5 0 .9 ( 1 0 .8 ) 4 6 6 0 .0 0 0 (m e a n , s ta n d a rd d e v ia tio n ) D e p re s s io n (> 15) 3 6 .8 % 2 5 8 3 1 .5 % 152 0 .1 9 9 H e a lth B e h a v io r s B M I L o w (< 2 4 .2 7 ) 3 2 .7 % 2 3 4 4 1 .4 % 188 0 .0 0 3 M e d iu m (2 4 .2 8 - 2 4 .7 6 ) 3 3 .4 % 2 3 9 3 2 .6 % 148 H ig h (> 2 7 .7 7 ) 3 3 .9 % 2 4 2 2 6 .0 % 118 S m o k in g S ta tu s N e v e r 4 3 .9 % 3 1 7 5 3 .8 % 251 0 .0 0 2 C u rre n t 1 5 .8 % 114 1 5 .2 % 71 P a st 4 0 .3 % 291 31.1 % 145 A lc o h o l C o n s u m p tio n N o n e in P a s t Y e a r 3 7 .9 % 271 6 3 .2 % 2 9 5 0 .0 0 1 < 1 g ra m e th y l a lc o h o l/m o n th 2 3 .2 % 166 1 7 .3 % 81 > 1 g ra m e th y l a lc o h o l/m o n th 3 9 .0 % 2 7 9 1 9 .5 % 91 P sy c h o lo g ic a l A ttr ib u te s M a ste ry 19 .0 (2 .5 ) 7 1 5 18.9 (2 .2 ) 4 6 4 0 .9 1 7 (m e a n , s ta n d a rd d e v ia tio n ) S e lf-E ffic a c y 2 6 .3 (2 .5 ) 7 1 6 2 6 .3 (2 .3 ) 4 6 3 0 .9 4 0 (m e a n , s ta n d a rd d e v ia tio n ) E. Summary This chapter described the individual measures selected from MacArthur Research o f Network on Successful Aging Community Study. First, the data and sample 115 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. used was described. Subsequently, detailed information on the operationalization o f physical functioning (self-reported and performance-based), demographic, health (conditions/events, cognitive performance, and depression), health behaviors, and psychological characteristics were presented. In general, among this relatively healthy community based sample of participants, significant decrement was found for both self- reported and performance-based measures o f physical functioning from 1988 to 1995. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter IV Discrepancies Between Self-Reported and Performance-Based Measures of Physical Functioning A. Introduction As detailed in Chapter II, studies examining the relationship between self-report and performance-based measures have found a moderate association (Jette and Branch, 1985; Kelly-Hayes et al„ 1992; Sager et al„ 1992; Myers et al„ 1993; Cress et al„ 1995; Reuben et al„ 1995; Kempen. Van Heuvelen et al„ 1996; Kempen, Steverink et al„ 1996; Mendes de Leon et al„ 1996; Merrill et al„ 1997; and Ferrer et al„ 1999). Research on their association over time, however, is limited. Only three studies have shown that their longitudinal association is also moderate (Myers et al„ 1993; Mendes de Leon et al„ 1996; Kempen, Sullivan et al„ 1999). O f the three studies with longitudinal components (Myers et al„ 1993; Mendes de Leon et al., 1996; Kempen. Sullivan et al„ 1999), only Kempen, Sullivan et al. (1999) investigated how the observed discordance between reported and performed measures is influenced by other factors. The purpose o f this chapter is to further our understanding o f the longitudinal association between self- reported and performance-based measures o f physical functioning and identify sources of discrepancies. As the MacArthur Study was not originally designed to examine the concordance between self-reported and performance-based measures, comparative self-reported and performance-based measures are not one o f the same. Thus, in order to select individual measures o f physical functioning for comparison, descriptive statistics are presented for seven a priori matched pairs o f self-reported and performance-based measures. 117 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The accuracy o f the change in reported functional problems is then estimated by calculating sensitivity and specificity for the comparisons selected. Finally, multivariate analyses test the effect of demographic, health, behavioral, and psychological characteristics on the observed discrepancies between change in self-reported and performance-based measures. As reviewed in Chapter II. the studies examining the potential sources of discrepancies between self-reported and performance-based measures o f physical functioning leave many questions unanswered. Although they suggest that demographic, health, and psychological attributes account for the discrepancies between reported and performed measures o f physical functioning, information is needed on how these characteristics influence discrepancies between change in reported and performed measures o f physical functioning. Although this information has recently been provided for depressive symptomatology (Kempen. Sullivan, et al.. 1999), the purpose o f this investigation is to examine how demographic, health, behavioral, and psychological characteristics influence the observed discrepancies between change in reported and performed measures o f physical functioning. While it is expected that persons with the certain characteristics will over-report the onset o f dysfunction (e.g., individuals who are older, female, with lower economic status, specific disease conditions or fractures, higher levels o f depression, and poorer health behaviors) it is also expected that persons with higher levels o f mastery, self-efficacy, and cognitive functioning will under-report the onset o f dysfunction, over time. For example, it is expected that compared to individuals with moderate levels o f BMI, persons with high BMI will report greater onset 118 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dysfunction than what is revealed from objective performance-based measures of physical functioning. B. Method Sample. Before analyzing the accuracy of self-reported items, conceptually similar self-reported and performance-based measures o f physical functioning were matched, on an a priori basis. As the participants of the MacArthur Study were selected on their high physical and cognitive abilities at baseline, it was important to identify comparisons that had variability in reported measures o f physical functioning for analysis. Four performance-based measures from the MacArthur Study were matched with four self-reported measures based on the face validity o f similar functioning. A total o f seven comparisons were made. The first comparison examined the reported ability to walk across a room (Katz et al.. 1963) with the timed (in seconds) performance measure o f walking a ten foot course at a normal pace (Nevitt et al.. 1989). Both o f these measures captured the ability to walk a relatively short distance at a normal pace. The second comparison examined the reported ability to "toilet" with the timed (in seconds) repeated chair stands exercise - without the use o f arms (Nevitte et al.. 1989). Although it is arguable whether individuals defined the ability to "toilet” as the ability to sit on the toilet without the use o f their hands, both o f these measures were paired together because they can involve similar mechanics o f lower body strength. Another comparison that seemed to involve mechanics o f lower body strength paired the reported ability to stoop, crouch, or kneel (Nagi, 1976) with the timed repeated chair stands. Again, although it is unclear 119 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. whether individuals defined their ability to stoop, crouch or kneel with (or without) the use o f their arms, this comparison seemed to capture similar physical mechanics. The ability to "toilet" could also involve fine motor strength or fine motor dexterity. Consequently, the reported ability to toilet was compared with maximum hand grip strength (in kilograms) measured by employing a dynamometer (Nevitt et al.. 1989) and the time (in seconds) required to sign one's name (Jebsen et al., 1969). Two additional comparisons were made o f fine motor activities: 1) the reported ability to write/handle small objects (Nagi. 1976) with the performed grip strength: and 2) the reported ability to write/handle small objects with the time required to sign one's name. These seven pairs of matched measurements are outlined in Figure IV -1. For each comparison, individuals without complete information on the reported and performed measures over the seven years o f the MacArthur Study were omitted from the present study. This reduced the analytic sample used here from 722 individuals (recall that the original sample o f 1189 individuals was reduced to 722 because participants from the Durham, North Carolina sub-sample were omitted because their measures o f physical functioning were not collected in 1995) to less than 450 for each comparison (see Figure IV -1 for specific sample sizes). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1V-1. .•) priori comparisons o f self-reported and performance-based measures C o m p a r iso n S e lf-R e p o r te d M e a s u r e P e r fo r m a n c e -B a se d M e a su r e 1 I N #1 Gait Function (Figure IV-2) Ability to walk across a room (Katz et al.. 1963) Walk '0 foot course Normal pace (Nevitt et al., 1989) 409 #2 Lower Body Strength (Figure IV-3) Ability to •‘toilet” (K atzetal., 1963) Repeated chair stands (Nevitt et al., 1989) 369 43 Lower Body Strength (Figure IV-4) Ability to stoop/crouch/kneel (Nagi, 1976) Repeated chair stands (Nevitt et al., 1989) 366 s4 Fine Motor Strength o f Gross Function (Figure IV-5) Ability to "toilet” (K atzetal., 1963) Maximum grip strength (kg) (Nevitt et al., 1989) 431 45 Fine M otor Dexterity o f Gross Function (Figure IV-6) Ability to "toilet” (Katz et al.. 1963) Hand Signature (Jebsen et al.. 1969) 431 46 Fine M otor Strength (Figure IV-7) Ability to write/handle small objects (Nagi, 1976) Maximum grip strength (kg) (Nevitt et al., 1989) 430 47 Fine M otor Dexterity (Figure IV-8) . . . - _ ,_______, Ability to write/handle small objects (Nagi, 1976) Hand Signature (Jebsen et al., 1969) 431 >p--------------------------------------------------------------------------- :------M— ------------s -----------------------------------------------------------------------------------------------------1 All perform ance-based measures were based on the number o f seconds individuals took to complete the task. For each o f the seven comparisons, a diagram o f boxes and arrows is constructed to best outline the descriptive statistics o f these measures, over time, see Figures IV-2. IV-3. IV-4. IV-5. IV-6. IV-7, and IV-8. By examining the arrows in these figures, we can follow the percentage of individuals who reported their ability (or inability) to perform a given task over time. These arrows direct the reader to boxes within which the mean, standard deviation, and range (in seconds or kilograms) o f the matched performance-based measure is found. In addition, the MacArthur Study distinctly identified individuals who were ‘"unable'’ to perform an exercise. Consequently, the number o f people who where unable to perform is also shown within the boxes. 121 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Figure IV-2. Gait Function: Self-reported ability to walk across a room and perform ance on the tim ed walking o f a 10 foot course, M acArthur Study (1988, 1991, and 1995). 9 2 .4 % A b le (n=378) 9 8 .8 % A b le (n=404) 6 .3 % (J n a b le /N e c d H e lf (n=26) 1 0 0 % A b le (n=409) 0 .9 % A b le (n=4) 1 .2 % U n a b le /N e e d H e lp (n=5) 0 .2 % U n a b le /N e e d H e lf (n=l) mean (s.d.) = 11.44 (5.14) range = 5.90 - 90.00 no. unable to perform = 4 mean (s.d.) = 11.62 (2.64) range = 9 .1 0 - 14.10 mean (s.d.) = 22.40 (--) range = not applicable mean (s.d.) - 26.51(17.42) range = 7.50 - 90.00 no. unable to perform = 11 mean (s.d .)1 = 9.49 ( 1.96) ran^e = 5 .7 0 -1 7 .3 0 no .' unable to perform = 2 mean (s.d.) = 9.84 (2.26) range = 5.30 - 23.70 no. unable to perform = 4 mean (s.d.) = 13.68(3.85) range = 10.00 - 19.50 1 9 8 8 1 9 9 1 1 9 9 5 Note: Percentage values indicate the percent o f reported ability from baseline. A rrow s indicate how individuals self-reported their ability to perform a given task over time (able or unable/need help). The arrows point to boxes that describe characteristics o f the perform ance-based measure: mean, 'standard deviation, range (in seconds), and 'num ber o f people unable to perform. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Figure IV-3. Low er Body Strength: Self-reported ability to use the toilet and perform ance on the lim ed 5 repeated chair stands, M acArthur Study (1988, 1991, and 1995). mean (s.d.) = 11.23(3.36) range = 3.40 - 29.00 no. unable to perform = 76 9 8 .4 % A b le (n -3 6 3 ) mean (s.d.) = 10.52 (2.55) range = 4.10 - 20.60 no.’ unable to perform = 4 1 9 9 .7 % A b le (n=368) mean (s.d.) = 18.02(11.91) range = 8.60 - 38.30 no. unable to perform = 24 1 .3 % U n a b le /N e e d H c lf (n = 5 ) m ean (s.d .)1 = 11.93 (2.86 range = 5 .2 0 -2 0 .2 0 1 0 0 % A b le (n=369) mean (s.d.) = 11.70 (— ) range = not applicable 0 .3 % A b le (n-l) m ean (s.d.) = 11.50 (— ) range = not applicable 0 .3 % U n a b le /N e e d H e lp (n= l) no. unable to perform = 5 1 9 9 5 1 9 9 1 1 9 8 8 Note: Percentage values indicate the percent o f reported ability from baseline. Arrows indicate how individuals self-reported their ability to perform a given task over tim e (able or unable/need help). The arrows point to boxes that describe characteristics o f the perform ance-based measure: mean, 'standard deviation, range (in seconds), and ’num ber o f people unable to perform. IO 0 4 Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Figure IV-4. Lower Body Strength: Self-reported ability to stoop, crouch, or kneel and perform ance on the tinted 5 repeated chair stands, M acArthur Study (1988, 1991, and 1995). 9 3 .4 % A b le (n=342) 9 6 .4 % A b le (n=353) 3 .0 % U n a b le (n=l1) 9 8 .6 % A b le (n = 3 6 l) 0 .8 % A b le (n=3) _ 2 .2 % U n a b le (n=8) 1 .4 % U n a b le mean (s.d.) = 11.87 (2.83) range = 5.20 - 20.20 mean (s.d.) = 12. 77 (3.04) range = 10.40 - 16.20 no. unable to perform = 26 mean (s.d.) = 10.44(2.56) range = 4 .1 0 - 20.60 no.* unable to perform = 31 mean (s.d.) = 11.18(3.52) range = 3.40 - 38.30 no. unable to perform = 71 mean (s.d.) = 12.77 (3.04) range = 10.40 - 16.20 no. unable to perform = 1 mean (s.d.) = 12.34 (3.70) range = 9.20 - 18.30 no. unable to perform = 7 mean (s.d.) = 13.33 (2.26) range = 9.20 - 15.60 no. unable to perform = 8 1 .4 % U n a b le (n=5) 1 .4 % A b le (n=5) 1 .4 % A b le (n=5) mean (s.d .)= 13.36(3.48) range = 7.90 - 16.60 mean (s.d.)= 9.00 (2.93) range = 4.00 - 11.20 mean (s.d.) = 11.32 (1.89) range = 8.20 - 13.10 1 9 8 8 1 9 9 1 1 9 9 5 Note: Percentage values indicate the percent o f reported ability from baseline. Arrow s indicate how individuals self-reported their ability to perform a given task over tim e (able or unable/need help). The arrows point to boxes that describe characteristics o f the perform ance-based measure: mean, 'standard deviation, range (in seconds), and 'num ber o f people unable to perform. to -s- Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Figure IV-5. Fine M otor Dexterity o f G ross Function: Self-reported ability to use the toilet and strength o f grip on a dynam ometer, M acArthur Study (1988, 1991, and 1995). 9 5 .1 % A b le (n=4IO) 9 9 .5 % A b le (n=429) 4 .4 % U n a b le /N e e d H e lf (n=19) 1 0 0 % A b le (n=431) 0 .2 % A b le (n = l) 0 .5 % U n a b le /N e e d H e lp (n=2) 0 .2 % U n a b le /N e e d H e lp (n=l) mean (s.d.) = 20.00 (— ) range = not applicable mean (s.d.) = 23.79 (9.80) range = 2.00 - 54.00 no.2 unable to perform = 1 mean (s.d.) = 16.11(7.16) range = 1.00 - 28.00 mean (s.d.) - 8.00 (— ) range = not applicable mean (s.d.)1 = 28.77(10.25 range = 10.00 - 62.00 mean (s.d.) = 28.49 (9.97) range = 8.00 - 58.00 mean (s.d.) = 16.50 (3.54) range = 14.00 - 19.00 1 9 8 8 1 9 9 1 1 9 9 5 Note: Percentage values indicate the percent o f reported ability from baseline. Arrows indicate how individuals self-reported their ability to perform a given task over tim e (able or unable/need help). The arrows point to boxes that describe characteristics o f the perform ance-based measure: mean, ’standard deviation, range (in seconds), and 2 num bcr o f people unable to perform. IO L/i Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Figure IV-6. Fine M otor Dexterity o f Gross Function: Self-reported ability to use the toilet and perform ance on hand signature, M acArthur Study (1988, 1991, and 1995). 9 4.9% A ble (n= 41I) 99.5% A ble (n=429) 4.2% U nable/N eed Hclf (n= 18) 100% A ble (n=431) 0.2 % A ble (n = l) 0 .5% U nable/N eed H elp (n=2) 0.2 % U nable/N eed Heir (n = l) mean (s.d.) = 10.19(3.76) range = 3.40 - 30.00 no." unable to perform = 2 mean (s.d.) = 11.90 (— ) range = not applicable mean (s.d.) = 26.10 (— ) range = not applicable mean (s.d.) = 15.34(7.12) range = 6.50 - 30.00 no. unable to perform = 2 m ean (s.d.) = 8.80 (2.44) range = 2.30 - 19.60 mean (s.d.)1 = 8.84 (2.27) range = 3.90 - 22.70 mean (s.d.) = 12.45 (1.63) range = 11.30 - 13 .60 1988 1991 1995 Note: Percentage values indicate the percent o f reported ability from baseline. Arrows indicate how individuals self-reported their ability to perform a given task over tim e (able or unable/need help). The arrows point to boxes that describe characteristics o f the perform ance-based measure: mean, standard deviation, range (in seconds), and I num ber o f people unable to perform. u O n Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Figure IV-7. Fine M otor Dexterity: Self-reported ability to write/handle sm all objects and strength o f grip on a dynam om eter, M acArthur Study (1988, 1991, and 1995). 9 9 .1 % A b le - (n=427) / 1 0 0 % A b le (n=430) 1 0 0 % A b le (n -4 3 0 ) 0 .7 % U n a b le mean (s.d.) = 28.43 (9.98) range = 8.00 - 58.00 mean (s.d .)-2 8 .7 7 (10.25) range = 10.00 - 62.00 mean (s.d.) = 23.44 (9.80) range = 1.00 - 54.00 (n=3) mean (s.d.) = 17.33 (14.47) range = 8.00 - 34.00 _jio_jjnabl£mj)£rform^J^ 1 9 8 8 1 9 9 1 1 9 9 5 Note: Percentage values indicate the percent o f reported ability from baseline. Arrows indicate how individuals self-reported their ability to perform a given task over tim e (able or unable/need help). The arrows point to boxes that describe characteristics o f the performance-based measure: mean, 'standard deviation, range (in seconds), and 'num ber o f people unable to perform. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Figure IV-8. Fine M otor Dexterity: Self-reported ability to w rite/handle sm all objects and perform ance on tim ed hand signature, M acArthur Study (1988, 1991, and 1995). mean (s.d.) = 10.39(4.06) ran^e = 3.40 - 30.0 no." unable to perform = 4 9 9 .5 % A b le <n=429) 1 0 0 % A b le (n = 4 3 I) 1 0 0 % A b le (n = 4 3 I) mean (s.d.) =8.84 (2.27) range = 3.90 - 22.70 mean (s.d.) = 8.82 (2.45) range = 2.30 - 19.60 0 .5 % U n a b le (n=2) mean (s.d.) = 22.65 (4.88) range = 19.20-26.10 1 9 8 8 1 9 9 1 1 9 9 5 Note: Percentage values indicate the percent o f reported ability from baseline. Arrows indicate how individuals self-reported their ability to perform a given task over tim e (able or unable/need help). The arrow s point to boxes that describe characteristics o f the perform ance-based measure: mean, 'standard deviation, and range (in seconds) and 2 num ber o f people unable to perform. f3 00 Measures. The measures o f physical functioning used to examine the accuracy of reported dysfunction are described here. In addition, in order to identify personal, health, behavioral, and psychological characteristics that predispose older persons to discrepancies in measures o f physical functioning, a series o f variables that represent these characteristics are described, as well. Physical Functioning. Table IV -1 and Table IV-2 shows the mean (standard deviation), median, and range o f values for among those who were able to perform the following tasks in 1988 and 1995: walking at a normal pace, rising from a seated position five times, signing one's name, and gripping a dynamometer. A median time score was chosen as a cut-point for the creation o f ordinal variables to describe these measures because, in the majority of cases, the distributions were slightly skewed (e.g.. walking in 1988 and 1995: repeated chair stands in 1995. signing one's name in 1988 and 1995). Although the distribution for grip strength was normal in both 1988 and 1995. the median score was chosen as a cut-point for this measure, for consistency. For the measures where poorer performance is based on a higher score (in seconds) (e.g.. walking, completing five repeated chair stands, and signing one's name) an ordinal variable was constructed ranging from 0 to 2: 0 = individuals who could not complete the task, 1= slower times (> median time) and 2 = faster times (< median time). A poorer score on grip strength, is based on a lower score (in kilograms). Consequently, the ordinal variable for grip strength was constructed ranging from 0 to 2: 0 = individuals 129 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. who could not complete the task, 1= weaker (< median weight) and 2= stronger (> median weight). Tables IV -1 and IV-2 shows the frequencies o f these ordinal variables once they are categorized, as indicated above. Table IV -1. Descriptive statistics for performance-based measures in 1988 P e r fo r m a n c e -B a se d M e a su r e s M e a n (S ta n d a r d D e v ia tio n ) M e d ia n R a n g e o f V a lu e s P e r c e n t % (n ) 10 Foot Walking Course Unable Slower Time (> median) Faster Time (< median) 9.64(2.18) 9.20 5 .1 0 -2 1 .6 0 0.7% (5) 49.6% (355) 49.7% (356) 5 Repeated Chair Stands Unable Slower Time (> median) Faster Time (< median) 11.96 (2.85) 11.70 5.20 - 20.40 0.0% (0) 49.3% (356) 50.7% (366) Hand Signature Unable Slower Time (> median) Faster Time (< median) 9.09 (2.59) 8.70 3.90 - 22.70 0.4% (3) 49.0% (352) 50.6% (364) Grip Strength Unable Weaker (< median) Stronger (> median) 29.26(9.93) 28.00 10.00-62.00 0.1% (1) 53.6% (382) 46.3% (330) Table IV-2. Descriptive statistics for performance-based measures in 1995. P e r fo r m a n c e -B a se d M e a su r e s M e a n (S ta n d a r d D e v ia tio n ) M e d ia n R a n g e o f V a lu e s P e r c e n t % (n ) 10 Foot Walking Course Unable Slower Time (> median) Faster Time (< median) 12.50(7.51) 10.90 5.90 - 90.00 4.2% (19) 47.0% (211) 48.8% (219) 5 Repeated Chair Stands Unable Slower Time (> median) Faster Time (< median) 11.32 (3.77) 10.70 3 .4 0 -3 8 .3 0 22.6% (115) 38.3% (195) 39.1% (199) Hand Signature Unable Slower Time (> median) Faster Time (< median) 10.56 (4.35) 9.40 3 JO - 30.00 0.9% (4) 48.6% (220) 50.6% (229) Grip Strength Unable Weaker (< median) Stronger (> median) 23.29 (9.87) 22.00 1 .0 0 -5 4 .0 0 22.6% (115) 55.5% (251) 44.2% (200) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dichotomous variables were then created for these performance measures reflecting the ability (or dysfunction) and the time required to complete the exercise (reference group in parentheses): unable (able) and slow/unable (fast). Table IV-3 shows the frequencies for these dichotomous variables both in 1988 and 1995. Based on data from 1988 and 1995, change scores were defined as dichotomous variables (reference group in parentheses): become unable, ” 1" (remain able. "0") and become slower or unable, “ 1” (remain fast. "(T). Table IV-3 shows that less than 25% of respondents became unable to perform on these four performance-based measures: walking 10 feet (4.0%). completing five repeated chair stands (22.6%). signing one's name (0.9%). and gripping a dynamometer (0.2%). Between 11.0% and 51.0%. however, became slower or unable (and in the case o f grip strength, became weaker/unable) to perform these exercises: walking 10 feet (35.4%). completing five repeated chair stands (50.8%). signing one's name (26.9%). and gripping a dynamometer ( 11.2%). Similarly, the four self-reported measures were first dichotomously scored to indicate being unable (“ 1") or able (”0”) both in 1988 and 1995. These variables were then used to develop change scores to indicate: becoming unable ("1”) or remaining able (“O ’") from 1988 to 1995. Table IV-4 shows that the majority o f participants reported the continued ability to function. Less than 15% reported becoming unable: walking across a room (11.7%), stooping, crouching, kneeling (10.9%), and using the toilet (6.7%). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table IV-3. Descriptive statistics for performance-based measure o f physical functioning in 1988. 1995, and from 1988 to 1995____________________________________________________ P e r fo r m a n c e -B a se d M e a su r e s 1 9 8 8 P e r c e n t % (n ) 1 9 9 5 P e r c e n t % (n ) 1 9 8 8 -1 9 9 5 P e r c e n t % (n ) 10 Foot W alking Course Unable Able Become Unable Remain Able 0.7% (5) 99.3% (711) 4.2% (19) 95.8% (430) 4.0% (18) 96.0% (428) Slow/Unable Fast Become Slower/Unable Remain Fast 50.3% (360) 49.7% (356) 51.2% (230) 48.8% (219) 35.4% (81) 64.6% (148) Repeated Chair Stands Unable Able Become Unable Remain Able 0.0% (0) 100.0% (722) 22.6% (115) 77.4% (394) 22.6% (115) 77.4% (394) Slow Fast Become Slower/Unable Remain Fast 49.3% (356) 50.7% (366) 60.9% (310) 39.1% (199) 50.8% (130) 49.2% (126) Hand Signature Unable Able Become Unable Remain Able 0.4% (3) 99.6% (716) 0.9% (4) 99.1% (449) 0.9% (4) 99.1% (448) Slow/Unable Fast Become Slower/Unable Remain Fast 49.4% (355) 50.6% (364) 49.4% (224) 50.6% (229) 26.9% (66) 73.1% (179) Grip Strength Unable Able Become Unable Remain Able 0.1% (1) 99.9% (712) 0.2% (1) 99.8% (451) 0.2% (1) 99.8% (448) W eak/Unable Stroke Become W eaker/Unable Remain Strong 53.7% (383) 46.3% (330) 55.8% (252) 44.2% (200) 11.2% (21) 88.8% (166) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table IV-4. Descriptive statistics for self-reported measures of physical functioning in 1988, 1995, and from 1988 to 1995 S e lf-R e p o r te d M e a s u r e s 1 9 8 8 P ercen t(n ) 1 9 9 5 P e r c e n t % (n ) 1 9 8 8 -1 9 9 5 P e r c e n t % (n ) Walk Across Room Unable/Need Help Able Become Unable/Need Help Remain Able 0.0% (0) 100.0% (721) 11.7% (63) 88.3% (477) 11.7% (63) 88.3% (477) Stoop/Crouch/Kneel Unable Able Become Unable Remain Able 1.0% (7) 9 9 .0 % (713) 10.8% (58) 89.2% (477) 10.9% (58) 89.1% (472) Use the Toilet Unable/Need Help Able Become Unable/Need Help Remain Able 0.0% (0) 100.0% (722) 6.7% (36) 93.3% (504) 6.7% (36) 93.3% (504) Hypothesized Independent Variables. Demographic characteristics. Studies have shown that age. gender, and socioeconomic status influence physical functioning among older persons (Manton. 1988; Guralnik and Kaplan. 1989; Strawbridge et al.. 1992; Crimmins and Saito. 1993; Guralnik et al„ 1993; Kaplan et al„ 1993; Boult et al.. 1994). For a more detailed discussion o f these studies, the author directs the reader to Chapter II. In the present analysis age was coded in years. In addition, dichotomous variables were used to indicate the following characteristics (reference group in parentheses): female (male), having less than 12 years o f education (at least 12 years o f education), having an annual income less than $10,000 (annual income o f at least $10,000). In this study, as the MacArthur Study is a sub-sample o f the Established Populations for Epidemiologic Studies (EPESE) study, income and education were dichotomized based on the findings from Guralnik et al. 133 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (1993). As discussed in Chapter II, Guralnik et al. (1993) used a longitudinal sample from the EPESE to assess the role o f demographic factors in maintaining mobility in older persons. They found that persons with lower (< $5,000 per year) and middle ($5,000 to $9,999 per year) income levels were associated with significantly lower rates of maintaining mobility (on reported measures o f walking up and down stairs and walking half a mile) than persons with high (> $10,000 per year) income levels. In addition, they found that persons with low (< 8 years) and medium (9 to 12 years) levels o f education were also associated with lower rates o f maintaining mobility compared to those with more than 12 years of education. Health conditions and events. As detailed in Chapter II. studies have shown that specific disease conditions influence decrement in physical functioning (Satarino et al.. 1990; W o lfetal.. 1992; Guralnik et al.. 1993; Kurtz etal., 1993: Ettinger et al., 1994; Mulrow et al.. 1994: Ormel et al.. 1998; Kurtz et al.. 1999). Consequently, five prevalent health conditions and 2 health events were examined: whether the respondent had ever had cancer (excluding minor skin cancers), diabetes, or high blood pressure (average systolic and diastolic measurements after three seated readings greater than 140/90 mm Hg or taking blood pressure medication), myocardial infarction, or a stroke. The two health events considered were: "having fractured a hip" and "having fractured other bones” as dichotomous independent variables. Preliminary analyses showed that due to the small number o f cases of these health conditions and events (at baseline) and the small sample size o f the predicted dependent variables, retaining these heaith characteristics as individual independent variables was not possible (see Table IV-5). 134 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table IV-5. Cross-tabulations between baseline levels o f health conditions/events and the three dependent variables o f interest: 1) over-reporting the onset o f disability walking; 2) over-reporting the onset o f disability toileting (with grip strength); and 3) over-reporting the onset o f disability toileting (with hand signature). 1. Over-reporting onset o f disability walking H e a lth C o n d itio n s/E v e n ts A c c u r a te R e p o r tin g O n se t of D isa b ility W a lk in g P e r c e n t % (n ) O v e r -R e p o r tin g O n se t of D isa b ility W a lk in g P e r c e n t % (n ) Cancer 20.0% (79) 26.7% (8) Diabetes 11.6% (46) 10.0% (3) High Blood Pressure 38.6% (152) 37.9% (11) M yocardial Infarction 12.1% (48) 6.7% (2) Stroke 1.8% (7) 0.0% (0) Hip Fracture 2.5% (10) 6.7% (2) O ther Bone Fractures 22.3% (88) 36.7% (11) 2. Over-reporting onset o f disability toileting (with grip strength) H e a lth C o n d itio n s/E v e n ts A c c u r a te R e p o r tin g O n se t of D isa b ility T o ile tin g (W ith Grip Strength) P e r c e n t % (n ) O v e r -R e p o r tin g O n se t of D isa b ility T o ile tin g (With Grip Strength) P e r c e n t % (n ) Cancer 20.9% (88) 82.6% (4) Diabetes 11.1% (47) 17.4% (4) High Blood Pressure 38.1% (160) 30.4% (7) M yocardial Infarction 11.6% (49) 13.0% (3) Stroke 1.7% (7) 0.0% (0) Hip Fracture 2.6% (1 1) 4.4% (1) O ther Bone Fractures 22.3% (94) 21.7% (5) 3. Over-reporting onset o f disability toileting (with hand signature) H e a lth C o n d itio n s/E v e n ts A c c u r a te R e p o r tin g O n se t of D isa b ility T o ile tin g (With Hand Signature) P e r c e n t % (n ) O v e r -R e p o r tin g O n se t of D isa b ility T o ile tin g (With Hand Signature) P e r c e n t % (n ) Cancer 20.6% (87) 17.4% (4) Diabetes 11.6% (49) 17.4% (4) High Blood Pressure 38.1% (160) 30.4% (7) Myocardial Infarction 11.8% (50) 13.0% (3) Stroke 1.7% (7) 0.0% (0) Hip Fracture 2.6% (11) 4.4% (1) Other Bone Fractures 22.5% (95) 21.7% (5) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Consequently, two dummy variables were used to indicate whether individuals have at least one o f the five diseases in 1988 and whether individuals have an incidence of disease(s) in 1991 or 1995. The category describing the absence o f any diseases from 1988 to 1995 served as the referent group. Similarly, two dummy variables are used to indicate whether individuals have fractured a hip and/or other bones in 1988 and whether they have an incidence o f fracture(s) in 1991 or 1995. The category describing the absence o f any fractures from 1988 to 1995 served as the referent category. Cognitive performance. In the MacArthur Study, the total cognitive score represented a summation o f the performance on five sub-scales: 1) spatial memory. 2) similarity o f abstract concepts. 3) language. 4) delayed verbal memory, and 5) spatial orientation. Spatial memory was assessed by implementing the delayed Span Test, which requires the placement o f circular disks on a board in a specific order and then asking individuals to identify the new disk that has been added (Moss et al.. 1986). This test identifies individuals with neurological diseases. By scoring individuals from 0 to 17. this easily administered test focused on measuring spatial memory separate from verbal related ability. The Wechsler Adult Intelligence Scale -Revised (WAIS-R) was used to evaluate abstract concepts (Wechsler. 1981). Participants were asked to explain how different objects or concepts (e.g., work and play) were similar and scored from 0 to 16 on this task. The Boston Naming Test tests language (Kaplan et al., 1983). Individuals were shown 18 drawings (e.g., broom, bench, tree, harp, etc.) and asked to name each object. Only items correctly named were scored within a time limit of 10 seconds per picture. Relying on the Boston Naming Test (Kaplan et al.. 1983), delayed verbal memory was tested. Individuals were asked to recall the images shown during the 136 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. naming test. Participants were challenged by this task as they were not asked explicitly to remember these images during the naming test. Finally, spatial orientation was tested through measurement o f subject’s ability to copy geometric figures including a diamond, a diamond in a square, and a three-dimensional cube (Rosen et al.. 1984). Participants were asked to draw these figures and a single rater scored drawings using a standardized system. A total cognitive score, representing the sum o f these five tests, was determined by the original investigators o f the MacArthur Study and ranged from zero to eighty-nine such that lower scores indicate poorer cognitive functioning (Inouye et al.. 1993). At baseline, total cognitive score was normally distributed (see Figure III-1 in Chapter 3) with a mean score o f 53.95 (standard deviation 9.54). Depression. The depression scale is adapted from the Hopkins Symptom Check List (Derogatis et al., 1974). Participants were assigned a score based on the sum of eleven items such that individuals with higher continuous scores (ranging from eleven to forty-four) have a higher depression score. Individuals had scores that ranged from eleven to thirty-two with a mean of 14.44 (standard deviation o f 3.41). Due to the skewed distribution of this scale at baseline (see Figure II1-2 in Chapter III), a dichotomous variable was created based on the highest fertile (> 15) to indicate levels of depression ("1” = high levels o f depression and ”0" = moderate/low levels). Health behaviors. Three health behaviors were also examined: body composition, cigarette smoking, and alcohol consumption. Body mass index (kg/m2) captured the relative weight o f each individual. Based on tertiles. a categorical variable describes individuals with low (< 24.27). medium (24.28 to 27.76), and high (> 27.77) BMI levels. 137 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Two dummy variables were used to indicate categories low BMI and high BMl with category medium BMI serving as the referent group. Smoking behavior is measured using two items that assessed the use and frequency of cigarette smoking. A categorical variable describes individuals who currently smoke, have smoked in the past, and have never smoked. Two dummy variables were used to indicate categories current smoker and past smoker with category never smoked serving as the referent. Finally, alcohol consumption is measured by four items that assessed the type and quantity o f alcohol consumed. Again two dummy variables were used to indicate: < 1 gram ethyl alcohol per month and > 1 gram ethyl alcohol per month with category none in the past year serving as the reference group. Psychological Characteristics. Two psychological resources were measured: self-efficacy and mastery. Self-efficacy and mastery were operationalized by the scale developed by Rodin and McAvay (1993) and Pearlin and Schooler (1978). respectively. Individuals can score between nine and thirty-six on the self-efficacy scale and between seven and twenty-eight on the scale for mastery. Both of these scales were normally distributed at baseline with higher scores indicating higher levels of these resources. Procedure. For the analyses examining change in reported dysfunction, changes in the performance-based measures are treated as the "gold standard." Although this is an arguable assumption, interviewers demonstrated the exercises before each participant was asked to walk the course, thereby reducing the chance o f a misunderstanding. Thus, the validity o f change in the self-reported measures should determine which individuals actually became disabled or slow down. The validity of the change in self-reported 138 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. measures was determined by calculating the sensitivity and specificity. A good indication o f which individuals actually report becoming disabled and those who remain able is referred to as the validity o f the test. Validity has two components: sensitivity and specificity. Sensitivity is defined as the ability to identify correctly those who have the disease. Specificity is defined as the ability to identify correctly those who do not have the disease. Ideally, measures with high specificity and sensitivity are sought. Rose and Barker (1978. p. 1071) defined these concepts as: Sensitivity: A/A+B (see Figure IV-9). Specificity: D/C+D (see Figure IV-9). Figure IV-9. Validity in a self-reported measure R e p o r te d O b se r v e d P e r fo r m a n c e Unable Able Unable A C Able B D Total A+B C+D Due to the longitudinal nature o f this study, these definitions are modified as follows (see Figure IV -10): • Sensitivity: The ability to correctly identify those who become unable to function (A/A+B). • Specificity: The ability to correctly identify those who remain able to function (D/C+D). Figure IV -10. Validity o f change in a self-reported measure R e p o r te d O b se r v e d P e r fo r m a n c e Become Unable Remain Able Become Unable A C Remain Able B D Total A+B C+D Agreement: A+D/(A+B)+(C+D) In addition, agreement is calculated by summing the number o f individuals who report and perform becoming unable (cell A o f Figure IV -10) and the number o f individuals 139 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. who report and perform remaining able (cell D o f Figure IV -10) divided by the total number o f individuals (cells A, B, C. D o f Figure IV-10). Logistic regression is used to investigate the impact o f lower levels o f cognitive functioning on the change in inaccurate reporting (i.e.. under or over-reporting onset o f dysfunction). By referring to Figure IV -10, two dichotomous dependent variables were constructed to identify individuals that (reference group in parentheses): under-reported the onset o f dysfunction, "B” (accurately estimated * ‘A") and over-reported the onset of dysfunction, "C” (accurately estimated “D). Thus individuals who under-report have lower self-reported levels o f dysfunction onset compared to performance-based levels; and individuals who over-report have higher self-reported levels of dysfunction onset compared to performance-based levels. The logistic regression models are first built by computing the correlations between the dependent variables (under and over-reporting o f dysfunction onset) and baseline levels o f each hypothesized covariate (demographic, health, behavioral, and psychological characteristics). The covariates that are significantly correlated are included. Coefficients from the logistic regression equations predict the log odds of under or over-reporting the onset o f dysfunction. Odds ratios and 95 % confidence intervals appear in Table IV-15. C. Results As shown in Figures IV-2. IV-3. IV-4. IV-5. IV-6. IV-7. and IV-8. from 1988 to 1995, the vast majority o f participants reported the continued ability to walk across a room (93.3%, see Figure IV-2), “toilet” (98.7%, 95.1%, and 95.1% - for 140 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. comparisons 2 ,4 , and 5 respectively; see Figures IV-3, IV-5, IV-6), stoop, crouch, or kneel (94.2%; see Figure IV-4), and write/handle small objects (99.1% and 99.5%- for comparisons 6 and 7, respectively; see figures IV-7 and IV-8). For all seven comparisons, the observed performance on the matched items changes in a predictable manner (i.e., among individuals reporting the inability to function, more time was needed to perform the activity or a weaker grip strength was recorded). Unfortunately, these comparisons show small levels of variability in reported functioning. By 1995, a small percentage o f individuals report dysfunction: 6.5% for walking (see Figure IV-2); 1.3%. 4.6%. and 4.3% for "toileting" (for comparisons 2, 4. and 5. respectively; see Figures IV- 3, IV-5. IV-6); 4.4% for stooping, crouching, or kneeling (see Figure IV-4); and for writing/handling small objects (0.7% for comparison 6 and 0.5% for comparison 7; see Figures IV-7 and IV-8). Since the MacArthur Study was not originally designed to examine the concordance between self-report and performance measures, these a priori comparisons may misrepresent similar movements or kinetics. Consequently, comparison o f reported and performed functioning are limited in this investigation. The remaining analyses focus on the comparison o f reported and performed functioning involving some component o f larger motor function: walking, "toileting.” and stooping, crouching, or kneeling. For the comparisons chosen, the sensitivity, specificity, and agreement levels are calculated both in cross-section and longitudinally. This information is presented in Tables IV-6 to IV-13. As participants were selected for participation in the MacArthur Study based on their high physical and cognitive functioning at baseline, it was not 141 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. possible to conduct cross-sectional analyses at baseline (in 1988) because these participants reported few, if any, problems with physical functioning. Instead, data from the follow-up assessment, in 1995. was used to examine the relationship between reported and performed measures of physical functioning, in the cross-section. Here, cross-sectional analysis are presented for comparison with earlier studies (Merrill et al.. 1997 and Ferrer et al.. 1999). In the cross-section, while the "top" portion of Tables IV-6 to IV-9 compares the reported and performed ability of the a priori matched measures, the "bottom" portion compares the reported ability to the speed (or strength) of the performance-based measures. The longitudinal comparisons follow a similar format. In Tables IV-10 to IV -13. while the "top" portion compares the change in reported and performed ability o f the a priori matched measures, the "bottom" portion compares the change in the reported ability to the change in speed (or strength) o f the performance- based measure. In the cross-section, the comparison between the reported ability to walk across a room and the ability to walk 10 feet is presented in the top portion of Table IV-6. Specificity (93%) was higher than sensitivity (79%) indicating that although the reported measure is less accurate in capturing those who had functional limitations, it accurately estimates those who were able to function. The agreement between these measures was 92%. For the comparison between the reported ability to walk and the speed of walking (bottom of Table IV-6), sensitivity was low (19%) and specificity was 100%. Thus, although 100% o f those who were able to walk across a room walk quickly, only 19% of those who were unable, walked slow or were unable to walk the ten foot course. Agreement between these measures is approximately 58%. 142 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table IV-6. Agreement between reported dysfunction and observed performance o f gait function in 1995 R e p o r te d F u n c tio n a l L im ita tio n s O b se r v e d P e r fo r m a n c e S e n sitiv ity S p e c ific ity % A g r e e m e n t 10 Foot W alking Course W alk Across Room Unable Able Unable/Need Help 15 30 0.79 0.93 9 2 % Able 4 400 Total 19 430 10 Foot Walking Course W alk Across Room Slower/Unable Fast Unable/Need Help 44 I 0.19 1.00 5 8 % Able 186 218 Total 230 219 Table IV-7. Agreement between reported dysfunction and observed performance o f lower body strength in 1995______ ___________________ R e p o r te d F u n c tio n a l L im ita tio n s O b se r v e d P e r fo r m a n c e S e n sitiv ity S p e c ific ity % A g r e e m e n t Stoop/Crouch/Kneel Unable Able Total Repeated Chair Stands Unable Able 35 17 80 374 115 391 0.30 0.96 81 % Stoop/Crouch/Kneel Unable Able Total Repeated Chair Stands Slower/Unable Fast 48 4 259 195 307 199 0.16 0.98 4 8 % 143 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The top portion o f Table IV-7 shows the comparison between the reported ability to stoop, crouch, or kneel and the observed ability to complete five chair stands. Specificity is 96% and sensitivity is 30% indicating that although the reported measure is less accurate in capturing those who were limited in their physical functioning, it accurately estimates those who were able to function. Agreement between these measures is 81%. The bottom portion o f Table IV-7 shows the comparison between the reported ability to stoop, crouch, or kneel and the observed time required to complete the five chair stands. Similar to the comparison based on the ability to complete the five chair stands, this comparison shows high specificity (98%). Furthermore, a lower proportion o f agreement is found for this comparison (48%). compared to the comparison based on the ability to complete five chair stands, and the sensitivity was only 16% (i.e.. 16% o f individuals who reported being unable to stoop, crouch, or kneel completed the 5 repeated chair stands at a slow pace or were unable to complete this task). Comparisons between the reported ability to ‘'toilet" and the: 1) ability and time to sign one’s name and 2) ability and strength of one’s grip on a dynamometer (see top portion o f Table IV-8 and Table IV-9) show agreement levels above 90%. The comparison between the reported ability to "toilet” and the ability to sign one’s name have the highest sensitivity (50%) among these comparisons indicating that the reported inability to "toilet” accurately reflects the observed inability to sign one's name for half o f the cases. The lower portions o f Table IV-8 and Table IV-9 also show low values of sensitivity when comparing the reported inability to "toilet" with the time required to sign one’s name (9%) and grip the dynamometer (8%). These values indicate that the reported inability to "toilet” reflects the poorer performance (in seconds) o f signing 144 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table IV-8. Agreement between reported dysfunction and observed performance o f fine motor dexterity o f gross m otor function, hand signature in 1995___________________________________ R e p o r te d F u n c tio n a l L im ita tio n s O b se r v e d P e r fo r m a n c e S e n sitiv ity S p e c ific ity % A g r e e m e n t Use the Toilet Unable/Need Help Able Total Hand Signature Unable Able 2 23 2 426 4 449 0.50 0.95 9 4 % Use the Toilet Unable/Need Help Able Total Hand Signature Slower/Unable Fast 21 4 203 225 224 229 0.09 0.98 5 4 % Table IV-9. Agreement between reported dysfunction and observed performance o f fine motor dexterity o f gross motor function, grip strength in 1995_____________________________________ R e p o r te d F u n c tio n a l L im ita tio n s O b s e r v e d P e r fo r m a n c e S e n sitiv ity S p e c ific ity % A g r e e m e n t Use the Toilet Unable/Need Help Able Total Grip Strength Unable 0 1 1 Able 23 428 451 0.00 0.95 95 % Use the Toilet Unable/Need Help Able Total Grip Strength Weaker/Unable 20 232 252 Strong 3 197 200 0.08 0.99 48 % 145 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. one's name and strength (in kilograms) in one’s grip in less than 10% o f the cases. The specificity for all four o f these comparisons was at least 95% (see Table IV-8 and Table IV-9) indicating that the reported ability to “toilet" accurately reflects the 1) ability to sign one's name and grip a dynamometer: and 2) ability to sign one's name quickly and have a strong grip strength. The comparison between change in the reported ability to walk across a room and change in the ability to walk 10 feet is presented in the top portion of Table IV -10. Sensitivity (83%) and specificity (93%) were high indicating that the reported change in dysfunction is accurate. The agreement between these measures was 93%. For the comparison between change in the reported ability to walk and the speed o f walking (bottom of Table IV-10). sensitivity was low (27%) and specificity was 100%. Thus although 100% o f those who remain able to walk across a room continue to walk fast, only 27% o f those who become unable, walk slower or become unable to walk. Agreement between these measures is approximately 72%. The top portion o f Table IV -11 shows the comparison between change in the reported ability to stoop, crouch, or kneel and the observed change in the ability to complete five chair stands. Specificity is 96% and sensitivity is 30% indicating that although the reported measure is less accurate in capturing those who become limited in their physical functioning, it accurately estimates those who remain able to function. Agreement between these measures is 81%. The bottom portion o f Table IV -11 shows the comparison between change in the reported ability to stoop, crouch, or kneel and the observed change in time required to complete the five chair stands. Similar to the comparison based on the change in the ability to complete the five chair stands, this 146 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table IV-10. Agreement between reported dysfunction and observed performance o f gait function (1988-1993)_____________________________________________ ______ R e p o r te d F u n c tio n a l L im ita tio n s O b se r v e d P e r fo r m a n c e S e n sitiv ity S p e c ific ity % A g r e e m e n t 10 Foot Walking Course Become Remain Walk Across Room Unable Able Become Unable/Need 15 30 0.83 0.93 9 3 % Help Remain Able 3 398 Total 18 428 10 Foot Walking Course Become Remain Walk Across Room Slower/Unable Fast Become Unable/Need 17 0 0.27 1.00 72 % Help Remain Able 64 148 Total 81 148 Table IV -11. Agreement between reported dysfunction and observed performance o f lower body strength (1988-1995)_____________________________________________________________________ R e p o r te d F u n c tio n a l L im ita tio n s O b se r v e d P e r fo r m a n c e S e n sitiv ity S p e c ific ity % A g r e e m e n t Repeated Chair Stands Become Remain Stoop/Crouch/Kneel Unable Able Become Unable 35 17 0.30 0.96 Remain Able 80 369 Total 115 386 Repeated Chair Stands Become Remain Stoop/Crouch/Kneel Slower/Unable Fast Become Unable 23 1 0.18 0.99 Remain Able 107 124 Total 130 125 81 % 5 8 ' 147 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. comparison shows high specificity (99%). Furthermore, a lower proportion of agreement is found for this comparison (58%), compared to the comparison based on the ability to complete five chair stands, and the sensitivity was only 18% (i.e.. 18% of individuals who reported becoming unable to stoop, crouch, or kneel completed the 5 repeated chair stands at a slower pace or became unable to complete this task). Comparisons between change in the reported ability to "toilet" and change in: the 1) ability and time to sign one's name and 2) ability and strength o f one’s grip on a dynamometer (see top portion of Table 1V-12 and Table 1V-13) show agreement levels above 90%. The comparison between change in the reported ability to "toilet" and change in the ability to sign one's name have the highest sensitivity (50%) among these comparisons indicating that reported decrement "toileting" accurately reflects the observed decrement in ability to sign one's name for half o f the cases. The lower portions ofTable IV-12 and Table IV-13 also show low values o f sensitivity when comparing decrement on the reported ability to "toilet” with the decrement on the time required to sign one's name (16%) and grip the dynamometer (14%). These values indicate that the reported decrement "toileting” reflects the decrement in time (in seconds) required to sign one's name and strength (in kilograms) in one's grip in slightly over 10% o f the cases. The specificity for all four o f these comparisons was at least 95% (see Table IV -12 and Table IV-13) indicating that the reported continued ability to "toilet” accurately reflects the continued 1) ability to sign one's name and grip a dynamometer; and 2) ability to quickly sign one's name and maintain a strong grip strength. 148 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table IV-12. Agreem ent between reported dysfunction and observed performance o f fine motor dexterity o f gross motor function, hand signature (1988-1995)_______________________________ R e p o r te d F u n c tio n a l L im ita tio n s O b se r v e d P e r fo r m a n c e S e n sitiv ity S p e c ific ity % A g r e e m e n t H a n d S ig n a tu re B e c o m e R e m a in U s e th e T o ile t U n a b le A b le B e c o m e U n a b le /N e e d 2 23 0 .5 0 0 .9 5 9 4 % H e lp R e m a in A b le 2 4 2 5 T o ta l 4 4 4 8 H a n d S ig n a tu re B e c o m e R e m a in U s e th e T o ile t S lo w e r/U n a b le F a st B e c o m e U n a b le /N e e d 9 0 .1 6 0 .9 8 7 6 % H e lp R e m a in A b le 5 7 176 T o ta l 6 6 179 T a b le IV -1 3 . A g r e e m e n t b e tw e e n re p o r te d d y s fu n c tio n a n d o b s e rv e d p e rfo rm a n c e o f fin e m o to r d e x te rity o f g r o s s m o to r fu n c tio n , g rip s tre n g th (1 9 8 8 - 1 9 9 5 )_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ R e p o r te d F u n c tio n a l L im ita tio n s O b se r v e d P e r fo r m a n c e S e n sitiv ity S p e c ific ity % A g r e e m e n t U s e th e T o ile t B e c o m e U n a b le /N e e d H e lp R e m a in A b le T o ta l G rip S tre n g th B e c o m e R e m a in U n a b le A b le 0 23 1 4 2 5 1 4 4 8 0 .0 0 0 .9 5 9 5 % U s e th e T o ile t B e c o m e U n a b le /N e e d H e lp R e m a in A b le T o ta l G rip S tre n g th B e c o m e R e m a in W e a k e r/U n a b le S tro n g 3 2 18 164 21 166 0 .1 4 0 .9 9 8 9 % Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. These tables (Tables IV-10 to IV-13) reveal that it is only possible to examine discrepancies between self-reported and performance-based measures in the direction o f over-reported onset of dysfunction for three comparisons: gait function (Table IV-10): 2) “toileting’'' (with grip strength, see Table IV-12) and 3) “toileting" (with signing one’s name, see Table IV-13). Unfortunately, all o f the discrepancies in the direction o f under reporting and those in the direction of over-reporting specific to speed and strength of performance were not possible due to limited sample sizes. Furthermore, the sample size for examining the over-reported discrepancy between the ability to stoop, crouch, or kneel and the ability to complete 5 repeated chair stands was too small for analysis. Table IV -14 shows the correlates o f over-reporting the onset of dysfunction walking across a room (based on change in reported and performed ability) and “toileting" (based on change in reported ability and change in grip strength and signing one's name). Covariates that are significantly correlated with over-reporting the onset o f dysfunction walking across a room are age, gender, fractured bones, cognitive functioning, and BtMI. Depression, smoking status, and mastery are the only covariates that are significantly correlated with the over-reporting the onset o f dysfunction “toileting." For comparison across the three dependent variables, all o f these variables were retained (i.e., age. gender, fractured bones, cognitive functioning, depression. BMI, smoking status, and mastery). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T a b le I V - 14. C o r re la tio n s b e tw e e n h y p o th e s iz e d c o v a ria te s a n d o v e r- re p o rtin g o n s e t o f fu n c tio n a l lim ita tio n s V a r ia b le O v e r - R e p o r tin g W a lk in g O v e r -R e p o r tin g T o ile tin g (w ith g rip s tre n g th ) O v e r -R e p o r tin g T o ile tin g (w ith h a n d s ig n a tu re ) D e m o g r a p h ic C h a r a c te r istic s A g e 0 .1 0 “ 0 .0 5 0 .0 5 F e m a le G e n d e r 0 .0 9 * 0 . 0 1 0.01 E d u c a tio n < 12 y e a rs 0 .0 6 0 .0 4 0 .0 2 In c o m e < S I 0 ,0 0 0 0 .0 8 0 .0 2 0 .0 4 In c o m e m is s in g - 0.01 -0 .0 2 -0 .0 2 H e a lth C h a r a c te r istic s D is e a s e s 0 .0 3 0.01 0 .0 0 F ra c tu re d B o n e s 0.11* 0 .0 5 0 .0 5 C o g n itio n -0 .1 0 * -0 .0 6 -0 .0 6 D e p re s s io n (1 5 < ) -0 .0 5 -0 .1 1 “ -0.12b H e a lth B e h a v io r s B o d y M a ss In d e x 0 .1 4 b 0 .0 6 0 .0 6 S m o k in g S ta tu s 0 .0 2 0 .1 0 “ 0 .1 0 “ A lc o h o l C o n s u m p tio n -0 .0 2 -0 .0 2 -0 .0 2 P sy c h o lo g ic a l A ttr ib u te s M a s te ry -0 .0 2 -0 .1 5 b -0 .1 5 b E ffic a c y -0 .0 2 -0 .0 2 -0 .0 2 “ p < . 0 5 , b p < .0 1 . ^ p < .001 The logistic regression models predicting the over-reported dysfunction walking across a room and 'Toileting'” are shown in Table IV-15. Older respondents are more likely (odds ratio 1.25 . 95% confidence interval 1.06 to 1.49) to over-report their ability to walk across a room compared to younger individuals. In addition, compared to individuals who never fractured any bones, individuals with a new/recurrent fracture are more than 3 times as likely to over-report the onset o f dysfunction walking across a room (odds ratio 3.66. 95% confidence interval 1.09 to 12.30). High BMI is also predictive of over-reporting. Individuals with high BMI are significantly more likely to over-report their ability to walk across a room (odds ratio 2.90,95% confidence interval 1.04 - 8.13) compared to individuals with medium BMI levels. 151 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le I V - 15. L o g is tic r e g r e s s io n o f o v e r - r e p o r tin g o n s e t o f f u n c tio n a l lim ita tio n s ( 1 9 8 8 - 1 9 9 5 ) V a r ia b le O v e r -R e p o r tin g O n se t of L im ita tio n s W a lk in g O d d s R a tio (9 5 % C .l.) O v er-R ep o rtin g O n se t of L im ita tio n s T o ile tin g ( W ith G r ip S tre n g th ) O d d s R a tio (9 5 % C .l.) O v e r -R e p o r tin g O n se t of L im ita tio n s T o ile tin g ( W ith H a n d S ig n a tu r e ) O d d s R a tio (9 5 % C .l.) D e m o g r a p h ic C h a r a c te r istic s A g e 1.25(1.06- l.49)b 1 .0 5 ( 0 . 8 7 - 1 .2 6 ) 1 . 0 5 ( 0 . 8 8 - 1 .2 6 ) F e m a le G e n d e r 1.62 (0 .6 2 - 4 .2 5 ) 0 . 6 8 ( 0 . 2 5 - 1 .8 5 ) 0 . 6 9 ( 0 . 2 5 - 1 .8 7 ) H e a lth C h a r a c te r istic s B o n e F ra c tu re s N o n e ( r e f e r e n c e ) In 1 9 88 1 .4 8 ( 0 . 5 9 - 3 . 7 2 ) 1 .4 5 ( 0 .4 8 - 4 . 3 6 ) 1 . 4 2 ( 0 . 4 7 - 4 . 2 9 ) N e w /R e c u r re n t 3.66(1.09-12.30)* 4 .1 1 (1.18- 1 4 .3 4 )’ 4.17(1.19-14.56)* C o g n itio n 0 . 9 6 ( 0 . 9 2 - 1 .0 1 ) 0 . 9 7 ( 0 .9 1 - 1 .0 2 ) 0 .9 6 ( 0 .9 1 - 1 .0 2 ) D e p r e s s io n ( < 15 ) 1 .4 0 ( 0 .6 1 - 3 . 2 5 ) 2 . 1 7 ( 0 . 8 4 - 5 . 5 9 ) 2 .2 0 (0 .8 5 - 5 .6 9 ) H e a lth B e h a v io r B o d y M a s s In d e x ( k g / m 2) M e d iu m ( r e f e r e n c e ) L o w ( < 2 4 .2 7 ) 0 . 4 9 ( 0 .1 3 - 1 .9 2 ) 1.0 5 ( 0 . 2 9 - 3 . 7 5 ) 1.07 ( 0 . 3 0 - 3 . 8 2 ) H ig h ( > 2 7 .7 7 ) 2.90(1.04-8.13)* 1 .6 9 ( 0 .5 4 - 5 . 2 8 ) 1 .6 8 ( 0 . 5 4 - 5 . 2 3 ) S m o k in g S ta tu s N e v e r S m o k e ( r e f e r e n c e ) . . . C u r re n t S m o k e r 1 .3 9 (0 .4 3 - 4 . 5 3 ) 0 . 1 5 ( 0 . 0 2 1 .2 8 ) 0 .1 5 (0 .0 2 - 1 .2 6 ) P a s t S m o k e r 0 . 6 6 ( 0 . 2 5 - 1 .7 5 ) 0 . 4 6 ( 0 . 1 6 - 1 .3 0 ) 0 .4 7 ( 0 . 1 7 - 1 .3 5 ) P sy c h o lo g ic a l A ttr ib u te M a s te ry 1 .0 0 ( 0 .8 3 - 1 .2 2 ) 0 .7 7 (0 .6 3 - 0 .9 6 )* 0 .7 7 (0 .6 3 - 0 .9 7 )* N 4 0 8 4 2 8 4 2 7 M a x - R e s c a le d R 2 0 .1 9 0 .1 7 0 .1 7 _ Moles: 'C o n f id e n c e In te rv a l. * p < .0 5 , b p < .0 1 , c p < .001 L /i Significant predictors of over-reporting the onset o f dysfunction "toileting" include bone fractures, and the psychological attribute of mastery. Compared to individuals who never fractured any bones, individuals with a new/recurrent fracture are more than 4 times as likely to over-report the onset o f dysfunction "toileting" when compared with grip strength (odds ratio 4.11. 95% confidence interval 1.18 to 14.34) and hand signature (odds ratio 4.17,95% confidence interval 1.19 to 14.56). Furthermore, individuals with higher levels of mastery are significantly less likely to overestimate their ability to use the toilet - when compared with grip strength (odds ratio 0.77. 95% confidence interval 0.63 to 0.96) and when compared to signing one's name (odds ratio 0.77, 95% confidence interval 0.63 to 0.97), than those with lower levels o f mastery. D. Discussion This study has examined the relationship between change in reported and performed measures of physical functioning and their sources of discordance. O f the comparisons examined, only the one specific to the onset o f dysfunction walking across a room (compared with change in the ability to walk a ten foot course) showed a relatively high sensitivity (83%). This indicates that 83% o f participants who reported becoming unable were actually observed becoming unable to complete the walking course. For all other comparisons, sensitivity was low ranging from 14% (comparing the reported change in ability to "toilet" with the change in the hand grip strength) to 50% (comparing change in the reported ability to "toilet” with change in the ability to sign one's name). Compared to earlier studies that have examined similar relationships in the cross- section, these findings seem relatively low. For instance, using data from the 153 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. EPESE, Merrill et al. (1997) found that, when comparing the reported difficulty to stoop, crouch or kneel with the performed ability to complete five chair stands, the measures for both men (53%) and women (58%) had moderate rates of sensitivity. Here, while comparing these same exercises, the overall (for both men and women) sensitivity o f the measures is 30% both in the cross-section and over-time. This value is approximately half o f that found by Ferrer et al. (1999) when they compared the reported difficulty to stand up from a chair with the observed performance o f five chair stands (63%). The lower association found in the cross section is probably due to the high functioning nature o f the sample. Recall that the cross-sectional analyses presented here are based on the MacArthur data in 1995 because participants in the this study were originally selected based on their high physical and cognitive functioning in 1988. It was anticipated that by 1995, some physical decrement would have surfaced. In 1995. however, the majority of participants continued to be able to stoop, crouch or kneel (89.2%) and complete the five repeated chair stands (77.4%). The increased measurement error in change variables also contributes to the lower association found longitudinally. As hypothesized, this study found that individuals with certain characteristics tend to over-report the onset of reported dysfunction. Specifically, individuals who are older over-report the onset o f dysfunction walking across a room. This finding supports the study by Kempen, Van Heuvelen et al. (1996). They found that in the cross-section, older persons over-report their dysfunction on a summary measure o f motor functioning and on individual measures o f hearing and vision. This study also suggests that older Americans who have a new or recurrent bone fracture (hip or otherwise) are more likely to over-repot the onset o f dysfunction walking and '‘toileting.” Although this is the 154 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. first known study to examine the affect o f fractures on the discordance between measurement types, this finding suggests that these persons might benefit from rehabilitation programs. Such programs might improve their reported ability to continue walking and "toileting" after they have fractured specific joints. Furthermore, the finding that participants with high BMI over-report the onset of dysfunction walking supports the Social Security Administration’s recent decision to no longer permit individuals to qualify for dysfunction based on obesity, alone (Kosterski and Zwillich, 1999). The finding here suggests that older persons with high BMI continue being able to walk a relatively short distance despite the fact that they report becoming disabled. Unfortunately, it was not possible to examine whether persons with higher levels of psychological attributes under-reported decrement on measures o f physical functioning (as Kempen Van Heuvelen. et al.. 1996; and Kempen. Steverink et al.. 1996 found in the cross-section). Here, the findings suggest that older person with higher levels o f mastery more accurately report decrement on their ability to "toilet." Finally, this study did not find an association between certain characteristics (e.g.. gender, economic status, cognitive functioning, and depressive symptomatology) and the observed discrepancies between change in reported and performed functioning. These findings are. perhaps, attributable to the high functioning nature o f the sample. Participants in the MacArthur Study were originally selected on their high physical and cognitive functioning. Consequently it is suggested that the associations found among other studies (Kelly- Hayes et al., 1992; Kempen, Steverink et al.. 1996; Kempen Van Heuvelen et al., 1996; Merrill et al., 1997; Ferrer et al., 1999; and Kempen, Sullivan et al., 1999) between these characteristics (i.e., gender, economic status, cognitive functioning, and depressive 155 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. symptomatology) and observed discrepancies between reported and performed measures of physical functioning may not manifest themselves. There are a number o f important limitations that must be acknowledged with respect to the available data. First, the MacArthur Study was not originally designed to examine the concordance between self-reported and performance-based measures. Consequently, measures o f physical functioning were chosen on an a priori basis for the best possible comparison. For example, instead o f comparing the reported onset of dysfunction "toileting” with the performed onset o f dysfunction "toileting.” the reported onset o f dysfunction "toileting” was compared with the performed onset of dysfunction specific to three tests: 1) the repeated chair stands, 2) grip strength, and 3) signing one's name. These three comparisons were made in the effort to best capture all o f the physical mechanics o f "toileting.” Unfortunately, however, this strategy most likely captures individual components o f the reported activity instead of the complete activity. The ability to "toilet” most likely involves an array of physical movements some o f which are captured by the performance-based measures used here, but most likely by others as well (e.g.. balance, upper and lower body coordination, and strength). Furthermore, physical functioning requires the coordination o f numerous physiological systems including muscular, nervous, and vestibular systems. Thus, when we compare the reported ability o f a specific activity to a performed "component” o f the activity, we most likely arrive at an incomplete understanding o f the relationship between reported and performed measures o f physical functioning. As participants in the MacArthur study were selected based on their high physical functioning, sensitivity was low for most comparisons and a limited number o f 156 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. participants became "unable” to perform a specific activity. Thus, the findings that change in the self-reported measures o f crouching, stooping, or kneeling and ""toileting” are poor indicators o f physical decrement should be interpreted cautiously. In addition, due to a limited sample size, it was not possible to completely examine the factors responsible for the discrepancies (i.e.. under and over-report the onset o f functional limitations) between change in reported and performed measures. As a result, the factors found to influence the over-reported onset of dysfunction should be carefully evaluated. Researchers with access to a longitudinal, nationally representative sample would benefit from further exploring the inaccurate reporting o f dysfunction onset by incorporating comparable measures o f reported and observed measures o f physical functioning. Despite these limitations, these analyses provide needed information on how demographic, health, behavioral, and psychological characteristics might influence the disagreement between change in self-reported and performance-based measures of physical functioning. Individuals who rely on self-reported measures, for the assessment o f physical functioning, should pay particular attention to the behavioral characteristics of their subjects. Specifically, persons with higher BMI seem more likely to over-report the onset of reported dysfunction walking. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter V Sources o f Discrepancies Within Different Domains o f Self-Reported Measures o f Physical Functioning A. Introduction An alternative strategy to examining the discordance between reported and performed measures of physical functioning is to rely on summary measures o f physical functioning. Although some studies have used summary reported measures for this type o f analysis, exploring whether different determinants influence observed discrepancies among different domains of physical functioning is absent from the literature. This chapter focuses on the demographic, health, behavioral and psychological characteristics responsible for the observed discrepancies within three different domains o f physical functioning: 1) activities of daily living. ADL (Katz et al.. 1963). 2) gross motor functioning (Rosow and Breslau. 1963), and what Allaire et al. (1990) calls "'physical activity” (Nagi et al.. 1976). Using the MacArthur longitudinal study o f high functioning Americans 70-79 at baseline, determinants are identified in the cross-section and longitudinally. B. Discrepancies Within Summary Measures As detailed in Chapter II, researchers examining the discrepancies between reported and performed measures o f physical functioning have either relied on single item comparisons (Elam et al., 1991; Dorevitch et al., 1992; Kelly Hayes et al., 1992; Sager et al., 1992; Myers et al., 1993; Kempen, Steverink et al.. 1996; Kempen, Van Heuvelen et al., 1996; Merril et al., 1997; and Ferrer et al., 1999) or contrasted 158 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. summary measures from various sub-scales (Cress et al., 1995; Reuben et al., 1995; Kempen. Steverink et al.. 1996; Kempen, Van Heuvelen et al., 1996; Kempen, Sullivan et al.. 1999). While most o f these studies were conducted in the cross-section (Cress et al.. 1995; Reuben et al.. 1995; Kempen, Steverink et al., 1996; Kempen, Van Heuvelen et al.. 1996), the study by Kempen, Sullivan, et al., (1999) examined the relationship between reported and performed measures, o f physical functioning, over time. Presented here is a description o f the summary reported and performed measures used in these studies. Cress et al., (1995) used data from community residents 62-98 years o f age to compare reported scores from four subs-scales o f the Sickness Impact Profile (SIP) (physical dimension, body care and movement, ambulation and mobility) to performance- based measures o f gait and balance. In addition to gait and balance, grip strength was compared on the reported sub-scale o f ambulation. Individuals were asked to walk from 6 meters to 40 meters (distance was not standardized). The time required to complete the walking course was recorded with non-ambulatory participants assigned a score o f zero seconds. To assess balance, participants were asked to perform a parallel semi-tandem and a tandem stand (each for a maximum o f ten seconds). Participants who could not perform the parallel semi-tandem stand were assigned a score of zero. If a tandem stand was maintained for ten seconds, a score o f five seconds was assigned. Grip strength was assessed using a dynamometer and the score was recorded in kilograms. In 1995, Reuben et al. used data from 83 community residing residents to compared reported measures on: 1) the Functional Status Questionnaire (FSQ) for basic ADL and intermediate activities o f daily living (IADL) and 2) a component o f the Medical Outcomes Survey (MOS SF-36) that examines eight domains of 159 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. functioning and well-being: physical functioning, role limitations as a result o f physical health problems, role limitations as a result o f emotional problems, energy/fatigue, emotional well-being, social functioning, pain, and general health perceptions. Scores on all three reported scales (the FSQ-ADL, FSQ-IADL. and the MOS SF36) range from 0 (worst functioning) to 100 (best functioning). The Physical Performance Test (PPT) was used to collect information on seven activities. The seven performed items included were: 1) writing a sentence “Whales live in the blue ocean.": 2) simulating eating by transferring five kidney beans from an emesis basin to a can. one at a time: 3) lifting a book and putting it on a shelf above shoulder height: 4) putting on and removing a jacket; 5) picking up a penny from the floor: 6) turning 360 degrees; and 7) a fifty foot walk test. Each o f the seven items was scored on a five point scale that ranged from zero to four with zero indicating “unable to do" and four indicating the quickest time or best function. These scores were then summed to create a summary performance-based score with a range from zero to twenty-eight. Kempen. Steverink. et al., (1996) used data from the Groningen Longitudinal Aging Study (GLAS), at baseline (1993) to compare summary reported measures o f ADL to performance-based measures on three exercises. The 753 participants in their study were 57 and older, residing in the community of the northern Netherlands. From an eleven item ADL subscale o f the Groningen Activity Restriction Scale, reported ADL was assessed. Examples o f these items include: a) “Can you. fully independently, get around in the house?"; b) “Can you, fully independently stand up from sitting in a chair"; c) “Can you, fully independently, get on and off the toilet?"; and d) “Can you. fully independently, dress yourself?” Scores on the summary reported ADL measure 160 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ranged from 11 (no limitations) to 44 (maximum limitations). This reported measure was compared with three performance-based tests: a) putting on and taking off a jacket; b) walking 6 meters (including a 180 degree turn after 3 meters); and c) standing up and sitting down five times from a kitchen chair, without using one's arms. For each o f these performance-based measures, the score reflects the number o f seconds required for the performance o f the task. In the case o f putting on and taking off a jacket the score reflects the sum o f time required to complete each segment o f this activity. Individuals who were unable to perform an exercises were assigned a score equal to the worst actual score. Kempen. Van Heuvelen, et al.. (1996) compared an OECD indicator of "freedom of movement" (consisting o f three items: carrying 5 kg. picking up something from the floor, and walking 400 meters) with five performance-based tests from the Groningen Fitness Test for the Elderly among a community sample o f persons 57 and older. These five performance-based measures included: a) walking: b) flexibility o f the hip; 3) mobility o f the shoulders; 4) balance; and 5) grip strength. Walking was tested on a rectangular course where the speed was increased from 4km/h by lkm /h every 3 minutes. The walking score reflects the total distance walked. A ''sit and reach" test was used to assess the flexibility o f hip joint. Participants sat on the floor with their legs outstretched and bent forward to push a slide as far as possible with their fingertips. The shift length of the slide (from the best o f three tries) was recorded. Mobility o f the shoulder joints was assessed by asking participants to hold both handles o f a cord and pas the cord from the front o f the body, over their heads, to behind the body, keeping their arms straight and as close together as possible. The persons' arm length was combined with the shift 161 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o f the sliding handle to determine the best o f score o f three trials. Balance was assessed by standing on a platform that tilted from side to side. The total time that balance was maintained (the platform did not touch the floor) during a thirty second interval was recorded. The best o f three trials was used. Grip strength was assessed by a dynamometer. The final score was the best o f three trials. Longitudinally, Kempen, Sullivan et al., (1999) used data from a sub-sample of the Groningen Longitudinal Aging Study. In 1993, 574 participants were community residents, 57 years o f age and older. Follow-up data collection occurred in 1995. Self- reported measures o f physical functioning were assessed with an eleven item ADL subscale o f the Groningen Activity Restriction Scale. Scores on this summary reported measure ranged from 11 (no limitations ) to 44 (maximum limitations). This reported measure was compared with three performance-based tests: a) putting on and taking off a jacket; b) walking 6 meters (including a 180 degree turn after 3 meters); and c) standing up and sitting down five limes from a kitchen chair, without using one's arms. For each o f these performance-based measures, the score reflects the number o f seconds required for the performance o f the task. In the case o f putting on and taking o ff a jacket the score reflects the sum o f time required to complete each segment o f this activity. Individuals who were unable to perform an exercises were assigned a score equal to the worst actual score. Principle component analysis for these three performance-based measures showed one underlying dimension in both waves (1993 and 1995). Consequently a summary score was created for each wave using the sum-scores o f the three tests in each wave. Together, although various types o f reported summary measures have been used (ADL, FSQ-BADL, FSQ-IADL, MOS SF-36, and the OECD indicator o f “ 'freedom 162 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o f movement,'") these studies have relied on similar performance-based measures of physical functioning. These performance-based measures have been used in the analyses as: 1) individual independent variables predicting the reported summary score (Cress et al.. 1995; Kempen, Steverink et al., 1996; Kempen, Van Heuvelen et al., 1996); 2)combined into a summary performance-based score and correlated with summary reported measures (Reuben et al.,1995); or 3) combined into a summary performance- based measure to predict change in a summary reported based measure (Kempen. Sullivan et al.. 1999). Both, in the cross-section (Cress et al.. 1995; Reuben et al.. 1995; Kempen. Steverink et al., 1996; Kempen, Van Heuvelen et al., 1996) and longitudinally (Kempen. Sullivan et al.. 1999) these studies have found a moderate association between summary reported and performed measures o f physical functioning. As discussed in the previous chapter (Chapter II) existing studies have identified some discrepant sources between reported and performed measures o f physical functioning. In the cross-section, for individual measures o f comparison, these sources include: age. gender, economic status, health status, cognitive functioning, depressive symptomatology, and psychological attributes (Kelly-Hayes et al., 1992: Sager et al.. 1992; Kempen. Steverink et al.. 1996; Kempen, Van Heuvelen et al.. 1996; Merril et al., 1997; and Ferrer et al., 1999). Specific to summary measures. Cress et al. (1995) examined the effect of cognitive functioning and depressive symptomatology while controlling for physical performance (gait speed and balance). They found that among a sample o f nursing home residents 62 to 98, both depressive symptomatology and poorer cognitive functioning (as assessed on the Mini Mental Status Exam) independently predict poorer repored 163 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. functioning on three scales o f the SIP (i.e., body care and movement, ambulation, and overall physical dimensions). Kempen, Van Heuvelen et al.. (1996) found that among individuals 57 and older (from the Groningen Longitudinal Aging Study), those who are older, women, with depressive symptomatology, and highly educated over-reported levels on the OECD indicator of "freedom o f movement,'' while those with higher verbal memory (as assessed by the National Adult Reading Test) and mastery under-reported on the OECD, compared to summary measures o f gait, flexibility o f the hip/spine and shoulder, balance and grip strength. Within the domain o f ADL, Kempen. Steverink et al. (1996) also found that, among this sample of older persons, those with higher levels of depressive symptomatology and mastery over-reported functional limitation as compared to performance on three exercises (walking 6 meters - including a 180 degree turn at three meters, putting on and taking off a jacket and five repeated chair stands - without the use o f arms). In addition they found that participants with higher levels of self- efficacy also over-reported dysfunction within the domain o f ADL. Most recently. Kempen, Sullivan et al. (1999) used the same data to examine the effect of depressive symptomatology on the change in reported (within the domain o f ADL) and performed (walking 6 meters - including a 180 degree turn at three meters, putting on and taking off a jacket and five repeated chair stands - without the use of arms) measures o f physical functioning. They found that controlling for age. gender, and baseline number of chronic conditions, change in depressive symptomatology is predictive o f change within the domain o f ADL. Furthermore they found that the strength o f the association between change in reported and performed measures o f physical functioning depends on the change in depressive symptomatology. Together, these studies suggest that, within 164 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the domains o f different summary reported measures, specific characteristics differentially influence the concordance between reported and performed measures. To the author's knowledge, we have yet to clearly understand how specific factors influence discrepancies within different domains o f reported functioning. As reported measures are more commonly used than performance-based measures to assess the physical functioning o f older (and younger) persons, the information collected from such an investigation could be used to improve our selection of the best domain o f reported functioning suited for specific circumstances. Alternatively, if restricted to a specific domain o f reported physical functioning, this information could serve to guide professionals on how specific factors may confound their results. The aim o f this investigation is to explore how specific factors influence discrepancies within different domains o f reported physical functioning. Attention is placed on the demographic, health, behavioral, and psychological characteristics responsible for the discrepancies between summary reported and performed measures in the cross-section and over time. It is expected that the factors responsible for the observed discrepancies between summary reported and performed physical functioning will differ for each domain of reported functioning. C. Method Sample. As participants were selected for participation in the MacArthur Study based on their high physical and cognitive functioning at baseline, it was not possible to conduct cross-sectional analyses at baseline (in 1988) because these participants reported few, if any, problems with physical functioning. Instead, data from the follow-up 165 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. assessment, in 1995. was used to examine the relationship between reported and performed measures o f physical functioning, in the cross-section. Recall that of the original sample o f 1189 individuals in the MacArthur Study has been reduced to 722 because measures o f physical functioning were not made for participants from the Durham. North Carolina sub-sample. O f these 722 individuals. 173 (24.0%) died before the follow-up interview in 1995. Thus, cross-sectional analyses were restricted to a sample o f 549 men and women. O f those who survived, complete data on reported physical functioning at both baseline and follow-up (in 1995) were available for the following number o f individuals per domain o f reported measure o f physical functioning: 539 within the domain of ADL (Katz, et al.. 1963), 540 within the domain of gross mobility (Rosow and Breslau, 1966), and 537 within the domain of physical activity (Nagi, 1976). These same individuals represent the sample used for the longitudinal analyses. Dependent Variables. Three domains of self-reported physical functioning were examined: 1) ADL (Katz et al.. 1963). 2) gross mobility (Rosow and Breslau, 1966). and 3) physical activity (Nagi, 1975). Summary scores o f reported dysfunction were constructed across all three domains. In the MacArthur Study, the Katz et al. (1963) scale of activities of daily living was used to ask individuals if they were able to walk across a room, bathe, groom, dress, eat. transfer from bed to chair, and use the toilet (gong to the bathroom as well as cleaning and getting dressed afterwards). The dichotomously scored items (“ 1” unable/need help, “0” able) were summed to construct a continuous summary score ranging from zero to seven, reflecting the number o f ADL items individuals are 166 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. unable/need help to perform. Change in ADL was defined as the difference between the 1995 and 1988 summary score such that higher (more positive) scores reflect decrement in functioning. A continuous summary score ranging from zero to three is also constructed to describe the number o f items individuals self-report the inability to perform on the Rosow and Breslau (1966) scale o f gross functional mobility. Specifically, the individual items ask participants about their ability to perform heavy house work, walk up and down a flight o f stairs, and walk half a mile without help. These items were first dichotomously scored: unable (“ 1”) or able CO”) and then summed such that higher scores reflect poorer functioning. Change within the domain of gross mobility was defined as the difference between the 1995 and 1988 summary score so that higher values indicate decrement in functioning. The third self-reported measure o f physical functioning relies on the construction of a summary measure across five Nagi (1976) items of physical activity. Individuals were asked to estimate their level o f difficulty (none, a little, some, a lot or unable) pushing/pulling large objects, stooping/crouching/kneeling, carrying ten pounds or more, reaching or extending their arms above their shoulders, and writing/handling small objects. These items were dichotomously scored to indicate: unable (” 1”) or able (“0". responding none, a little, some, a lot) and then summed (with a range from zero to five) to quantify the number o f Nagi items individuals were unable to perform. As with the other two domains o f reported functioning, change within the domain o f physical activity was defined as the difference between the 1995 and 1988 summary score such that higher scores indicate decrement. 167 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hypothesized Independent Variables. Physical performance. A summary measure reflecting the performance on five distinct tasks was used here: balance, gait speed, foot tapping, repeated chair stands, and picking up a pencil and signing one’s name. Balance represents the sum of the scores (in seconds) on the tandem stand (eyes open) and single leg stand, ranging from zero to twenty seconds. The MacArthur Study also asked subjects to walk ten feet at a normal pace, turn and return at a fast pace. Measures were based on time (in seconds) to complete the course. Participants in the study were instructed to perform this activity without the use of any aids (e.g.. a cane), but were permitted to do so. if necessary. In order to examine coordination and leg strength, respondents were asked to complete two exercises: foot tapping and repeated chair stand, respectively. When seated, individuals were asked to alternate tapping a foot, between two. two inch circles placed one foot apart, as fast as possible (Nevitt et al., 1989). Up to ten taps were noted and the minimum time (up to thirty seconds) required to complete the exercise was documented. Leg strength was measured by the number of seconds individuals required to perform five chair stands without the use o f their arms (Nevitt et al., 1989). Fine motor coordination was assessed by asking respondents to pick up a pencil and sign their name and timing the number o f seconds (up to thirty) required to perform this activity (Jebsen et al.. 1969). The summary measure o f physical performance has been previously developed by Seeman et al. (1994) for the MacArthur Study. Seeman et al. (1994) first re-scaled each o f these five sub-scales to have a range from zero (worst performance) to one (best performance). These re-scaled items were weighted equally and summed so that 168 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the summary score had a range from zero to five. Here, this summary score was multiplied by - I so that higher scores reflect poorer performance. Furthermore, change for this summary variable was defined as the difference between the 1995 and 1988 summary scores multiplied by - I so that higher change scores reflect decrement in physical performance. Change scores range from -0.80 to 3.33 with a mean of 0.55 and a standard deviation o f 0.70. Demographic characteristics. Studies have shown that age, gender, and socioeconomic status influence physical functioning among older persons (Manton. 1988; Guralnik and Kaplan. 1989; Strawbridge et al., 1992; Crimmins and Saito, 1993; Guralnik et al.. 1993; Kaplan et al.. 1993; Boult et al.. 1994). For a more detailed discussion o f these studies, the author directs the reader to Chapter II. In the present analysis age is coded in years. In addition, dichotomous variables are used to indicate the following characteristics (reference group in parentheses): female (male), having less than 12 years o f education (at least 12 years o f education), having an annual income less than SI 0,000 (annual income o f at least $10,000). In this study, income and education serve as a proxies for economic status. In this study, as the MacArthur Study is a sub sample o f the Established Populations for Epidemiologic Studies (EPESE) study, income and education were dichotomized based on the findings from Guralnik et al. (1993). As discussed in Chapter II, Guralnik et al. (1993) used a longitudinal sample from the EPESE to assess the role of demographic factors in maintaining mobility in older persons. They found that persons with lower (< $5,000 per year) and middle ($5,000 to $9,999 per year) income levels were associated with significantly lower rates o f maintaining mobility (on reported measures o f walking up and down stairs and 169 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. walking half a mile) than persons with high (> $ 10,000 per year) income levels. In addition, they found that persons with low (< 8 years) and medium (9 to 12 years) levels o f education were also associated with lower rates o f maintaining mobility compared to those with more than 12 years of education. Health conditions and events. As detailed in Chapter II. studies have shown that specific disease conditions influence decrement in physical functioning (Satarino et al.. 1990; W olf et al., 1992; Guralnik et al.. 1993; Kurtz et al.. 1993; Ettinger et al.. 1994; Mulrow et al.. 1994; Ormel et al.. 1998; Kurtz et al.. 1999). Consequently, five prevalent health conditions and 2 health events are examined: whether the respondent had ever had cancer (excluding minor skin cancers), diabetes, or high blood pressure (average systolic and diastolic measurements after three seated readings greater than 140/90 mm Hg or taking blood pressure medication), myocardial infarction, or a stroke. The two health events considered are: "having fractured a hip" and "having fractured other bones" as dichotomous independent variables. Cognitive performance. In the MacArthur Study, the total cognitive score represented a summation of the performance on five sub-scales: 1) spatial memory. 2) similarity o f abstract concepts, 3) language, 4) delayed verbal memory, and 5) spatial orientation. Spatial memory was assessed by implementing the delayed Span Test, which requires the placement o f circular disks on a board in a specific order and then asking individuals to identify the new disk that has been added (Moss et al.. 1986). This test identifies individuals with neurological diseases. By scoring individuals from 0 to 17. this easily administered test focused on measuring spatial memory separate from verbal 170 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. related ability. The Wechsler Adult Intelligence Scale -Revised (WAIS-R) was used to evaluate abstract concepts (Wechsler. 1981). Participants were asked to explain how different objects or concepts (e.g., work and play) were similar and scored from 0 to 16 on this task. The Boston Naming Test tests language (Kaplan et al., 1983). Individuals were shown 18 drawings (e.g., broom, bench, tree. harp, etc.) and asked to name each object. Only items correctly named were scored within a time limit o f 10 seconds per picture. Relying on the Boston Naming Test (Kaplan et al.. 1983), delayed verbal memory was tested. Individuals were asked to recall the images shown during the naming test. Participants were challenged by this task as they were not asked explicitly to remember these images during the naming test. Finally, spatial orientation was tested through measurement o f subject's ability to copy geometric figures including a diamond, a diamond in a square, and a three-dimensional cube (Rosen et al., 1984). Participants were asked to draw these figures and a single rater scored drawings using a standardized system. A total cognitive score, representing the sum o f these five tests, was determined by the original investigators of the MacArthur Study and ranged from zero to eighty-nine such that lower scores indicate poorer cognitive functioning (Inouye et al.. 1993). At baseline, total cognitive score was normally distributed (see Figure III-1 in Chapter 3) with a mean score of 53.95 (standard deviation 9.54). Continuous scores o f cognitive functioning were used and change in cognitive functioning was defined as the difference between the 1995 and 1988 scores multiplied by -1 so that higher scores reflect declining cognitive functioning. Depression. The depression scale is adapted from the Hopkins Symptom Check List (Derogatis et al., 1974). Participants were assigned a score based on the sum 171 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of eleven items such that individuals with higher continuous scores (ranging from eleven to forty-four) have a higher depression score. Due to the skewed distribution o f this scale at baseline (see Figure III-2 in Chapter III), a dichotomous variable was created based on the highest tertile (> 15) to indicate levels o f depression (“ I" = high levels o f depression and "0” = moderate/low levels). Health behaviors. Three health behaviors were also examined: body composition, cigarette smoking, and alcohol consumption. Body mass index (kg/m2) captured the relative weight of each individual. Based on tertiles. a categorical variable describes individuals with low (< 24.27), medium (24.28 to 27.76). and high (> 27.77) BMI levels. Two dummy variables were used to indicate categories low BMI and high BMI with category medium BMI serving as the referent group. Smoking behavior is measured using two items that assessed the use and frequency of cigarette smoking. A categorical variable describes individuals who currently smoke, have smoked in the past, and have never smoked. Two dummy variables were used to indicate categories current smoker and past smoker with category never smoked serving as the referent. Finally, alcohol consumption is measured by four items that assessed the type and quantity of alcohol consumed. Again two dummy variables were used to indicate: < 1 gram ethyl alcohol per month and > 1 gram ethyl alcohol per month with category none in the past year serving as the reference group. Psychological Characteristics. Two psychological resources were measured: self-efficacy and mastery. Self-efficacy and mastery were operationalized by the scale developed by Rodin and McAvay (1993) and Pearlin and Schooler (1978). respectively. 172 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Individuals can score between nine and thirty-six on the self-efficacy scale and between seven and twenty-eight on the scale for mastery. Both o f these scales were normally distributed at baseline with higher scores indicating higher levels of these resources. Procedure. Multiple regression analysis was used to examine the extent of variance explained in summary self-reported measures o f physical functioning by a summary performance-based measure o f physical functioning. This technique has been used by other (Kempen. Steverink et al.. 1996; Kempen. Van Heuvelen et al.. 1996). In the cross-section, these analyses determined whether additional variance is explained by- demographic characteristics, cognitive functioning, health conditions or events, and health behaviors. Unfortunately, information on depressive symptomatology (as assessed by the Hopkins Symptom Check List) and psychological attributes (self-efficacy and mastery) were not collected at follow-up, in 1995. prohibiting the exploration o f the role o f these factors in the cross-section. In addition, although information on the maximum level o f education attained was not collected at follow-up, for the analyses in the cross- section. baseline information specific to education is used as a proxy. For the longitudinal analyses, the extent o f variance explained by change in summary self-reported measures by change in a summary performance-based measure of physical functioning is also achieved through multiple regression analyses. Additional determinants, examined here include levels o f psychological resources, levels of depressive symptomatology, and change in cognitive functioning. For both cross-sectional and longitudinal analyses, determinants were tested separately for their contribution (with all variables within a group entered simultaneously). 173 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In the cross-section, researchers (Kempen. Steverink et al., 1996; Kempen Van Heuvelen et al., 1996) have used a technique involving two different types o f analyses to determine whether individual determinants 1) affect the observed discrepancies between reported and performance-based measures in a ''consistent" manner (e.g., '‘consistently higher or lower levels o f self-reported compared to performance-based levels” Kempen. Steverink et al.. 1996. p. P257); and 2) are associated with the magnitude o f the discrepancies ("either positively or negatively indicating inaccuracy" Kempen. Steverink. et al., 1996. p. P257). In the first set of analyses. Kempen. Van Heuvelen et al.. (1996) obtain standardized regression coefficients for each o f the determinants while controlling for performance-based functioning and predicting self-reported measures. Only one predictor was included in each regression equation predicting reported functioning. For the present investigation, unstandardized coefficients are also provided for the reader. This strategy was pursued in order to isolate each predictor and reduce the confounding effects different predictors may have on each other. In the second part o f the analyses. Kempen, Steverink et al.. (1996) and Kempen. Van Heuvelen et al.. (1996) obtained absolute residual scores o f the regression analysis with the performance-based measure as the predictor o f the self-reported measurement. These residual scores were then correlated with each of the determinants to determine whether the determinants are associated with the magnitude o f the discrepancies between reported and performed measures. This investigation uses both of these techniques in the cross-section, and longitudinally, to examine the effect of specific determinants on the relationship between self-reported and performance-based measures o f physical functioning. In the 174 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cross-section these determinants include demographic characteristics, cognitive functioning, health conditions or events, and health behaviors. Longitudinally, the additional determinants examined are levels of psychological resources (e.g., mastery and self-efficacy), levels of depressive symptomatology, and change in cognitive functioning. D. Results Descriptive Characteristics of the Sample. Table V-l shows the physical functioning characteristics o f the sample at baseline and at follow-up in 1995. Although there is not a significant difference between baseline and follow-up levels on the summary measure of performance, participants do report decrement within the three domains examined. Demographic, health, behavioral, and psychological characteristics o f interest are presented in Table V-2. As one would expect, individuals who survive to 1995 are on average, significantly older (81 years o f age. standard deviation o f 3 years) compared to participants at baseline (74 years of age. standard deviation o f 3 years). In addition, a significantly higher proportion o f women survive to the final follow-up o f the study (59.1% in 1995 compared to 52.8% in 1988). A significantly higher proportion of participants in 1995 also have annual income levels of S10.000 or more (63.7%) compared to participants at baseline (56.5%). Education levels were only measured at baseline, with 48.0% having less than twelve years o f education. O f the health conditions and events considered, participants reported having had a significantly higher proportion o f cancer, high blood pressure, hip fractures, and other bone fractures in 1995 compared to 1988, see Table V-2. Differences specific to myocardial infarction and stroke 175 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T a b le V - 1 . D e s c rip tiv e s ta tis tic s o f p h y s ic a l fu n c tio n in g fo r th e stu d y 1 9 8 8 I 9 9 S P - V a r ia b le P e r c e n t (% ) n P e r c e n t (% ) n v a lu e P h y sic a l P e r fo r m a n c e S u m m a r y S c a le (m e a n , s ta n d a rd d e v ia tio n ) -2 .8 3 , 0 .4 8 7 1 4 -2 .8 5 , 0 .3 6 2 5 4 0 .3 4 7 S e lf-R e p o r te d S u m m a r y S c a le s K a tz Ite m s S u m U n a b le /N e e d H e lp 0 100.0 % 721 8 1 .8 % 441 0.001 I 0.0% 0 7 .8 % 4 2 2 0 .0% 0 3.5 % 19 3 0 .0% 0 1.9 % 10 4 0 .0% 0 0 .9 % 5 5 0 .0% 0 1.3 % 7 6 0 .0% 0 0 .9 % 5 7 0 .0% 0 1 .9 % 10 R o s o w -B r e s la u Ite m s S u m U n a b le 0 9 1 .0 % 6 5 3 5 5 .4 % 2 9 9 0.001 1 8 .9 % 6 4 20.0 % 108 2 0.1 % 1 1 3 .5 % 73 3 0.0 % 0 11. 1 % 6 0 N a g i Ite m s S u m U n a b le 0 96 .1 % 6 9 3 7 3 .2 % 393 0.001 1 3 . 9 % 2 8 1 1 .9 % 6 4 2 0 .0% 0 7 .6 % 41 3 0 .0% 0 4 .3 % 23 4 0 .0% 0 2.2 % 12 5 0 .0% 0 0 .7 % 4 C h a n g e S c o r e s P h y s ic a l P e rfo rm a n c e S u m m a ry S c a le (m e a n , s ta n d a rd d e v ia tio n ) 0 .5 9 .0 .7 0 4 2 0 S e lf -R e p o rte d M e a s u re s K a tz C h a n g e S c o re (1 9 8 8 - 1 9 9 5 ) 0 1 8 1 .8 % 7 .8 % 441 4 2 2 3 .5 % 19 J 1 .9 % 10 4 0 .9 % 5 5 1.3 % 7 6 0 .9 % 5 7 1 .9 % 10 R o s o w -B re s la u C h a n g e S c o re (1 9 8 8 -1 9 9 5 ) -1 1 .7 % 9 0 5 6 .5 % 3 0 5 1 20.0 % 108 2 12.2 % 66 3 9 .6 % 5 2 N a g i C h a n g e S c o re (1 9 8 8 -1 9 9 5 ) -1 2.2 % 12 0 71 .1 % 3 8 2 I 1 2 .7 % 68 -i 7.1 % 3 8 j 4.1 % 22 4 2.0 % 11 5 0 .7 % 4 176 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table V-2. Descriptive characteristics o f the study V a r ia b le 1 9 8 8 P e r c e n t (% ) n 1 9 9 5 P e r c e n t (% ) n P - v a lu e D e m o g r a p h ic C h a r a c te r istic s A g e (m e a n , s ta n d a rd d e v ia tio n ) 7 4 .1 9 , 2 .5 7 7 2 2 8 1 .1 1 .2 .5 7 4 8 9 0.000 G e n d e r F e m a le 5 2 .8 % 381 5 9 .1 % 3 0 0 0 .0 2 9 E d u c a tio n < 12 y e a rs 4 8 .0 % 3 4 6 — — — in c o m e < $ 10,000 4 3 .5 % 2 7 6 3 6 .3 % 186 0 .0 1 3 H e a lth C h a r a c te r istic s H e a lth C o n d itio n s /E v e n ts C a n c e r 2 0 .4 % 146 2 5 .9 % 142 0.021 D ia b e te s 1 4 .2 % 102 1 5 .7 % 86 0 .4 7 6 H ig h B lo o d P re ss u re 4 i .2 % 295 4 6 .6 % 2 5 6 0 .0 5 4 M y o c a rd ia l In fa rc tio n 13.1 % 94 1 6 .8 % 9 2 0 .0 6 9 S tro k e 2 .7 % 19 4 .6 % 2 5 0 .0 6 8 H ip F ra c tu re 2.1 % 15 4 .6 % 25 0 .0 1 3 O th e r B o n e F ra c tu re s 2 3 .0 % 165 2 8 .1 % 154 0 .0 4 1 C o g n itiv e F u n c tio n in g (m e a n . std. 5 3 .9 5 . 9 .5 4 721 4 7 .1 1 . 1 6 .00 4 9 0 0.000 d e v .1 ) D e p re s s io n (> 15) 3 6 .8 % 25 8 . . . — — H e a lth B e h a v io r s B M I L o w (< 2 4 .2 7 ) 3 2 .7 % 2 3 4 39 .1 % 190 0 .0 7 5 M e d iu m (2 4 .2 8 - 2 4 .7 6 ) 3 3 .4 % 2 3 9 2 9 .8 % 145 H ig h (> 2 7 .7 7 ) 3 3 .8 % 2 4 2 3 1 .1 % 151 S m o k in g S ta tu s N e v e r 4 3 .9 % 31 7 5 0 .8 % 2 5 7 0.001 C u rre n t 1 5 .8 % 114 7 .7 % 3 9 P a st 4 0 .3 % 291 4 1 .5 % 210 A lc o h o l C o n s u m p tio n N o n e in P a st Y e a r 3 7 .8 % 271 4 2 .3 % 2 1 4 0 .1 5 8 < 1 g ra m e th y l a lc o h o l/m o n th 2 3 .2 % 166 2 3 .9 % 121 > 1 g ra m e th y l a lc o h o l/m o n th 3 9 .0 % 2 7 9 3 3 .8 % 171 P sy c h o lo g ic a l A ttr ib u te s M a s te ry (m e a n , s ta n d a rd d e v ia tio n ) 1 8 .96 , 2 .3 5 715 — — — S e lf-E ffic a c y (m e a n s ta n d a rd d e v ia tio n ) 2 6 .2 9 .2 .5 1 7 1 6 . . . — . . . — ------- i ■ ■ - I — — . Notes: S ta n d a rd d e v ia tio n . approached significance with 16.8% of participants reporting having had a myocardial infarction in 1995 (compared to 13.1% in 1988) and 4.6% reporting having had a stroke in 1995 (compared to 2.7% in 1988). There were no significant differences specific to having had diabetes: 14% in 1988 and 16% in 1995. In terms o f cognitive performance, participants in 1995 had, on average, a significantly lower cognitive score (mean o f 47, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. standard devotion o f 16) than those in 1988 (mean o f 54, standard deviation o f 10) indicating that those who survived to 1995 had higher cognitive functioning than the sample at baseline. Furthermore, over a third o f the sample (37%) had higher levels of depression on the Hopkins Symptom Check List, at baseline. Based on tertiles levels of baseline BMI, there were no significant differences in BMI between two samples. In addition, there were no significant differences based on alcohol consumption. For both years o f the study, over a third of the samples reported not having had any alcohol in the past year and consuming at least 1 gram of alcohol in the past year. Less than a quarter percent o f the participants (in 1988 and 1995) reported consuming less than a gram of alcohol in the past year. There were, however, significant differences between 1988 to 1995 in terms o f smoking behavior. Although roughly the same proportion reported having smoked in the past (40% in 1988 and 42% in 1995). a higher proportion reported having never smoked in 1995 (51% in 1995 and 44% in 1988) and a smaller proportion reported currently smoking in 1995 (8% in 1995 and 16% in 1988). Finally, individuals had relatively high levels of mastery and self-efficacy, at baseline. On average, participants had a score o f 19 (range of 7 to 28) on mastery and a score of 26 (range o f 9 to 36) on self-efficacy. 1. Cross-Sectional In the cross-section in 1995, Tables V-3 to V-5 show the mean and standard deviation o f the summary performance-based measure by the number of items participants are unable to perform for each reported scale (ADL, gross motor, and 178 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le V -3 . M e a n a n d s ta n d a r d d e v ia tio n s c o re o f p h y s ic a l p e r f o r m a n c e b y th e n u m b e r o f A D L ite m s in d iv id u a ls re p o r t b e in g u n a b le /n e e d h e lp w ith in 19 95_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __________ 0 1 2 3 4 5 6 7 M e a n (std . d ev .1 ) n M e a n (std . d ev .) n M e a n (std . d e v .) n M e a n (std . d ev .) n M e a n (std . d ev .) n M e a n (std . d e v .) n M e a n (std . d ev .) n M e a n (std . d e v .) n P h y s ic a l P e rfo m ia n c e -2 .8 5 ( 0 .3 6 ) 2 4 5 -2 .8 5 ( 0 .3 3 ) 8 . . . . . . -2 .3 8 ( . . . ) 1 . . . . . . _ _ _ _ . . . . . . ___ ___ Notes: S ta n d a rd d e v ia tio n T a b le V -4 . M e a n a n d s ta n d a r d d e v ia tio n s c o re o f p h y s ic a l p e r f o r m a n c e b y th e n u m b e r o f g r o s s m o to r ite m s in d iv id u a ls r e p o r t b e in g u n a b le w ith in 19 9 5 0 1 2 3 M e a n (std . d ev .1 ) n M e a n (std . d ev .) n M e a n (std . d e v .) n M e a n (std . d ev .) n P h y s ic a l P e rfo rm a n c e -2 .9 0 ( 0 .3 7 ) 1 8 6 -2 .7 2 ( 0 .2 9 ) 4 7 -2 .6 9 ( 0 .3 1 ) 18 -2 .5 9 (0 .2 8 ) 3 Notes'. 1 S ta n d a rd d e v ia tio n T a b le V -5 . M e a n a n d s ta n d a r d d e v ia tio n s c o re o f p h y s ic a l p e r f o r m a n c e b y th e n u m b e r o f p h y s ic a l ite m s in d iv id u a ls r e p o r t b e in g u n a b le w ith in 19 9 5 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 0 1 2 3 4 5 M e a n (std . d ev .1 ) n M e a n (std . d ev .) n M e a n (std . d e v .) n M e a n (std . d ev .) n M e a n (std . d e v .) n M e a n (std . d ev .) n P h y s ic a l P e rfo rm a n c e -2 .8 7 ( 0 .3 7 ) 2 2 6 -2 .7 3 ( 0 .2 6 ) 21 -2 .7 9 ( 0 .4 0 ) 5 -3 .0 1 ( - ) 1 . . . . . . . . . . . . Notes'. 1 S ta n d a r d D e v ia tio n o physical activity). On a scale of-3.82 to -2.29. individuals who report dysfunction on a fewer number o f items have lower (more negative) scores indicating better performance. The variance explained in self-reported physical functioning by each separate groups o f determinants is shown in Table V-6. As the MacArthur Study did not measure levels o f depression or psychological resources in 1995. it was not possible to examine the role o f these determinants in cross-section. An association between reported and performed functioning is found only specific to the domain o f gross motor functioning. The discrepancies within this domain are affected by demographic, cognitive, health, and behavioral characteristics. Surprisingly, no association was found between reported and performed functioning specific to the domains o f ADL or physical activity. This finding may be attributable to the high functioning nature o f the participants. As shown in Table V -l, a higher percentage o f individuals report no limitations on the ADL (81.8%) and physical activity scales (73.2%) in 1995 than on the Rosow-Breslau scale o f gross mobility (55.4%). Thus, the absence o f a significant association between reported and performance-based measures within the domains o f ADL and physical activity may be due to the reduced variance on these scales. T a b le V -6 . P e rc e n ta g e o f e x p la in e d v a r ia n c e in s e lf-re p o rt b y p e r fo rm a n c e -b a s e d m e a s u re s o f p h y s ic a l fu n c tio n in g a n d e x p la in e d v a ria n c e a d d e d b y g ro u p s o f d e m o g ra p h ic , c o g n itiv e , h e a lth , a n d b e h a v io ra l m e a s u re s in 1995 A d ju ste d R 2 S u m m a r y A D L D y sfu n c tio n (n = 5 3 9 ) S u m m a r y G r o ss-M o to r D y sfu n c tio n (n = 5 4 0 ) Summary P h y s ic a l A c tiv ity D y sfu n c tio n (n = 5 3 7 ) P e rfo rm a n c e -B a s e d P h y sic a l F u n c tio n in g 0.0 5 .4 * * * 0.1 D e m o g ra p h ic C h a ra c te ris tic s 1 .6 * 5 .0 * * * 3 .9 * * * C o g n itiv e F u n c tio n in g 4 ^ * * * 3 .0 * * * 1 .3 * * H e a lth C o n d itio n s a n d E v e n ts 7 .0 * * * H 9 * * * 8 .9 * * * H e a lth B e h a v io rs 1.4 5 .5 * * * 5 .8 * * * * F v a lu e o r F c h a n g e . p < 0 .0 5 ; ** F v a lu e o r F c h a n g e . p < 0 .0 1 ; * * * F v a lu e o r F c h a n g e , p < 0 .0 0 1 . 180 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table V-7 presents the unstandardized and standardized regression coefficients for each individual determinant with the self-report measure as the outcome variable and controlling for physical performance (only one predictor was included in each regression equation). Within the domain of activities of daily living, none of the determinants are significant. Participants with annual income levels less than $10,000 report better functioning on the summary score o f dysfunction specific to gross mobility compared to those with higher levels o f annual income. Specific to the domain o f physical activity, individuals who have not had any alcohol in the last year report poorer functioning within the domain o f physical activity compared to persons who have had moderate amounts o f alcohol consumption. Table V-8 shows the correlation coefficients between the absolute residuals of summary performance on summary self-report with the determinants of discrepancies. Here, the coefficients reflect the influence o f demographic, health and behavioral characteristics on the magnitude o f the discrepancies between reported and performed measure o f physical functioning. Discrepancies between reported and performed measures within the domain o f ADL are smaller for individuals with high BMI compared to those with moderate BMI. In addition, discrepancies are smaller within the domain o f gross mobility for older persons with annual income levels below $10,000 per year. The discrepancies within the domain o f physical activity is affected by alcohol consumption. While discrepancies are larger for those who have not had any alcohol within the past year, discrepancies are smaller for those who drink heavily (> 1 gram ethyl alcohol/month), as compared to those who drink moderately. 181 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le V -7 . U n s ta n d a r d iz e d a n d s ta n d a r d iz e d re g re s s io n c o e f f ic ie n ts a n d s ta n d a r d e r r o r s fo r in d iv id u a l d e m o g ra p h ic , c o g n itiv e , h e a lth , a n d b e h a v io r a l c h a r a c te r is tic s w ith s u m m a ry s e lf - r e p o r t le v e ls o f d y s fu n c tio n a s th e o u tc o m e a n d c o n tr o llin g f o r p e r f o r m a n c e - b a s e d le v e ls in 1 9 95_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ S u m m a r y A D L D y sfu n c tio n (n = 5 3 9 ) S u m m a r y G r o ss- M o to r D y sfu n c tio n (n = 5 4 0 ) S u m m a r y P h y sic a l A c tiv ity D y sfu n c tio n (n = 5 7 3 ) V a r ia b le s B 1 B J (S E 3 ) B 1 B J (S E 3 ) B 1 B 1 (S E 3 ) D e m o g r a p h ic C h a r a c te r istic s A g e 0.00 0 .0 3 ( 0 .0 1 ) 0.01 0 .0 4 (0 .0 2 ) 0.00 0 .0 0 (0.01 ) F e m a le G e n d e r 0.02 0 .0 3 ( 0 .0 3 ) -0.11 -0 .0 8 ( 0 .0 8 ) 0 .0 9 0 .1 0 ( 0 .0 5 ) E d u c a tio n < 12 y e a rs 0 .0 4 0 .0 8 ( 0 .0 3 ) 0 .0 9 0 .0 7 (0 .0 8 ) 0 .0 6 0 .0 7 (0 .0 5 ) In c o m e < $ 1 0 ,0 0 0 -0 .0 3 - 0 .0 6 ( 0 .0 4 ) -0 .2 1 ' -0 .1 4 (0 .1 0 )* 0 .0 8 0 .0 9 (0 .0 6 ) H e a lth C h a r a c te r istic s C a n c e r -0 .0 5 -0 .0 8 ( 0 .0 4 ) 0 .0 5 0 .0 4 (0 .0 9 ) 0.11 0 .1 2 ( 0 .0 6 ) D ia b e te s 0 .0 3 0 .0 3 ( 0 .0 5 ) - 0 .0 7 - 0 .0 4 ( 0 .1 3 ) -0.00 -0 .0 0 (0 .0 9 ) H ig h B lo o d P r e s s u r e 0.01 0 .0 2 ( 0 .0 3 ) 0 .0 9 0 .0 7 (0 .0 8 ) 0 .0 4 0 .0 5 ( 0 .0 5 ) M y o c a r d ia l In fa r c tio n -0 .0 3 -0 .0 4 ( 0 .0 4 ) 0 .1 9 0.11 (0 . 10) 0 .0 6 0 .0 6 (0 .0 7 ) S tro k e -0 .0 4 -0 .0 3 ( 0 .0 8 ) 0 .1 5 0 . 0 4 ( 0 .2 1 ) -0 .0 3 - 0 .0 2 ( 0 .1 4 ) H ip F r a c tu re -0 .0 5 -0 .0 3 ( 0 .1 0 ) 0 .0 3 0 .0 1 (0 .2 5 ) 0.00 0.00 (0 .1 6 ) O th e r B o n e F r a c tu re s -0.02 -0 .0 3 ( 0 .0 4 ) 0 .0 4 0 .0 2 (0 .0 9 ) -0.01 -0.01 (0 .0 6 ) C o g n itiv e Im p a ir m e n t 0.00 0 .0 9 ( 0 .0 0 ) 0.00 0.02 (0 .00 ) -0.00 -0.01 (0 .00 ) H e a lth B e h a v io r s B M I M e d iu m ( 2 4 .2 8 - 2 4 .7 6 ) L o w (< 2 4 .2 7 ) -0 .0 3 -0 .0 5 ( 0 .0 4 ) 0.01 0.00 (0 . 10) 0 .0 6 0 .0 6 (0 .0 7 ) H ig h (> 2 7 .7 7 ) 0 .0 7 0 .1 3 ( 0 . 0 4 ) -0 .0 4 -0 .0 3 ( 0 .1 0 ) 0 .0 4 0 .0 4 (0 .0 7 ) S m o k in g S ta tu s N e v e r . . . C u r re n t -0 .0 6 -0 .0 7 ( 0 .0 6 ) 0.10 0 .0 5 ( 0 .1 4 ) -0.01 -0.01 (0 . 10) P a st -0 .0 4 -0 .0 8 ( 0 .0 4 ) 0 .0 7 0 .0 6 ( 0 .0 9 ) 0 .0 4 0 .0 5 (0 .0 6 ) A lc o h o l C o n s u m p tio n < 1 g r a m e th y l a lc o h o l/m o n th . . . . . . . . . . . . _ ___ N o n e in P a s t Y e a r -0.02 -0 .0 5 ( 0 .0 4 ) -0.00 -0 .0 0 (0 . 11) 0 .1 5 “ 0.16(0 .0 7 )' > 1 g r a m e th y l a lc o h o l/m o n th 0.01 0 .0 2 ( 0 .0 4 ) -0.12 -0 .0 9 (0 .1 0 ) -0 .0 5 -0 .0 6 (0 .0 7 ) Notes: 'U n s t a n d a r d i z e d c o e f f i c i e n t .: S ta n d a r d iz e d c o e f f ic ie n t. 5 S ta n d a r d E rro r. *p< 0 5 , bp < O I, c p<.OI T a b le V -8. C o r re la tio n c o e ffic ie n ts b e tw e e n a b s o lu te re s id u a ls (o b ta in e d w ith re g re s s io n a n a ly s is o f s e lf-re p o rt a n d p e r fo rm a n c e b a s e d m e a s u re s ) a n d d e m o g ra p h ic , c o g n itiv e , h e a lth , a n d b e h a v io ra l c h a ra c te ris tic s in 19 9 5 S u m m a r y S u m m a r y S u m m a r y A D L G r o ss-M o to r P h y sic a l A c tiv ity D y sfu n c tio n D y sfu n c tio n D y sfu n c tio n V a r ia b le s "(n = 5 3 9 ) (n = 5 4 0 ) (n = 5 3 7 ) D e m o g r a p h ic C h a r a c te r istic s A g e 0 .0 3 0 .0 4 0.00 F e m a le G e n d e r 0 .0 3 -0 .0 8 0.10 E d u c a tio n < 12 y e a rs 0 .0 8 0 .0 7 0 .0 7 In c o m e < S I 0 ,0 0 0 -0 .0 5 -0 .1 4 “ 0 .0 8 H e a lth C h a r a c te r istic s H e a lth C o n d itio n s /E v e n ts C a n c e r -0 .0 8 0 .0 4 0.12 D ia b e te s 0 .0 3 -0 .0 4 -0.00 H ig h B lo o d P re s s u re 0.02 0 .0 7 0 .0 5 M y o c a rd ia l In fa rc tio n -0 .0 4 0.12 0 .0 6 S tro k e -0 .0 3 0 .0 4 -0.02 H ip F ra c tu re -0 .0 3 0.01 0.01 O th e r B o n e F ra c tu re s -0 .0 3 0.0 3 -0.01 C o g n itiv e Im p a irm e n t 0 .0 8 0.02 -0.01 H e a lth B e h a v io r s B M I M e d iu m (2 4 .2 8 - 2 4 .7 6 ) — — — L o w (< 2 4 .2 7 ) -0.12 0 .0 4 0 .0 5 H ig h (> 2 7 .7 7 ) -0 .1 6 b 0.02 0 .0 4 S m o k in g S ta tu s N e v e r — — — C u rre n t -0 .0 5 0 .0 3 -0.02 P a st -0 .0 6 0 .0 4 0 .0 5 A lc o h o l C o n s u m p tio n < 1 g ra m e th y l a lc o h o l/m o n th — — — N o n e in P a st Y e a r -0 .0 6 0 .0 5 0 .2 0 b > 1 g ra m e th y l a lc o h o l/m o n th 0 .0 4 -0 .0 9 -0 .1 6 b Notes: ap < .0 5 . bp < .0 1. cp< .01 2. Longitudinal Tables V-9 to V-l I show the mean and standard deviation o f the change in summary performance-based measure by the change in the number o f items participants are unable to perform for each reported scale (ADL. gross motor, and physical activity). On a scale o f -0.80 to 3.33 individuals who report dysfunction on a fewer number of items have lower scores indicating better performance. 183 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. T a b le V -9 . M e a n a n d s ta n d a r d d e v ia tio n s c o re o f c h a n g e in p h y s ic a l p e r f o r m a n c e b y th e n u m b e r o f A D L ite m s in d iv id u a ls re p o rt b e c o m in g u n a b le /n e e d h e lp w ith fr o m 1 9 8 8 to 1 9 95_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 0 I 2 3 4 5 6 7 M e a n ( s id . d e v . 1) n M e a n ( s t d . d e v .) n M e a n ( s t d . d e v .) n M e a n ( s t d . d e v .) n M e a n ( s td . d e v .) n M e a n ( s t d . d e v .) n M e a n ( s td . d e v .) n M e a n ( s td . d e v .) n P h y s ic a l P e rfo rm a n c e 0 .2 9 (0 .5 0 ) 361 0 .9 4 ( 0 .7 7 ) 2 7 1.84 (0 .6 6 ) 10 1.72 (0 .6 5 ) 7 1.5 6 ( 1 1 0 ) 4 2 .5 2 ( 0 .6 5 ) 4 2 .4 8 (0 .4 4 ) 3 2.31 ( 0 .7 1 ) 4 Noles: S ta n d a r d d e v ia tio n T a b le V -IO . M e a n a n d s ta n d a r d d e v ia tio n s c o re o f c h a n g e in p h y s ic a l p e r f o r m a n c e b y th e n u m b e r o f g r o s s m o to r ite m s in d iv id u a ls re p o r t b e c o m in g u n a b le w ith fro m 1 9 8 8 to 1995 -1 0 1 2 3 M e a n ( s td . d e v . 1) n M e a n ( s t d . d e v .) n M e a n ( s t d . d e v .) n M e a n ( s t d . d e v .) n M e a n ( s td . d e v .) n P h y s ic a l P e r f o r m a n c e 0.20 ( 0 .3 7 ) 8 0 .3 2 ( 0 .4 6 ) 2 5 0 0 .6 3 ( 0 .6 0 ) 86 0 .9 9 (0 .86 ) 4 4 1.58 (0 .9 8 ) 3 2 Notes'. S ta n d a r d d e v ia tio n T a b le V - l I. M e a n a n d s ta n d a rd d e v ia tio n s c o re o f c h a n g e in p h y s ic a l p e r f o r m a n c e b y th e n u m b e r o f p h y s ic a l a c tiv ity ite m s in d iv id u a ls re p o r t b e c o m in g u n a b le w ith fro m 1 9 8 8 to 1 9 9 5 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -1 0 1 2 3 4 5 M e a n ( s td . d e v .1) n M e a n ( s td . d e v .) n M e a n ( s t d . d e v .) n M e a n ( s td . d e v .) n M e a n ( s td . d e v .) n M e a n ( s t d . d e v .) n M e a n ( s td . d e v .) n P h y s ic a l P e r f o r m a n c e 0 .3 6 (0 .4 0 ) I I 0 .3 8 ( 0 .5 1 ) 311 0 .6 9 ( 0 .7 6 ) 4 9 1 .3 0 ( 0 .7 6 ) 2 4 1.62 ( 1.0 2 ) 17 2.10 ( 0 .6 4 ) 7 — Notes'. S ta n d a r d d e v ia tio n O © Table V-12 shows the variance explained in change within each domain o f self- reported physical functioning by each group of determinants, separately. Across all three types o f reported functioning, the longitudinal association between reported and performed measures o f physical functioning is moderate. A stronger association exists within the domain o f activities of daily living than within the domains o f gross motor functioning or physical activity. The discrepancies within the domain of ADL are affected by change in cognitive functioning, and baseline levels of 1) demographic characteristics, 2) depressive symptomatology, and 3) psychological attributes. Baseline psychological attributes also influence the discrepancies within the domain on gross motor functioning. In contrast, to the domain o f ADL. the discrepancies within the domain o f gross motor functioning are also influenced by baseline cognitive functioning and health behaviors. Baseline health behaviors are also responsible for the discrepancies seen within the domain o f physical activity, as are change in cognition functioning, and demographic characteristics. T a b le V -1 2 . P e rc e n ta g e o f e x p la in e d v a ria n c e o f c h a n g e in s e lf -r e p o r t b y c h a n g e in p e rfo rm a n c e -b a s e d m e a s u re s o f p h y s ic a l fu n c tio n in g a n d e x p la in e d v a ria n c e a d d e d b y c h a n g e in c o g n itiv e fu n c tio n in g an d b a s e lin e m e a s u re s o f d e m o g ra p h ic , c o g n itiv e , d e p re s s iv e , h e a lth , b e h a v io r a l, a n d p s y c h o lo g ic a l c h a ra c te ris tic s fro m 19 88 to 1995 S u m m a r y S u m m a r y S u m m a r y A D L G r o ss-M o to r P h y sic a l A c tiv ity D y sfu n c tio n D y sfu n c tio n D y sfu n c tio n A d ju ste d R 2 (n = 5 3 9 ) (n = 5 4 0 ) (n = 5 3 7 ) C h a n g e in : P h y s ic a l P e rfo rm a n c e 3 6 .4 * * * 2 7 .0 * * * 2 8 .2 * * * C o g n itiv e F u n c tio n in g 3 .1 * * * 0 .8 * 1 .2 * * B a se lin e : D e m o g ra p h ic C h a ra c te r is tic s 1 .9 * * 4 .1 * * * 4 .0 * * * C o g n itiv e F u n c tio n in g 0 .9 | g*** -0.0 D e p re s s iv e S y m p to m a to lo g y 1 .0 * * 0.2 0.3 H e a lth C o n d itio n s a n d E v e n ts -0 .5 0.6 0 .7 H e a lth B e h a v io rs 0.1 2 .0 * * 2 .4 * * P s y c h o lo g ic a l A ttrib u te s 2 |* * * 1 .2 * * .0-7 _______ * F v a lu e o r F c h a n g e . p < 0 .0 5 ; ** F v a lu e o r F c h a n g e . p < 0 .0 1 ; * * * F v a lu e o r F c h a n g e , p < 0 .0 0 1 . 185 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The unstandardized and standardized regression coefficients for each individual determinant with the change in self-report measure as the outcome variable and controlling for change in physical performance are shown in Table V-13. Here only one predictor was included in each regression equation (while controlling for change in physical performance). Participants with less than a high school education report declining physical functioning within the domain of gross mobility, compared to older participants with at least a high school education. In addition, women and individuals with an annual income o f less than $10,000 report declining physical functioning within the domain o f physical activity. Specific disease conditions are only predictive of declining physical functioning within the domain of physical activity. While individuals with cancer, at baseline, reported declining physical functioning within this domain, those with high blood pressure report improved physical functioning, compared to individuals without these conditions. Although persons with higher baseline cognitive functioning report improved functioning within the domain of gross mobility, individuals with declining cognitive ability report declining physical functioning within the domain of ADL. Individuals with higher levels o f depression also report declining physical functioning within the domain o f ADL. In contrast, although baseline levels of depression do not predict change within the domains of gross mobility or physical activity, health behaviors influence change in reporting within these domains. A high BMI is predictive o f decrement within the domains of gross mobility and physical activity, compared to individuals with moderate BMI. Compared to those who have never smoked, participants who currently smoke report decrement within the domains of gross motor functioning. Furthermore, compared to individuals who consume a 186 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Table V-13. Unstandardized, standardized regression coefficients and standard errors for individual characteristics with change in sum m ary self-report levels o f dysfunction as the outcom e and controlling for change in perform ance-based levels V a r ia b le s S u m m a r y A D L D y sfu n c tio n (n = 5 3 9 ) S u m m a r y G r o ss-M o to r D y sfu n c tio n (n = 5 4 0 ) S u m m a r y P h y sic a l A c tiv ity D y sfu n c tio n (n = 5 7 3 ) D e m o g r a p h ic C h a r a c te r istic s B ' B J (S E 3 ) B 1 B J (S E 3 ) B 1 B 2 (S E 3 ) Age 0.00 0.01 (0.02) 0.01 0.02 (0.02) 0.00 0.01 (0.02) Female G ender -0.10 -0.04 (0.09) 0.04 0.02 (0.08) 0 .1 7 " 0 .0 9 (0 .0 8 )* Education < 12 years 0.11 0.05 (0.09) 0 .2 2 b 0 .1 1 (0 .0 8 )b 0.12 0.07 (0.08) Income < $10,000 0.06 0.02(0.10) 0.12 0.06 (0.09) 0 .2 2 b 0 .1 2 (0 .0 8 )b H e a lth C h a r a c te r istic s Cancer 0.11 0.04(0.11) 0.19 0.08 (0.10) 0 .2 5 b 0 .1 1 (0 .0 9 )b Diabetes -0.09 -0.02 (0.14) -0.00 -0.00 (0.13) -0.06 -0.02 (0.12) High Blood Pressure -0.14 -0.06 (0.09) -0.06 -0.03 (0.08) -0 .2 2 b -0 .1 2 (0 .0 8 )b M yocardial Infarction 0.04 0.01 (0.14) 0.04 0.01 (0.12) -0.03 -0.01 (0.12) Stroke -0.25 -0.03 (0.35) 0.07 0.01 (0.32) -0.28 -0.04 (0.30) Hip Fracture -0.04 -0.01 (0.28) 0.24 0.04 (0.25) -0.26 -0.05 (0.24) O ther Bone Fractures -0.02 -0.01 (0.11) 0.08 0.03 (0.10) -0.06 -0.03 (0.09) Cognitive Impairment -0.00 -0.01 (0.01) -0 .0 1 " -0 .0 9 (0 .0 0 )* 0.00 0.01 (0.00) Change/Cognitive Impairment 0 .0 2 c 0 .1 3 (0.0I)C -0.00 -0.02 (0.00) 0.00 0.04 (0.00) Depressive Symptomatology 0 .2 2 " 0 .0 9 (0 .0 9 )" 0.12 0.06 (0.08) 0.03 0.02 (0.08) H e a lth B e h a v io r s BMI M edium (24.28 24.76) Low (< 24.27) -0.07 -0.03 (0.11) -0.01 -0.01 (0.10) 0.05 0.03 (0.09) High (> 27.77) 0.15 0.06(0.11) 0 .2 1 " 0.10(0.10)' 0 .1 9 " 0 .1 0 (0 .0 9 )* Sm oking Status Never Current -0.10 -0.03(0.14) 0 .2 4 " 0 .0 9 (0 .1 2 )" 0.19 0.07 (0.12) Past -0.07 -0.03 (0.10) 0.10 0.05 (0.09) 0.03 0.01 (0.08) Alcohol Consumption < 1 gram ethyl alcohol/m onth . . . . . . . . . . . . . . . . . . N one in Past Year -0.00 -0.00 (0.12) 0.17 0.08 (0.11) 0 .2 5 b 0 .1 3 (0.l0)b > 1 gram ethyl alcohol/m onth 0.14 -0.06 (0.12) -0.03 -0.01 (0.11) -0.06 -0.03 (0.10) P sy c h o lo g ic a l A ttr ib u te s M astery -0 .0 6 b -0 .1 2 (0 .0 2 )b -0.03 -0.06 (0.02) -0.02 -0.05 (0.02) Self-Efficacy 0.02 0.05 (0.02) -0 .0 3 * -0 .0 8 (0 .0 2 )* -0.00 0.00 (0.01) Notes: 'Unslandardized co efficien t.2 Standardized co efficien t.1 Standard Error. ap < 0 5 , bp < 0 1 , cp<.01 moderate amount o f alcohol, participants who have not had any alcohol, within the past year, report declining physical functioning within the domain o f physical activity. Furthermore those with higher levels o f mastery report improved physical functioning within the domain o f ADL while those with higher levels o f self-efficacy report improved functioning within the domain o f gross mobility. Neither o f these psychological resources influence change within the domain o f physical activity. Table V-14 presents the correlation coefficients between the absolute residuals of change in performance on change in self-report with the determinants o f discrepancies. Here, the coefficients reflect the influence o f demographic, health, behavioral, and psychological characteristics on the magnitude of the discrepancies between change in reported and performed measures of physical functioning. Although none o f the demographic characteristics influence the discrepancies within the domain o f ADL over time, discrepancies within the domain o f gross motor functioning are higher for those with less than 12 years o f education. In addition, the discrepancies with the domain of physical activity are higher for women and those with annual income levels below $ 10.000. The discrepancies w ithin the domain of physical activity are also affected by specific diseases. Although the discrepancy is larger for those who have had cancer, the discrepancy in smaller for those who have had high blood pressure. While baseline cognitive functioning reduces the discrepancy within the domain o f gross motor functioning, change in cognitive functioning increases the discrepancy within the domain o f ADL as does higher levels o f depressive symptomatology. Interestingly . across all three domains o f physical functioning, discrepancies are higher for individuals with high BMI. In addition, discrepancies are higher within the domain o f gross mobility and 188 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table V-14. Correlation coefficients between absolute residuals (obtained with regression analysis o f change self-report and change in performance based measures) and demographic, cognitive, depressive, health, behavioral, and psychological characteristics from 1988-1995_______________________________ S u m m a r y S u m m a r y S u m m a r y A D L G r o ss-M o to r P h y sic a l A c tiv ity D y sfu n c tio n D y sfu n c tio n D y sfu n c tio n V a r ia b le (n=539) (n=540) (n=537) D e m o g r a p h ic C h a r a c te r istic s Age 0.01 0.02 0.01 Female Gender -0.05 u.03 o . i r Education < 12 years 0.06 0.13b 0.08 Income < S I0.000 0.03 0.07 0.13b H e a lth C h a r a c te r istic s Health Conditions/Events Cancer 0.05 0.09 O.I3b Diabetes -0.03 - 0.01 -0.02 High Blood Pressure -0.07 -0.04 -0.14b Myocardial Infarction 0.01 0.02 - 0.01 Stroke -0.03 0.01 -0.05 Hip Fracture - 0.01 0.05 -0.05 Other Bone Fractures - 0.01 0.04 -0.03 Cognitive Functioning -0.02 - 0. 10* 0.02 Change Cognitive Functioning 0.14b -0.03 0.04 Depressive Symptomatology 0.12" 0.07 0.02 H e a lth B e h a v io r s BMI Medium (24.28 - 24.76) — — — Low (< 24.27) -0.07 0.07 0.02 High (> 27.77) 0. 10" 0.12b 0.10’ Smoking Status Never — — — Current 0.03 0.08 0.08 Past 0.03 0.02 -0.01 Alcohol Consumption < 1 gram ethyl alcohol/month — — — None in Past Year -0.05 0. 11’ o . i r > 1 gram ethyl alcohol/month 0.07 -0.08 -0.13b P sy c h o lo g ic a l A ttr ib u te s Mastery -0.14b 0.07 -0.06 Self-Efficacy 0.07 -0.10“ -0.00 Notes: “p<.05, bp< .01, cp< 0 1 physical activity for individuals who have not consumed any alcohol within the past year. Finally, while higher levels o f mastery seem to reduce the discrepancy within the domain o f ADL. higher levels o f self-efficacy reduce the discrepancy within the domain o f gross motor functioning. 189 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. E. Discussion An understanding of what factors contribute to the observed discrepancies within different domains o f reported physical functioning and performance-based measures is important if we seek to improve our assessment o f physical functioning. This research has focused on identifying the impact o f selected determinants on the discrepancies between self-reported and performance-based measures, across three domains o f reported functioning: ADL. gross motor functioning, and physical activity. In contrast to other cross-sectional studies (Elam et al.. 1991; Dorevitch et al.. 1992; Kelly Hayes et al., 1992; Sager et al.. 1992; Myers et al.. 1993; Cress et al., 1995; Reuben et al.. 1995: Kempen, Steverink et al., 1996; Kempen. Van Heuvelen et al.. 1996: Merril et al., 1997; and Ferrer et al., 1999), the results here suggest that there is a relatively weak association between self-reported and performance-based measures of physical functioning. As mentioned, the lack o f an association between reported and performed measures within the domain o f ADL and physical activity may be attributable to the relatively high functioning nature o f the sample. Participants in the MacArthur Study were originally selected based on their high physical and cognitive functioning. Seven years after the baseline selection o f these participants, over fifty percent o f the sample continue to report no functional limitation within each o f the domains examined: 81.8% ADL, 73.2% physical activity, and 55.4% gross mobility. Consequently, as speculated earlier, the author suggests that lack o f an association (in the cross-section) between reported and performed measures within the domain o f ADL and physical activity may be due to the lack of variability within these domains o f reported functioning. Longitudinally, however, a moderate association is found, which 190 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. supports the studies by Myers et al. (1993), Mendes de Leon et al. (1996), and Kempen. Sullivan, et al. (1999). This investigation shows that the determinants examined differentially influence discrepancies among the three domains both in the cross-section and longitudinally. In the cross-section, reasonable contributions from demographic, cognitive, health, and behavioral characteristics are made to reported levels o f functioning within all three domains. This study, supports the findings from Kempen. Steverink et al., 1996, that age, gender, and education do not influence the observed discrepancies within the domain of ADL. Although they found that depressive symptomatology and psychological attributes influence the observed discrepancy within the domain of ADL. as these characteristics were not measured at follow-up in the MacArthur Study, it was not possible to provide supportive (or contradictory) evidence to their findings. Instead, this study explored the influence o f specific disease conditions or events and three different health behaviors. The domain o f physical activity is particularly affected by health conditions or events and health behaviors. While heavy drinkers under-report dysfunction within the domain of physical activity, those who have not had any alcohol in the past year over-report dysfunction. Longitudinally, baseline demographic characteristics contribute to change in reported levels o f functioning within ADL. gross motor, and physical activity domains. Discrepancies within the domain o f ADL and gross motor functioning are particularly affected by psychological attributes, while those within the domain o f gross motor functioning and physical activity are influenced by health behaviors. While individuals with higher levels o f mastery under-report decrement within the domain o f ADL. 191 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. those with higher levels o f self-efficacy under-report decrement within the domain of gross motor functioning, as compared to performance. In addition, within the domains of gross motor functioning and physical activity, those who have not had any alcohol within the past year over-report decrement in physical functioning, as do those with higher BMI. Furthermore, change in cognitive functioning and levels o f depressive symptomatology influence discrepancies within the domain o f ADL. alone. Those with declining cognitive functioning and higher levels of depressive symptomatology over-report decrement within the domain of ADL. These findings contradict the cross-sectional findings o f Rozzini et al. (1993) and Kempen, Steverink et al. (1996). but are in line with the result o f Cress et al. (1995). This study is limited in its use of composite summary measures o f reported and performed physical functioning. As Wolinsky et al. (2000) has recently discussed, the use o f these aggregate measures, in longitudinal analyses, prohibits the determination of what individual item(s) participants have declined. In addition, the reported and performed measures used here are not the same. Despite these limitations, the performance-based measures can be considered as representative indicators o f physical functioning within the three different domains examined. As hypothesized, determinants differentially influence the observed discrepancies within different domains of reported functioning. When applying measures within the domain o f ADL to assess physical decrement, evaluators should pay particular attention to participant’s cognitive functioning, depressive symptomatology. BMI. and psychological attributes because these characteristics are likely to influence this type o f reported measure. Although BMI and psychological attributes are also likely to 192 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. influence reported measures within the domain o f gross motor functioning, so too is alcohol consumption. Finally, when applying measures from the domain o f physical activity, gender, economic status. BMI. and alcohol consumption are likely to confound one's results. These results suggest that overall, reported measures are more strongly confounded by psychological attributes and health behaviors than performance-based measures. Although self-reported measures are easier to administer, the effects o f self- efficacy, mastery, BMI, and alcohol consumption should be considered in the application o f reported measures among older persons. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter VI The Effect of Change in Cognitive Performance on Change in Performance Based Measures o f Physical Functioning A. Introduction Research on the relationship between cognitive impairment and physical functioning is scarce. Furthermore, we have yet to examine whether change in cognitive performance is related to decrement on more physically demanding or novel tasks. This investigation focuses on the cognitive, health (conditions/events and behaviors), and demographic characteristics predictive of change on twelve physical performance exercises. Using this longitudinal study o f high functioning Americans 70-79 at baseline, risk factors for older persons declining on specific types o f physical performance activities are identified. As the present study focuses on the relationship between the cognitive impairment and physical functioning among older persons, physical performance based tests, that provide an objective assessment o f physical functioning, are chosen instead o f the commonly used self-reported measures of ADL because reported measures o f physical functioning may yield inaccurate results among older persons with poor cognitive functioning (Guralnik. Branch et al., 1989). The twelve performance based tests investigated here include: walking at a fast pace, walking at a normal pace, turning in a circle, repeated chair stands, foot tapping, balance, tandem stand-with eyes open, tandem stand-with eyes closed, tandem walk, standing on a single leg, signing one’s name, and grip strength. 194 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. As presented in Chapter II, although existing studies have shown an association between cognitive impairment and physical functioning (Lawton and Brody, 1969; Wilson et al., 1973; Kuriansky and Gurland, 1976; Pfeffer et al.. 1982; Teresi et al., 1984; Winograd, 1984; Fillenbaum. 1985; Scherr et al., 1988; Jagger et al.. 1989; Barberger- Gateau et al.. 1992; Rozzini et al.. 1993; Galanos. 1994; Seeman et al.. 1994; Cress et al.. 1995; Binder et al., 1999), it remains unclear whether cognitive functioning might disproportionately influence more physically challenging or novel activities. Research suggests that performance on many information-processing tasks assessing cognitive or perceptual process is slowed with age (Cerella. 1985). Salthouse (1994) suggests that this slower processing speed explains lower scores on measures of working memory with age. Kaufman and colleagues (1989) suggest that aging seems to decrease fluid intelligence (e.g., tasks requiring sensorimotor coordination, new learning, and speedy performance— Cattell. 1943) while preserving crystallized intelligence (e.g.. tasks requiring language skills and the use o f established habits— Cattell. 1943). Furthermore. Albert and Kaplan (1980) suggest that decreased fluid intelligence is related to deficits in functions o f initiative, flexibility and planning. Based on these findings, Barberger-Gateau and Fabrigoule (1997) proposed a classification scheme for the physical performance o f older persons: a loss in efficiency for attentional demanding or novel tasks; and good performance on more automatic or over-leamed skills. The present investigation tests this classification. Researchers have already identified some factors that may be responsible for the observed association between cognitive and physical functioning. These factors include 195 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. age, education (Pinsky, 1987; Victor, 1989; Barberger-Gateau et al., 1992; Dartigues et al.. 1992), chronic conditions (Forette and Boiler, 1991; Dey. 1997; Guo et al.. 1997; Palombo et al., 1997; Perros et al., 1997;), depression (Winograd 1984; Kay et al.. 1985; Berkman et al., 1986; Barberger-Gateau et al.. 1992; Fuhrer et al.. 1992; Lichtenberg et al.. 1995; Alexopoulous et al., 1996;) and health behaviors (Abalan et al.. 1984; Guralnik et al.. 1989b; Colsher and Wallace, 1990; Graves et al., 1991; Letenneur et al.. 1993; Letenneur et al., 1994). For a more detailed discussion of these studies the author directs the reader to Chapter II. This investigation advances knowledge concerning the characteristics o f older people in the United States who are at greater risk of decrement on specific physical tasks. Focus is placed on the cognitive, health, behavioral, and demographic risk factors predictive o f decrement on twelve specific physical exercises. While it is expected that individuals with declining cognitive abilities will manifest decrement on more physically demanding or novel tasks, it is also expected that cognitive decline will not impact more routine physical activities. The following research question is asked: How does change in cognitive performance relate to change on an array of physical exercises when baseline physical performance, cognitive functioning, demographic, health, and behavioral characteristics are controlled for? B. Method Sample. From the original sample of 1189 individuals in the MacArthur Study, longitudinal information on physical functioning was available for 722 (participants from the Durham, North Carolina sub-sample were not measured and thereby omitted). 196 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. O f these 722 individuals. 173 (24.0%) died before follow-up interview in 1995. This reduced the analytical sample to 549 individuals. Although the MacArthur Study selected persons based on their high physical and cognitive functioning, there were a handful o f individuals who were “unable” to perform specific physical exercises, at baseline (1988). As a component o f these analyses investigates what characteristics predict becoming “unable” to perform some o f these exercises, the sample was then restricted to those who were able to perform these exercises, at baseline. The sample size for each performed measure o f physical functioning thereby ranges from 434 (walking at a fast pace) to 549 (grip strength), see Table VI-1. T a b le V I - 1. S a m p le siz e fo r e a c h p h y s ic a l p e rfo rm a n c e e x e rc is e _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ P h y s ic a l P e r f o r m a n c e T a s k T o t a l S a m p le S iz e W a lk a t a F a st P a c e 4 3 4 W a lk a t a N o rm a l P a c e 4 4 5 T u rn in a C irc le 4 4 7 R e p e a te d C h a ir S ta n d s 4 5 2 F o o t T a p p in g 4 5 4 B a la n c e 5 2 8 T a n d e m S ta n d . E y e s O p e n 5 3 8 T a n d e m S ta n d . E y e s C lo s e d 541 T a n d e m W a lk 541 S in g le L e g S ta n d 5 4 4 H a n d S ig n a tu re 5 4 8 G rip S tre n g th 5 4 9 Dependent Variables. The MacArthur Study included twelve performance based measures o f physical functioning. Physical functioning of the upper body was determined by measures o f grip strength and fine motor coordination. By employing a dynamometer, maximum hand grip strength (in kilograms) was documented in the dominant arm, based on the average o f the best three measures (Nevitt et al., 1989). Fine motor coordination was assessed by asking respondents to pick up pencils and sign their names and timing 197 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the number o f seconds (up to thirty) required to perform this activity (Jebsen et al., 1969). Assuming a constant name length, respondents with good coordination were able to complete this task in less time than those with poor coordination. A large number o f performance based tests measured lower body functioning (balance, gait, coordination, and strength) in the MacArthur Study. First, Nevitt et al. (1989) detailed four tests of balance that were used: 1) tandem stand, eyes open (up to 10 seconds); 2) tandem stand, eyes closed (up to 10 seconds); 3) tandem walk (up to 10 steps); and 4) single leg stand (up to 10 seconds). In addition, a summary score for balance was developed by the original investigators o f the MacArthur Study (Seeman et al.. 1994). This summary balance measure represents the sum o f the scores (in seconds) on the tandem stand (eyes open) and single leg stand, ranging from zero to twenty seconds. Higher scores for all five o f these measures represented better balance. Previous researchers o f the MacArthur Study reported good reliability o f the summary balance measure (two month test, re-test reliability coefficient o f 0.61) (Seeman et al., 1994). Turning balance (360 degrees) was also assessed by asking participants to turn around in a circle (Tinetti, 1986) and timing the number of seconds required. The performance based measures o f gait in the MacArthur Study asked subjects to walk ten feet at a normal pace, turn and return (Nevitt et al.. 1989). They were then asked to complete the same walk at a fast pace. Measures were based on time (in seconds) to complete the course. Participants in the study were instructed to perform this activity without the use of any aids (e.g., a cane), but were permitted to do so, if necessary. 198 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In order to examine coordination and leg strength, respondents were asked to complete two exercises: foot tapping and repeated chair stand, respectively. When seated, individuals were asked to alternate tapping a foot, between two. two inch circles placed one foot apart, as fast as possible (Nevitt et al., 1989). Up to ten taps were noted and the minimum time (up to thirty seconds) required to complete the exercise was documented. Leg strength was measured by the number of seconds individuals required to perform five chair stands without the use o f their arms (Nevitt et al., 1989). Respondents were first asked to stand from a seated position without the use o f their arms. If respondents were able to complete this activity, they were asked to repeat it five consecutive times. Change for all twelve o f these measures is defined as the difference between the 1995 and 1988 continuous score. Thus, two broad sub-categories are observed: 1) measures for which decrement is based on a reduction in average value (e.g., balance, single leg stand, tandem stand/eyes open, tandem stand/eyes closed, tandem walk, and grip strength); and 2) measures for which decrement is based on an increase in average value (e.g.. chair stands, hand signature, right/left foot tap. turning in a circle, walking at a fast pace and walking at a normal pace). Independent Variables. Cognitive performance. In the MacArthur Study, the total cognitive score represented a summation o f the performance on five sub-scales: 1) spatial memory. 2) similarity o f abstract concepts, 3) language, 4) delayed verbal memory, and 5) spatial orientation. Spatial memory was assessed by implementing the delayed Span Test, which requires the placement o f circular disks on a board in a specific order and then 199 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. asking individuals to identify the new disk that has been added (Moss et al.. 1986). This test identifies individuals with neurological diseases. By scoring individuals from 0 to 17, this easily administered test focused on measuring spatial memory separate from verbal related ability. The Wechsler Adult Intelligence Scale -Revised (WAIS-R) was used to evaluate abstract concepts (Wechsler, 1981). Participants were asked to explain how different objects or concepts (e.g., work and play) were similar and scored from 0 to 16 on this task. The Boston Naming Test tests language (Kaplan et al.. 1983). Individuals were shown 18 drawings (e.g.. broom, bench, tree. harp, etc.) and asked to name each object. Only items correctly named were scored within a time limit o f 10 seconds per picture. Relying on the Boston Naming Test (Kaplan et al.. 1983), delayed verbal memory was tested. Individuals were asked to recall the images shown during the naming test. Participants were challenged by this task as they were not asked explicitly to remember these images during the naming test. Finally, spatial orientation was tested through measurement o f subject's ability to copy geometric figures including a diamond, a diamond in a square, and a three-dimensional cube (Rosen et al., 1984). Participants were asked to draw these figures and a single rater scored drawings using a standardized system. A total cognitive score, representing the sum of these five tests, was determined by the original investigators o f the MacArthur Study and ranged from zero to eighty-nine with higher scores indicative of better functioning (Inouye et al., 1993). Preliminary analyses indicated that this summary cognitive score is normally distributed, at baseline. At baseline, total cognitive score was normally distributed (see Figure III-1 in Chapter 3) with a mean score o f 53.95 (standard deviation 9.54). Consequently, change in cognitive performance was defined as the difference between the 1995 and 1988 continuous 200 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. score multiplied by -1 , so that higher (or more positive) change scores reflect declining cognitive ability. Control Variables. In order to identify personal and health characteristics that predispose older persons to change in physical performance, a series o f variables that represent demographic, health, and behavioral characteristics are included. These variables also serve as controls for factors that are confounded with cognitive perform ance-such as age (Letenneur et al.. 1994. and Letenneur et al.. 1993 et al.), education (Pinsky, 1987; Barberger-Gateau et al., 1992), chronic conditions (Perros et al.. 1997: Dev et al.. 1997; Guo et al.. 1997; Palomobo et al.. 1997). depression (Kay et al., 1985: Fuhrer et al., 1992; Alexopoulous et al., 1996; Lichtenberg et al., 1995), and health behaviors (Abalan. 1984; Cosher and Wallace 1990; and Letenneur et al.. 1994) which may exert independent influence on change in physical performance. Demographic characteristics. Studies have shown that age. gender, and socioeconomic status influence physical functioning among older persons (Manton. 1988; Guralnik and Kaplan. 1989; Strawbridge et al.. 1992; Crimmins and Saito. 1993; Guralnik et al.. 1993; Kaplan et al.. 1993: Boult et al., 1994). For a more detailed discussion o f these studies, the author directs the reader to Chapter II. In the present analysis age is coded in years. In addition, dichotomous variables are used to indicate the following characteristics (reference group in parentheses): female (male). African American (Caucasian), having less than 12 years of education (at least 12 years o f education), having an annual income less than $10,000 (annual income of at least $10,000). In this study, as the MacArthur Study is a sub-sample o f the Established Populations for Epidemiologic Studies (EPESE) study, income and education were 201 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dichotomized based on the findings from Guralnik et al. (1993). As discussed in Chapter II. Guralnik et al. (1993) used a longitudinal sample from the EPESE to assess the role of demographic factors in maintaining mobility in older persons. They found that persons with lower (< $5,000 per year) and middle ($5,000 to $9,999 per year) income levels were associated with significantly lower rates o f maintaining mobility (on reported measures o f walking up and down stairs and walking half a mile) than persons with high (> $10,000 per year) income levels. In addition, they found that persons with low (< 8 years) and medium (9 to 12 years) levels o f education were also associated with lower rates o f maintaining mobility compared to those with more than 12 years o f education. Health conditions and events. Earlier studies have found that diabetes (Dey. 1997; Perros et al.. 1997) and hypertension (Lindholm. 1990; Guo et al.. 1997; Palombo et al., 1997) pose particular threats to the cognitive well-being o f older adults. In addition to these conditions, the potential influence of other health conditions and events are considered. As detailed in Chapter II. studies have shown that specific disease conditions influence decrement in physical functioning (Satarino et al.. 1990; W olf et al.. 1992; Guralnik et al.. 1993; Kurtz et al., 1993; Ettinger et al., 1994; Mulrow et al.. 1994; Ormel et al., 1998; Kurtz et al., 1999). Consequently, five prevalent health conditions examined are: whether the respondent had ever had cancer (excluding minor skin cancers), diabetes, or high blood pressure (average systolic and diastolic measurements after three seated readings greater than 140/90 mm Hg or taking blood pressure medication), myocardial infarction, or a stroke. These chronic conditions are each treated as dichotomous independent variables, with their absence coded as "0" and their 202 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. occurrence coded as "1." The two health events considered are: "having fractured a hip" and "having fractured other bones" as dichotomous independent variables. Depression. The depression scale is adapted from the Hopkins Symptom Check List (Derogatis et al.. 1974). Participants were assigned a score based on the sum of eleven items such that individuals with higher continuous scores (ranging from eleven to forty-four) have a higher depression score. Due to the skewed distribution o f this scale at baseline (see Figure III-2 in Chapter III), a dichotomous variable was created based on the highest tertile (> 15) to indicate levels of depression ("1" = high levels o f depression and "0" = moderate/low levels). Health behaviors. Three health behaviors were also examined: body composition, cigarette smoking, and alcohol consumption. Body mass index (kg/m: ) captured the relative weight of each individual. Based on tertiles. a categorical variable describes individuals with low (< 24.27), medium (24.28 to 27.76). and high (> 27.77) BMI levels. Two dummy variables were used to indicate categories low BMI and high BMI with category medium BMI serving as the referent group. Smoking behavior is measured using two items that assessed the use and frequency o f cigarette smoking. A categorical variable describes individuals who currently smoke, have smoked in the past, and have never smoked. Two dummy variables were used to indicate categories current smoker and past smoker with category never smoked serving as the referent. Finally, alcohol consumption is measured by four items that assessed the type and quantity of alcohol consumed. Again two dummy variables were used to indicate: < 1 gram ethyl Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. alcohol per month and > 1 gram ethyl alcohol per month with category none in the past year serving as the reference group. Procedure. Multivariate analyses are used to predict the change in the continuous variables o f the twelve physical performance measures considered. These analyses determine the effect o f change in cognitive performance on the change in physical performance, after controlling for baseline: 1 )physical performance o f the exercise predicted, 2) cognitive functioning; 3) demographic characteristics; 4) health conditions and events; and 5) health behaviors. As mentioned earlier, the sample was restricted to persons who were "able"’ to complete these exercises at baseline (1988). Table VI-2 shows the percentage o f persons who became “unable” and remained “able” to perform each of the twelve physical performance exercises from 1988 to 1995. T a b le V l-2 . P e rc e n ta g e o f p e rs o n s w h o b e c o m e " u n a b le " to p e rfo rm s p e c ific ta s k s . ^ -•--..•.-•a .-,-. — P h y sic a l P e r fo r m a n c e T a sk B e c a m e U n a b le % (n ) R e m a in e d A b le % (n ) B a la n c e 1 2 .7 % (6 7 ) 8 7 .3 % (4 6 1 ) S in g le L e g S ta n d 1 4 .5 % (7 9 ) 8 5 .5 % (4 6 5 ) T a n d e m S ta n d , E y e s O p e n 12.6% (68) 8 7 .4 % (4 7 0 ) T a n d e m S ta n d , E y e s C lo s e d 1 9 .2 % (1 0 4 ) 8 0 .8 % (4 3 7 ) T a n d e m W a lk 0 .0% (0 ) 1 0 0 % (5 4 1 ) G rip S tre n g th 0 .2% ( 1) 9 9 .8 % (5 4 8 ) R e p e a te d C h a ir S ta n d s 1 4 .8 % (6 7 ) 8 5 .2 % (3 8 8 ) H a n d S ig n a tu re 0 .7 % (4 ) 9 9 .3 % (5 4 4 ) F o o t T a p p in g 3 .1 % (1 4 ) 9 6 .9 % (4 4 0 ) T u rn in a C irc le 6 .0 % (2 7 ) 9 4 .0 % (4 2 0 ) W a lk a t a N o rm a l P a c e 3 .8 % (1 7 ) 9 6 .2 % (4 2 8 ) W a lk a t a F a st P a ce 5 .1 % (2 2 ) 9 4 .9 % (4 1 2 ) In the statistical literature, data which includes information (independent variables) on observations where information on the dependent variable is unknown is referred to as “right” or “left” censored data. According to Allison (1995, p.9), data is “right” censored, when “ ...all you know about (a variable) is that it is greater than some 204 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. value.” Similarly, the data is described as "left” censored when all we know about a variable is that it is less than some value. In this investigation, change scores on the various performance-based measures are based on censored data. For instance, we know that participants who are "unable” to perform a specific exercise (i.e.. completing the five repeated chair stands: turning in a circle; foot tapping; walking at a normal pace; walking at a fast pace: and signing one's name) would take a longer period o f time to complete the task, if they could, than those who are "able.” The data on these variables are thereby called "right" censored data. Furthermore, we know that participants who are "unable" to perform other exercises (i.e.. grip strength, tandem stand-eyes open, tandem stand - eyes closed, and tandem walk, balance, and single leg stand) would take a shorter period of time (or walk fewer steps, or have a weaker grip strength) to complete the exercise, if they could, than those who are "able.” The data on these variables are called "left” censored data. In order to incorporate the observations of those "unable" to perform each of the exercises, the MacArthur Study assigned values for both the "right” and "left” censored data. For the "right” censored variables (i.e., completing the five repeated chair stands, turning in a circle, foot tapping, walking at a normal pace, walking at a fast pace, and signing one's name), the MacArthur Study assigned those "unable” to perform a specific exercise, an extreme value to indicate the worst possible performance. For five o f the six "right” censored variables, this extreme value is equal to 25% beyond the poorest score achieved among those who were " ‘able” to complete the task (i.e., completing the five repeated chair stands, turning in a circle, walking at a normal pace, walking at a fast pace and signing one’s name). As the foot tapping exercise (the sixth "right” censored 2 0 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. variable) was terminated after 30 seconds, individuals were assigned a maximum score of 30 seconds for this exercises if they were "unable” to perform this task. For the "left” censored variables (i.e.. grip strength, tandem stand-eyes open, tandem stand-eyes closed, and tandem walk, and single leg stand), persons "unable" to perform the exercise were assigned a value o f "0” in kilograms or seconds (depending on the type o f measurement). Finally, the balance item is neither "left” nor "right” censored because it was calculated based on the sum o f the valid scores (i.e.. those who were "able” to perform) on the tandem stand-eyes open and single leg stand exercises. The procedural question for this investigation is how to use the sample data to estimate the relationship between change in physical performance and the explanatory variables o f interest? For five o f the models investigated (e.g.. predicting change in chair stands, foot tapping, turning in a circle, walking at a normal pace, and walking at a fast pace), preliminary analyses using OLS regression, yielded relative unstable equations (evident from highly skewed residuals). Recall that these five dependent variables are based on "right” censored data (i.e., individuals who are "unable” to perform have a score higher than those who are "able”). To determine how to analyze this data, four models were tested and compared against the original OLS regression model where change scores are based on scores where persons who are "unable” to perform are assigned a score o f 25% above the highest score among those who are "able.” For consistency, although individuals who are "unable” to tap their feet are assigned the maximum score of 30 seconds, a score 25% above the highest valid score (29.88 seconds) is used instead for these comparisons. First, instead o f assigning those who are "unable” a score o f 25% above the highest 206 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. score, these individuals are assigned a score o f 3% above the highest score to determine whether the change scores would yield a more stable model. Table VI-3 shows the values assigned at 25% and 3% above the poorest score achieved (in 1995) for the five performance measures examined. Then, in order to examine the characteristics o f those who continued to be "able” to perform an exercise, those who became "unable” were set to missing. A fourth model uses logistic regression to examine the characteristics of persons who became "unable” versus those who "remained able." For the logistic regression models, coefficients from these equations predict the log odds o f becoming "unable” which are exponentiated and presented as odds ratios. The final model uses a Tobit regression to analyze the sample data. This technique is used when information is available (independent variables) on the observations where the dependent variable is censored (Breen. 1996). T a b le V l-3 . S u m m a r y o f v a lu e s a s s ig n e d to p e rs o n s w h o b e c a m e “ u n a b le " o n re s p e c tiv e m e a s u re s . P e r fo r m a n c e M e a su r e P o o r e s t S c o r e in 1 9 9 5 (se c o n d s) 3 % A b o v e P o o r e st S c o r e (se c o n d s) 2 5 % A b o v e P o o r e st S c o r e (se c o n d s) C h a ir S ta n d s 3 8 .3 0 3 9 .4 5 4 7 .8 8 F o o t T a p p in g 2 3 .9 0 2 4 .6 2 2 9 .8 8 T u rn in a C irc le 2 7 .4 0 2 8 .2 2 3 4 .2 5 W a lk /N o rm a l P a c e 9 0 .0 0 9 2 .7 0 1 1 2 .5 0 W a lk /F a s t P a c e 4 5 .2 0 4 6 .5 6 5 6 .5 0 A Tobit (short for "Tobin's Probit.” Breen. 1996) is a two-stage modeling technique that was originally developed by Tobin in 1958. He developed this technique to examine the ratio o f expenditures on durable goods to the total disposable income among 735 households. In his sample, 183 households had a score o f zero for total disposable income (his dependent variable). Although his data was "left” censored, this technique can be used for "right” censored data, as well. The two stages involved in this technique 2 0 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. involve 1) a selection stage and 2) an outcome stage (Breen. 1996). For the "right'' censored variables in this investigation, the first stage models the probability o f an individual remaining "able" to complete the task. This is the probability that the "right” censored observation we are interested has a change score that is in the selected sample (i.e., within the range o f those remaining "able” to complete the task). This probability is conditional on the variables thought to influence it. The second stage o f this technique models the expected change score, in physical performance for the “right” censored observations, based on the continued ability to perform the task. By using the SAS statistical package, these two steps are estimated jointly. Using SAS. a Tobit regression is accomplished by specifying a "lifereg” procedure that allows for several varieties of censoring (Allison. 1995). Using this procedure for "right” censored data, one needs to specify the value o f the dependent variable beyond which the data is censored and request a normal distribution. A normal distribution is specified because in the first stage of this technique we are asking for the probability that a normally distributed random value is less than a specified value. For a more advanced discussion o f this technique the reader is directed to Breen (1996). C. Results The results from these analyses are presented in two parts. Part I summarizes the findings from the alternative modeling techniques for the five dependent variables that predict change in: repeated chair stands, foot tapping, turning in a circle, walking at a normal pace and walking at a fast pace. Part II describes the predictors o f change for all twelve performance-based measures. 208 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1. P arti Case-Studies. The strengths and weaknesses o f the comparative analytical strategies for the five cases under investigation are summarized in Table VI-4. Overall, assigning individuals who became “unable” to perform (the repeated chair stands, turning in a circle, and walking at a normal and fast pace) a value o f 3% or 25% above the poorest score is problematic because we are creating a cluster o f observations at a defined point which results in a bimodal distribution of the change score in physical performance for these tasks. This biases the OLS coefficients. If individuals who became "unable” to perform a task are excluded from the analysis (Model 3). two problems surface. First, this model is problematic because it excludes the cases of greatest interest, namely those who decline the most on physical performance by becoming "unable" to complete the task. Second, the estimates are biased because they are based on a non-randomly selected sub-set (Breen. 1996). Model 4 compared those who remained "able" to perform the task to those who became "unable.” Due to the small number o f individuals who became “unable" to tap their feet (3.1%). turn in a circle (6.3%). and walk at a normal (4.0%) and fast (5.3%) pace, it was only possible to examine the characteristics of those who became "unable” to complete the five repeated chair stands. Consequently this technique is limited by the sample size. The Tobit (censored) regression in Model 5 addresses the limitations o f the earlier models by including the cases of interest (those who become "unable”) and adjusting for the "censored” observations. In Tables VI-5 to VI-9 the sample size is shown as the number o f uncensored cases followed by the number o f censored cases in parentheses. These two values should be summed to arrive at the total sample size for each case. 2 0 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T a b le V I-4 . S tre n g th s a n d w e a k n e s s e s o f th e c o m p a ra tiv e a n a ly tic a l s tra te g ie s M o d e l T e c h n iq u e U s e d to T r e a t T h o se W h o B e c a m e “U n a b le ” ...,--------------- j ----- ------- 0 S tr e n g th s W e a k n e sse s M o d e l 1 A ss ig n a t 2 5 % a b o v e w o rs t sc o re R e ta in s c a s e s o f in te re s t B ia s e d c o e ffic ie n ts M o d e l 2 A s s ig n a t 3 % a b o v e w o rst sc o re R e ta in s c a s e s o f in te re s t B ia s e d c o e ffic ie n ts M o d e l 3 A s s ig n to m is s in g R e s id u a ls a re n o rm a lly d is trib u te d a m o n g re m a in in g c a s e s O m its c a s e s o f in te re s t: B ia s e d c o e ffic ie n ts M o d e l 4 L o g is tic re g re s s io n : p re d ic t b e c o m in g " u n a b le ” R e ta in s c a s e s o f in te re s t S a m p le to o sm a ll to a n a ly z e in 4 o f th e 5 c a s e s M o d e l 5 T o b it (c e n s o re d ) re g re s s io n R e ta in s c a s e s o f in te re st: U n c e n s o re d (c e n s o re d ) ca s e s N o n e In Tables VI-5 to VI-9, the models show that for each o f the five cases, the same covariates significantly predict change in performance when using "right” censored regression (Model 5) and the OLS regression with assignment at 25% (Model I) or 3% (Model 2) above the poorest valid score. Since the OLS regressions with assignment (Model 1 and Model 2) yield biased coefficients because we are creating a bimodal distribution, the "right” censored regression (Model 5) is a better technique to use. When comparing the "right" censored regression (Model 5) to the OLS regression model that excludes individuals who became "unable" (Model 3). different baseline demographic and health characteristics predict change on these five performance-based measures of physical functioning. As shown in Tables VI-5 to VI-9, in contrast to the "right" censored regression (Model 5), increasing age significantly predicts greater decrement on four o f these five performance tasks (i.e., five repeated chair stand, turning in a circle, walking at a normal pace, and walking at a fast pace) in the restricted sample (Model 3). In addition, although gender is not significant when predicting change in performance on 210 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table VI-5. Predictors o f change on 5 repeated chair stands M o d e l 1 M o d e l 2 M o d e l 3 V a ria b le B ‘ (S E )2 B (S E ) B (S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n 0 .3 6 (0 .0 7 )c 0 .2 9 (0 .0 6 )c 0 .0 9 (0 .0 2 )c B a s e lin e C o g n itio n -0 .1 6 (0 .0 8 )" -0 .1 3 (0 .0 6 )" -0.01 (0 .02) P h y s ic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e -0 .S 4 (0 .2 3 )’ -0 .5 5 (0 .1 8 )b -0 .5 2 (0 .0 6 )c D e m o g ra p h ic C h a ra c te r is tic s A g e 0 .4 5 (0 .2 6 ) 0 .3 9 (0 .2 0 ) 0 .2 0 (0 .0 7 )b F e m a le G e n d e r -0 .2 8 (1 .5 4 ) - 0 .2 8 ( 1 .2 0 ) -0 .2 8 (0 .4 2 ) E d u c a tio n < 12 y e a rs - 1 .5 7 ( 1 .4 4 ) - 1 .1 9 ( 1 .1 3 ) 0 . 0 1 (0 .3 9 ) In c o m e < 5 1 0 ,0 0 0 2 .5 6 ( 1 .5 0 ) 2 .0 4 ( 1 .1 7 ) 0 .4 3 (0 .4 1 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 3 .4 6 (1 .6 1 )" 2.6 7 (1 .2 6 )’ - 0 .1 0 ( 0 .4 5 ) D ia b e te s 1.07 (2 .0 9 ) 1 .1 5 ( 1 .6 3 ) 1 .3 7 (0 .5 7 )" H ig h B lo o d P re ss u re 1 .1 2 ( 1 .3 5 ) 0 .9 2 ( 1 .0 5 ) 0 .3 8 ( 0 .3 7 ) M y o c a rd ia l In fa rc tio n - 2 .1 8 ( 2 .0 3 ) -1 .7 5 (1 .5 8 ) - 0 .1 8 ( 0 .5 5 ) S tro k e 1.54 (5 .2 7 ) 0 .9 6 (4 .1 0 ) - 1 .3 2 ( 1 .4 6 ) H ip F ra c tu re 1 9 .2 4 (4 .8 l)c 1 5 .1 2 (3 .7 5 )c 3.51 (1 .8 4 ) O th e r B o n e F ra c tu re s 2 .8 0 ( 1 .5 9 ) 2 .0 9 ( 1 .2 4 ) -0 .0 5 (0 .4 5 ) D e p re s s io n (> 15 ) 0 .2 0 ( 1 .3 5 ) 0 .1 6 ( 1 .0 5 ) 0 .0 8 (0 .3 7 ) H e a lth B e h a v io rs B o d y M a ss In d e x - M e d iu m (re f.4) — — — L o w (< 2 4 .2 7 ) -0 .4 3 (1 .6 5 ) - 0 .2 6 ( 1 .2 9 ) 0 .3 3 (0 .4 5 ) H ig h ( > 2 7 .7 7 ) 0 .2 2 ( 1 .5 9 ) 0 .2 5 ( 1 .2 4 ) 0 .4 7 (0 .4 3 ) S m o k in g S ta tu s - N e v e r (r e f .) — — — C u rre n t S m o k e r 2 .6 4 ( 1 .9 9 ) 2 .0 6 ( 1 .5 5 ) 0 .1 4 ( 0 .5 5 ) P a st S m o k e r - 0 .1 0 ( 1 .4 4 ) - 0 .1 4 ( 1 .1 2 ) -0 .4 4 (0 .3 9 ) A lc o h o l. < 1 g ra m /m o n th (r e f .) — — — N o n e in p a s t y e a r 1 .9 7 ( 1 .6 9 ) 1 .6 6 ( 1 .3 2 ) 0 .5 6 (0 .4 6 ) > 1 g ra m /m o n th 2 .7 4 ( 1 .7 4 ) 2 .2 4 ( 1 .3 6 ) 0 .5 9 (0 .4 8 ) N o r N o n - C e n s o r e d C a s e s (R t3 C e n s o r e d ) 381 381 3 2 7 A d ju s te d R 2 o r L o g L ik e lih o o d 0.12 0 .1 4 0 .2 4 Notes: ' U n s ta n d a rd iz e d c o e ff ic ie n t. 2 S ta n d a rd E rro r. 5 R ig h t. R e f e r e n c e ." p < .0 5 . b p < 0 1 . c p < .001 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table VI-5 (cont.). Predictors o f change on 5 repeated chair stands M o d e l 4 M o d e l 5 V a ria b le O d d s R a tio (9 5 % C l) B (S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n 1 .0 6 (1 .0 3 -1.10)c 0 .4 1 (0 .0 8 )c B a s e lin e C o g n itio n 0 .9 7 ( 0 . 9 3 - 1 . 0 1 ) -0 .1 9 (0 .0 9 )* P h y s ic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e 1.02 (0 .9 1 - 1.13) -0 .5 3 (0 .2 6 )* D e m o g r a p h ic C h a ra c te ris tic s A g e 1.07 (0 .9 5 - 1.23) 0 .4 9 (0 .3 0 ) F e m a le G e n d e r 1 . 0 4 ( 0 . 4 8 - 2 . 3 0 ) -0 .3 2 (1 .7 4 ) E d u c a tio n < 12 y e a rs 0 . 6 9 ( 0 . 3 4 - 1.42) -1 .8 3 (1 .6 3 ) In c o m e < S 1 0 ,0 0 0 1.80 (0 .8 5 - 3 .7 8 ) 2 .9 4 (1 .7 0 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 2.22(1.07-4.59)“ 3 .9 7 (1 .8 3 )* D ia b e te s 1.05 ( 0 . 3 6 - 3 . 0 7 ) 0 .9 6 (2 .3 6 ) H ig h B lo o d P re ss u re 1 . 1 9 ( 0 . 6 0 - 2 . 3 6 ) 1 .3 0 ( 1 .5 2 ) M y o c a rd ia l In fa rc tio n 0 . 6 0 ( 0 . 1 9 - 1.85) -2 .4 2 (2 .2 9 ) S tro k e 2 .0 4 ( 0 . 2 0 - 2 1 . 4 4 ) 1 .8 6 ( 5 .9 6 ) H ip F r a c tu re 1 9 .2 2 (3.44- I0 7 .1 2 )1 2 3 .8 3 (5 .8 2 )c O th e r B o n e F ra c tu re s 1.85 ( 0 . 8 9 - 3 . 8 4 ) 3 .3 0 ( 1 .8 1 ) D e p r e s s io n (> 15 ) 1.03 (0 .5 3 - 1.99) 0 .2 4 ( 1 .5 3 ) H e a lth B e h a v io rs B o d y M a s s In d e x - M e d iu m (re f.4) — — L o w (< 2 4 .2 7 ) 0 .7 9 ( 0 . 3 4 - 1 . 8 3 ) - 0 .4 9 ( 1 .8 7 ) H ig h ( > 2 7 .7 7 ) 0 .9 4 (0 .4 3 - 2 .0 7 ) 0 .2 5 ( 1 .8 0 ) S m o k in g S ta tu s - N e v e r (re f.) — — C u r re n t S m o k e r 1 .8 8 ( 0 .7 2 - 4 . 8 9 ) 3 .0 7 ( 2 .2 6 ) P a s t S m o k e r 1.07 (0 .5 1 - 2 . 2 2 ) - 0 .1 0 ( 1 .6 2 ) A lc o h o l. < 1 g ra m /m o n th (re f.) — — N o n e in p a s t y e a r 1.55 (0 .6 4 - 3 . 7 5 ) 2 .1 3 ( 1 .9 1 ) > 1 g ra m /m o n th 1.74 ( 0 . 6 9 - 4 . 3 8 ) 3 .0 8 ( 1 .9 7 ) N o r N o n - C e n s o r e d C a s e s (R tJ C e n s o re d ) 381 3 2 7 (5 4 ) | A d ju s te d R : o r L o g L ik e lih o o d 0.20 -1 3 9 3 .5 4 Notes: 1 U n s ta n d a r d iz e d c o e ffic ie n t. 1 S ta n d a rd E rro r. JR ig h t. R e f e r e n c e .* p < .05, b p < 0 l , cp < . 0 0! Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table Vl-6. Predictors o f change on foot tapping M o d e l l M o d e l 2 M o d e l 3 V a ria b le B 1 (S E )1 B (S E ) B (S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n 0 .2 4 (0 .0 3 )c 0 .2 0 (0 .0 2 )c 0 .0 9 (0 .0 2 )c B a s e lin e C o g n itio n -0 ,1 5 (0 .0 3 )c -0 .1 3 (0 .0 3 )c -0 .0 7 (0 .0 2 )c P h y sic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e -0 .5 6 (0 .2 2 )b -0 .4 9 (0 .1 9 )b -0 .2 6 (0 .1 2 )“ D e m o g ra p h ic C h a ra c te r is tic s A g e -0 .0 9 (0 .1 0 ) -0 .0 8 (0 .0 9 ) -0 .0 4 (0 .0 5 ) F e m a le G e n d e r 0 .5 9 ( 0 .6 0 ) 0 .6 3 (0 .5 0 ) 0 .8 0 (0 .3 2 )b E d u c a tio n < 12 y e a rs -1 .2 8 (0 .5 6 )“ -1 .2 1 (0 .4 7 )b -0 .9 3 (0 .3 l)b In c o m e < S I 0 ,0 0 0 -0 .3 0 (0 .5 8 ) -0 .2 9 (0 .4 9 ) -0 .2 6 (0 .3 2 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 1 .2 1 (0 .6 2 )“ 1.01 (0 .5 2 ) 0 .3 3 (0 .3 4 ) D ia b e te s -0 .4 7 ( 0 .8 1 ) -0 .2 6 (0 .68 ) 0 .4 6 (0 .4 3 ) H ig h B lo o d P re s s u re -0 .1 5 ( 0 .5 2 ) -0 .0 9 (0 .4 4 ) 0 .1 7 ( 0 .2 8 ) M y o c a rd ia l In fa rc tio n -0 .6 5 (0 .7 9 ) -0 .4 3 (0 .6 6 ) 0 .3 0 (0 .4 2 ) S tro k e 0 .7 3 (2 .0 4 ) 0 .8 2 ( 1 .7 2 ) 1 .1 2 ( 1 .0 9 ) H ip F ra c tu re 0 .3 0 ( 2 .0 3 ) 0 .3 0 ( 1 .7 1 ) 0 .3 0 (1 .0 9 ) O th e r B o n e F ra c tu re s -0 .6 5 (0 .6 2 ) -0 .4 8 (0 .5 2 ) 0 .1 3 (0 .3 3 ) D e p re s s io n (> 15 ) - 0 .1 4 ( 0 .5 2 ) - 0 .1 0 ( 0 .4 4 ) 0.01 (0 .2 9 ) H e a lth B e h a v io rs B o d y M a ss In d e x -M e d iu m (re f.4) — — — L o w (< 2 4 .2 7 ) 0 .3 0 (0 .6 4 ) 0 .2 5 (0 .5 4 ) 0 .1 6 ( 0 .3 5 ) H ig h ( > 2 7 .7 7 ) 0 .0 6 (0 .6 2 ) 0 .1 2 ( 0 .5 2 ) 0 .3 5 (0 .3 3 ) S m o k in g S ta tu s - N e v e r (re f.) — — — C u rre n t S m o k e r -0 .2 8 (0 .7 7 ) - 0 .1 8 ( 0 .6 5 ) 0.21 (0 .4 2 ) P a st S m o k e r -1 .0 2 ( 0 .5 6 ) -0 .9 2 (0 .4 7 )“ -0 .5 3 (0 .3 0 ) A lc o h o l, < I g r a m /m o n th (re f.) — — — N o n e in p a s t y e a r 0 .9 7 (0 .6 6 ) 0 .8 6 ( 0 .5 5 ) 0 .4 3 (0 .3 5 ) > I g r a m /m o n th 1 .1 8 ( 0 .6 7 ) 1.01 (0 .5 7 ) 0 .4 4 (0 .3 7 ) N o r N o n - C e n s o r e d C a s e s (R tJ C e n s o re d ) 381 381 3 6 8 A d ju s te d R : o r L o g L ik e lih o o d 0 .1 7 0 .1 7 0 .0 9 Notes: 1 U n s ta n d a r d iz e d c o e f f ic i e n t.: S ta n d a rd E rro r. 3R i g h t .4R e f e re n c e . '1 p < . 0 5 . b p < .0 1. c p < .0 0 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table Vl-6 (cont.). Predictors o f change on foot tapping M o d e l 4 M o d e l 5 V a ria b le O d d s R a tio (9 5 % C l) B (S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n N o t P o ssib le 0 .2 5 (0 .0 3 )c B a s e lin e C o g n itio n -0 .1 6 (0 .0 3 ) c P h y s ic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e -0 .5 7 (0 .2 2 )b D e m o g ra p h ic C h a ra c te ris tic s A g e -0 .0 9 (0 .1 0 ) F e m a le G e n d e r 0 .5 8 (0 .6 0 ) E d u c a tio n < 12 y e a rs -1 .2 9 (0 .5 7 )“ In c o m e < S I 0 .0 0 0 - 0 .3 0 ( 0 .5 9 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 1 .2 6 (0 .6 3 )* D ia b e te s -0 .5 2 (0 .8 2 ) H ig h B lo o d P re ss u re -0 .1 5 (0 .5 3 ) M y o c a rd ia l In fa rc tio n -0 .7 0 (0 .7 9 ) S tro k e 0.71 (2 .0 5 ) H ip F ra c tu re 0.31 (2 .0 5 ) O th e r B o n e F ra c tu re s -0.68 (0 .6 2 ) D e p re s s io n (> 15 ) - 0 .1 4 ( 0 .5 3 ) H e a lth B e h a v io rs B o d y M a ss In d e x - M e d iu m (re f.4) — L o w (< 2 4 .2 7 ) 0.31 (0 .6 5 ) H ig h ( > 2 7 .7 7 ) 0 .0 4 (0 .6 2 ) S m o k in g S ta tu s - N e v e r (re f.) — C u r re n t S m o k e r - 0 .3 0 ( 0 .7 8 ) P a s t S m o k e r -1 .0 4 (0 .5 6 ) A lc o h o l. < 1 g ra m /m o n th (re f.) — N o n e in p ast y e a r 1.00 (0 .66) > 1 g ra m /m o n th 1.22 (0 .6 9 ) N o r 3 6 8 (1 3 ) N o n - C e n s o r e d C a s e s (R t C e n s o re d ) A d ju s te d R 'o r L o g L ik e lih o o d -1 1 2 3 .5 1 Notes: 1 U n s ta n d a rd iz e d c o e ff ic ie n t. 2 S ta n d a rd E rro r. ’R ig h t. R e f e r e n c e .1 p < .05. b p < 0 l , cp < . 0 0l 214 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table VI-7. Predictors o f change on turning in a circle r— - -------- ------------- ------ »...... .... M o d e l 1 M o d e l 2 M o d e l 3 V a r ia b le B' (S E ): B (S E ) B (S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n 0 .2 4 (0 .0 4 )c 0 .1 9 (0 .0 4 )c 0 .0 2 (0 .0 1 )* B a s e lin e C o g n itio n -0.l4(0.05)b -0 .1 1 (0 .0 4 )b -0.02 (0 .01) P h y s ic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e -0 .2 3 (0 .4 3 ) -0 .2 6 (0 .3 5 ) -0 .3 3 (0 .1 1 )b D e m o g ra p h ic C h a ra c te ris tic s A g e 0 .0 2 ( 0 .1 6 ) 0 .0 5 (0 .1 3 ) 0 .1 4 (0 .0 4 )1 F e m a le G e n d e r -0 .6 3 (0 .9 0 ) -0 .4 1 (0 .7 3 ) 0 .4 7 (0 .2 3 )“ E d u c a tio n < 12 y e a rs - 1 .1 8 ( 0 .8 5 ) -1 .0 2 (0 .6 9 ) - 0 .3 7 ( 0 .2 1 ) In c o m e < $ 1 0 ,0 0 0 1 .3 2 ( 0 .8 8 ) 1 .1 0 ( 0 .7 2 ) 0 .2 5 (0 .2 2 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 2 .1 0 (0 .9 4 )“ 1 .7 4 (0 .7 6 )“ 0 .3 3 (0 .2 4 ) D ia b e te s 0 .5 6 ( 1 .2 3 ) 0 .4 9 ( 1 .0 0 ) 0 .1 4 ( 0 .3 1 ) H ig h B lo o d P re s s u re -0 .5 6 (0 .7 9 ) -0 .4 4 (0 .6 4 ) 0 .0 4 (0 .2 0 ) M y o c a rd ia l In fa rc tio n -0 .4 8 ( 1 .1 9 ) -0 .4 5 (0 .9 8 ) -0 .0 8 (0 .3 0 ) S tro k e -1 .2 7 ( 3 .0 7 ) -1 .0 2 (2 .5 0 ) -0 .0 4 (0 .7 4 ) H ip F ra c tu re 0 .0 4 ( 3 .0 3 ) 0 .3 9 (2 .4 6 ) 1 .7 8 (0 .7 3 )“ O th e r B o n e F ra c tu re s 0 .9 8 (0 .9 3 ) 0 .8 2 (0 .7 6 ) 0 .2 2 ( 0 .2 3 ) D e p re s s io n (> 15 ) -0 .2 1 (0 .7 9 ) -0 .0 8 (0 .6 4 ) 0 .4 4 (0 .2 0 )“ H e a lth B e h a v io rs B o d y M a s s In d e x - M e d iu m (re f.4) — — — L o w (< 2 4 .2 7 ) -0 .9 2 (0 .9 7 ) -0 .7 4 (0 .7 9 ) -0 .0 3 (0 .2 4 ) H ig h ( > 2 7 7 7 ) -0 .0 7 (0 .9 4 ) -0 .0 5 (0 .7 6 ) 0 .0 2 (0 .2 3 ) S m o k in g S ta tu s - N e v e r (r e f .) — — — C u rre n t S m o k e r 0 .3 8 ( 1 .1 6 ) 0 .3 3 (0 .9 4 ) 0 .0 5 (0 .2 9 ) P a st S m o k e r 0 .1 7 ( 0 .8 5 ) 0 .1 6 ( 0 .6 9 ) 0 .1 6 ( 0 .2 1 ) A lc o h o l. < 1 g ra m /m o n th (re f.) — — — N o n e in p a s t y e a r 0 .8 7 (0 .9 9 ) 0 .6 8 ( 0 .8 1 ) -0 .0 8 (0 .2 5 ) > I g ra m /m o n th 0 .7 2 ( 1 .0 2 ) 0 .6 3 (0 .8 3 ) 0.31 (0 .2 5 ) N o r N o n - C e n s o r e d C a s e s (R t3 C e n s o re d ) 3 7 7 37 7 3 5 4 | A d ju s te d R : o r L o g L ik e lih o o d 0 .0 8 0 .0 8 0.11 Notes: 1 U n s ta n d a r d iz e d c o e ff ic ie n t. 1 S ta n d a rd E rro r. 3R ig h t. 4R e f e re n c e .a p < . 0 5 , b p < .0 1 , “ p < .001 215 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table VI-7 (cont.). Predictors o f change on turning in a circle M o d e l 4 M o d e l 5 V a ria b le O d d s R a tio (9 5 % C l) B (S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n N o t P o ssib le 0 .2 5 (0 .0 5 )c B a s e lin e C o g n itio n -0 .1 5 (0 .0 5 )b P h y s ic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e -0 .2 2 (0 .4 5 ) D e m o g ra p h ic C h a ra c te ris tic s A g e 0.02 (0 .1 6 ) F e m a le G e n d e r -0 .71 (0 .9 3 ) E d u c a tio n < 12 y e a rs -1 .2 4 (0 .8 8 ) In c o m e < 5 1 0 ,0 0 0 1.40 (0 .9 1 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 2 .2 3 (0 .9 7 )a D ia b e te s 0 .5 9 ( 1 .2 8 ) H ig h B lo o d P re ss u re - 0 .6 0 ( 0 .8 2 ) M y o c a rd ia l In fa rc tio n -0 .5 8 (1 .2 4 ) S tro k e -1 .3 5 (3 .1 7 ) H ip F ra c tu re - 0 .0 7 ( 3 .1 3 ) O th e r B o n e F ra c tu re s 1.02 (0 .9 6 ) D e p re s s io n ( > 1 5 ) -0 .2 5 (0 .8 2 ) H e a lth B e h a v io rs B o d y M a ss In d e x - M e d iu m ( r e f .4) — L o w (< 2 4 .2 7 ) - 0 .9 7 ( 1 .0 0 ) H ig h ( > 2 7 .7 7 ) -0 .0 7 (0 .9 7 ) S m o k in g S ta tu s - N e v e r (re f.) — C u rre n t S m o k e r 0 .4 0 ( 1 .2 0 ) P a s t S m o k e r 0 .1 7 ( 0 .8 8 ) A lc o h o l. < 1 g r a n v m o n th (re f.) . . . N o n e in p a s t y e a r 0 .9 3 (1 .0 3 ) > 1 g ra m /m o n th 0 .7 7 ( 1 .0 6 ) N o r N o n -C e n s o re d C a s e s (R tJ C e n s o r e d ) 3 5 4 ( 2 3 ) A d ju s te d R : o r L o g L ik e lih o o d -1 2 5 0 .7 3 Notes: 1 U n s ta n d a rd iz e d c o e f f ic i e n t.: S ta n d a rd E rro r. 3 R ig h t. 4 R e fe re n c e .a p < .0 5 . bp < . 0 l . cp < . 0 0l Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table V!-8. Predictors o f change on walking at a normal pace M o d e l 1 M o d e l 2 M o d e l 3 V a ria b le B ' (S E )2 B (S E ) B (S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n 0 .6 9 (0 .1 2 )c 0 .5 7 (0 .1 0 )c 0 .0 9 (0 .0 4 )* B a s e lin e C o g n itio n -0 .3 6 (0 .1 4 )b -0 .3 0 (0 .1 l)b -0 .0 5 (0 .0 4 ) P h y s ic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e 0 .3 0 (0 .5 3 ) 0 .2 0 (0 .4 4 ) - 0 .1 0 ( 0 .1 8 ) D e m o g r a p h ic C h a ra c te r is tic s A g e -0 .0 8 (0 .4 1 ) 0 .0 0 (0 .3 4 ) 0 .3 2 (0 .1 3 )* F e m a le G e n d e r -3 .4 3 (2 .4 8 ) -2 .6 1 (2 .0 3 ) 0 .6 9 (0 .8 0 ) E d u c a tio n < 12 y e a rs -2 .6 5 (2 .3 1 ) -2 .0 3 (1 .8 9 ) 0 .3 5 (0 .7 4 ) In c o m e < S 1 0 .0 0 0 2 .3 0 (2 .4 0 ) 2 .0 0 ( 1 .9 7 ) 0 .8 3 (0 .7 7 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 3 .3 9 ( 2 .5 4 ) 2 .7 8 (2 .0 8 ) 0 .3 3 (0 .8 2 ) D ia b e te s 1 .3 0 ( 3 .3 4 ) 1 .1 0 ( 2 .7 4 ) 0 .2 0 ( 1 .0 8 ) H ig h B lo o d P re s s u re -2 .0 7 (2 .1 4 ) -1 .5 2 (1 .7 5 ) 0 .6 4 (0 .6 9 ) M y o c a rd ia l In fa rc tio n -2 .3 3 (3 .2 5 ) - 1.88 (2 .66 ) - 0 .0 2 ( 1 .0 3 ) S tro k e -2 .6 3 (8 .3 2 ) -2 .1 5 (6 .8 2 ) -0 .2 5 (2 .6 2 ) H ip F ra c tu re 1.03 (7 .6 6 ) 1.84 (6 .2 8 ) 4 .8 8 (2 .4 2 )“ O th e r B o n e F ra c tu re s -1 .2 5 (2 .5 2 ) -0 .8 1 (2 .0 6 ) 0 .9 9 (0 .8 0 ) D e p r e s s io n ( > 1 5 ) 1.68 (2 .1 5 ) 1.42 (1 .7 6 ) 0.41 (0 .7 0 ) H e a lth B e h a v io rs B o d y M a s s In d e x - M e d iu m ( r e f 1) — — — L o w (< 2 4 .2 7 ) - 3 .5 5 ( 2 .6 4 ) -3 .0 0 (2 .1 6 ) -0 .7 7 (0 .8 5 ) H ig h ( > 2 7 .7 7 ) - 0 .1 6 ( 2 .5 3 ) 0 .1 0 ( 2 .0 8 ) 1 .0 4 ( 0 .8 2 ) S m o k in g S ta tu s - N e v e r (re f.) . . . — — C u r re n t S m o k e r 2 .2 9 ( 3 .1 9 ) 1.88 (2 .6 2 ) 0 .1 8 ( 1 .0 4 ) P a st S m o k e r - 1 .3 7 ( 2 .2 9 ) -1 .0 8 (1 .8 7 ) 0 .0 3 (0 .7 3 ) A lc o h o l, < 1 g r a m /m o n th (re f.) — — — N o n e in p a s t y e a r 3 .4 7 (2 .6 9 ) 3 .0 5 (2 .2 0 ) 1 .4 5 ( 0 .8 6 ) > 1 g ra m /m o n th 1.77 (2 .7 6 ) 1 .5 7 ( 2 .2 6 ) 0 .9 0 (0 .8 9 ) N o r N o n - C e n s o r e d C a s e s ( R t3 C e n s o r e d ) 3 7 6 3 7 6 361 A d ju s te d R : o r L o g L ik e lih o o d 0 .0 9 0.10 0 .0 6 Notes: 1 U n s ta n d a r d iz e d c o e ff ic ie n t. 2 S ta n d a rd E rro r. 3 R ig h t. ^ R efere n ce.* p < . 0 5 . b p < .0 1 , 0 p < .001 217 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T a b le V l-8 (c o n t.). P re d ic to rs o f c h a n g e o n w a lk in g a t a n o rm a l p a c e M o d e l 4 M o d e l 5 V a ria b le O d d s R a tio (9 5 % C l) B ( S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n N o t P o s s ib le 0 .7 2 (0 .1 2 )c B a s e lin e C o g n itio n -0 .3 7 (0 .1 4 )b P h y s ic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e 0 .3 3 (0 .5 4 ) D e m o g r a p h ic C h a ra c te ris tic s A g e -0 .1 1 (0 .4 2 ) F e m a le G e n d e r -3 .6 2 (2 .5 0 ) E d u c a tio n < 12 y e a rs - 2 .7 9 ( 2 .3 3 ) In c o m e < S I 0 .0 0 0 2 .3 7 (2 .4 3 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 3 .5 2 (2 .5 7 ) D ia b e te s 1.34 (3 .3 8 ) H ig h B lo o d P re s s u re - 2 .1 9 ( 2 .1 6 ) M y o c a rd ia l In fa rc tio n - 2 .4 4 ( 3 .2 8 ) S tro k e - 2 .7 4 ( 8 .4 0 ) H ip F ra c tu re 0 .8 4 (7 .7 4 ) O th e r B o n e F ra c tu re s -1 .3 3 (2 .5 4 ) D e p re s s io n (> 1 5 ) 1.74 (2 .1 7 ) H e a lth B e h a v io rs B o d y M a s s In d e x - M e d iu m ( r e f .4) — L o w (< 2 4 .2 7 ) -3 .6 6 (2 .6 7 ) H ig h ( > 2 7 .7 7 ) -0 .2 3 (2 .5 6 ) S m o k in g S ta tu s - N e v e r (re f.) — C u r re n t S m o k e r 2 .3 8 (3 .2 3 ) P a st S m o k e r - 1 .4 2 ( 2 .3 1 ) A lc o h o l. < 1 g ra m /m o n th (re f.) — N o n e in p a s t y e a r 3 .5 7 (2 .7 2 ) > 1 g ra m /m o n th 1.83 (2 .7 9 ) N o r N o n - C e n s o r e d C a s e s (R t3 C e n s o r e d ) 3 6 1 ( 1 5 ) A d ju s te d R : o r L o g L ik e lih o o d -1 6 1 5 .1 0 Notes: 1 U n s ta n d a rd iz e d c o e f f ic i e n t.: S ta n d a rd E rro r. 3R ig h t. R e f e r e n c e .1 p < .0 5 . b p < .01 . c p < .001 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table VI-9. Predictors o f change on walking at a fast pace M o d e l 1 M o d e l 2 M o d e l 3 V a ria b le B 1 (S E )2 B (S E ) B (S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n 0 .4 1 (0 .0 6 )c 0 .3 3 (0 .0 5 )c 0 .0 4 (0 .0 2 )3 B a s e lin e C o g n itio n -0 .2 7 (0 .0 7 )c -0 .2 2 (0 .0 6 )c -0 .0 5 (0 .0 2 )b P h y sic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e - 0 .1 8 ( 0 .4 0 ) - 0 .1 9 ( 0 .3 2 ) - 0 .1 8 ( 0 .1 0 ) D e m o g ra p h ic C h a ra c te r is tic s A g e 0 .0 3 (0 .22 ) 0 .0 6 (0 .1 8 ) 0 .2 1 (0 .0 5 )c F e m a le G e n d e r - 0 .4 7 ( 1 .3 3 ) -0.31 (1 .0 7 ) 0.31 ( 0 .3 2 ) E d u c a tio n < 12 y e a rs - 1 .4 9 ( 1 .2 0 ) -1 .2 3 (0 .9 6 ) -0 .3 3 (0 .2 9 ) In c o m e < S I 0 ,0 0 0 0 .5 4 ( 1 .2 5 ) 0.51 (1 .0 1 ) 0 .4 4 (0 .3 0 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 2 .3 7 ( 1 .3 4 ) 2.01 (1 .0 7 ) 0 .6 6 (0 .3 3 )* D ia b e te s 1 .4 9 ( 1 .7 8 ) 1 .1 9 ( 1 .4 2 ) 0 .0 6 ( 0 .4 4 ) H ig h B lo o d P re s s u re - 0 .9 3 ( 1 .1 2 ) -0 .6 6 (0 .9 0 ) 0 .3 4 ( 0 .2 7 ) M y o c a rd ia l In fa rc tio n -0 .6 5 (1 .7 0 ) -0 .6 3 (1 .3 6 ) -0 .5 1 (0 .4 1 ) S tro k e -0 .0 7 (4 .3 0 ) 0 .2 4 (3 .4 5 ) 1 .3 8 ( 1 .0 1 ) H ip F ra c tu re 0 .0 6 ( 3 .9 7 ) 0 .5 7 ( 3 .1 8 ) 2 .4 1 (0 .9 4 )b O th e r B o n e F ra c tu re s -0 .3 4 (1 .3 3 ) - 0 .1 6 ( 1 .0 5 ) 0 .5 4 (0 .3 2 ) D e p re s s io n ( > 1 5 ) 0 .7 5 ( 1 .1 3 ) 0 .6 6 (0 .9 0 ) 0.31 (0 .2 7 ) H e a lth B e h a v io rs B o d y M a ss In d e x - M e d iu m (re f.4) — — — L o w (< 2 4 .2 7 ) -1 .4 6 (.1 3 7 ) - 1 .2 6 ( 1 .1 0 ) - 0 .5 6 ( 0 .3 3 ) H ig h ( > 2 7 .7 7 ) 0 .0 5 (1 .3 4 ) 0 .0 8 (1 .0 7 ) 0 .1 6 ( 0 .3 2 ) S m o k in g S ta tu s - N e v e r (re f.) — — — C u rre n t S m o k e r 1 .2 6 ( 1 .6 5 ) 1.13 (1 .3 3 ) 0 .6 0 (0 .4 0 ) P a s t S m o k e r - 0 .0 0 ( 1 .1 9 ) 0 .0 3 (0 .9 6 ) 0 .1 7 ( 0 .2 9 ) A lc o h o l, < I g r a m /m o n th (re f.) — — — N o n e in p a s t y e a r 0 .3 4 ( 1 .4 0 ) 0 .3 2 ( 1 .1 2 ) 0 .1 5 ( 0 .3 4 ) > 1 g r a m /m o n th 0 .5 5 (1 .4 3 ) 0 .5 3 (1 .1 5 ) 0 .4 5 (0 .3 5 ) N o r N o n - C e n s o r e d C a s e s (R t ’ C e n s o re d ) 3 6 6 3 6 6 3 4 8 A d ju s te d R 2 o r L o g L ik e lih o o d 0 .1 1 0.11 0 .1 3 Notes: 1 U n s ta n d a r d iz e d c o e ff ic ie n t. 2 S ta n d a rd E rro r. JR i g h t .4R e fe re n c e .1 p < .0 5 . b p < .0 1 , 1 p < .001 219 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table VI-9 (cont.). Predictors o f change on walking at a fast pace M o d e l 4 M o d e l S V a ria b le O d d s R a tio ( 9 S % C l ) B ( S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n N o t P o s s ib le 0 .4 3 (0 .0 6 )c B a s e lin e C o g n itio n -0 .2 8 (0 .0 7 ) c P h y s ic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e - 0 .1 8 ( 0 .4 1 ) D e m o g ra p h ic C h a ra c te ris tic s A g e 0.02 (0 .22 ) F e m a le G e n d e r - 0 .5 0 ( 1 .3 6 ) E d u c a tio n < 12 y e a rs - 1 .5 6 ( 1 .2 2 ) In c o m e < S I 0 ,0 0 0 0 .5 5 (1 .2 8 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 2 .4 7 (1 .3 6 ) D ia b e te s 1 .5 7 ( 1 .8 1 ) H ig h B lo o d P re ss u re - 0 .9 9 ( 1 .1 4 ) M y o c a rd ia l In fa rc tio n -0 .6 5 (1 .7 4 ) S tro k e - 0 .1 4 ( 4 .3 8 ) H ip F ra c tu re -0 .0 6 (4 .0 4 ) O th e r B o n e F ra c tu re s -0 .3 8 (1 .3 5 ) D e p re s s io n (> 15 ) 0 .7 7 ( 1 .1 5 ) H e a lth B e h a v io rs B o d y M a ss In d e x - M e d iu m (re f.'1 ) — L o w (< 2 4 .2 7 ) 1 .5 0 ( 1 .3 9 ) H ig h ( > 2 7 . 7 7 ) 0 .0 4 (1 .3 6 ) S m o k in g S ta tu s - N e v e r (re f.) — C u rre n t S m o k e r 1 .2 9 ( 1 .6 9 ) P a st S m o k e r -0 .0 2 ( 1.2 ) A lc o h o l, < 1 g r a m /m o n th (re f.) — N o n e in p a s t y e a r 0 .3 4 ( 1 .4 3 ) > I g ra m /m o n th 0 .5 6 ( 1 .4 6 ) N o r N o n - C e n s o r e d C a s e s (R t3 C e n s o re d ) 3 4 8 (1 8 ) A d ju s te d R : o r L o g L ik e lih o o d -1 3 3 5 .3 4 Notes: 1 U n s ta n d a rd iz e d c o e f f ic ie n t.: S ta n d a rd E rro r. "R ig h t. ‘R e fe re n c e .1 p < .0 5 . bp < . 01, cp < . 0 0l the full sample using a Tobit (Model 5), women showed greater decrement than men specific to foot tapping and turning in a circle when the sample is limited to the valid cases (Model 3). In contrast to the “right” censored regression (Model 5), specific health conditions/events also significantly predict decrement in performance in the restricted samples (Model 3): 1) diabetics take more time to complete the five repeated chair stands (Table VI-5); 2) individuals who have fractured a hip take more time to turn 220 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in a circle, walk at a normal pace, and walk at a fast pace (Tables VI-7. VI-8, and VI-9); 3) depressed older persons take more time to turn in a circle (Table VI-7); and 4) participants reported having had cancer take more time to walk at a fast pace (VI-9). As mentioned earlier the estimates are biased in Model 3 because they are based on a non- randomly selected sub-set (Breen. 1996). In addition, the estimates in Model 3 exclude the cases o f greatest interest (those who become '■unable”). Consequently, the Tobit "right” censored regression (Model 5) is, once again, the better technique. By comparing the results from the different statistical techniques for these five cases, the Tobit censored regression is the best procedure to use for the current investigation. A "left” censored regression model is used to predict change in the following performance-based exercises because these physical exercises were "left” censored: single leg stand, tandem stand-eyes open, tandem stand-eyes closed, tandem walk, and grip strength. Recall that these "left” censored variables are assigned a value o f zero (in seconds, steps or kilograms) if individuals were "unable” to perform the activity. Similarly, a "right” censored regression model is used to predict change on the "right” censored performance measures: standing from a seated position five times, signing one's name, foot tapping, turning in a circle, walking at a normal pace, and walking at a fast pace. Here, the "right” censored variables are assigned a value equal to 25% above the poorest score among those where were "able” to perform the activity. For the case specific to balance, an OLS regression is used to predict change on this task because this performance measure was calculated based on the valid (i.e., those who continue to be able to perform the exercise) scores o f the single leg stand and the tandem stand-eyes open. Consequently there are no censored cases for the physical 221 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. performance exercise o f balance and an OLS regression is the best statistical procedure for this case. 2. Part II Predictors o f Change for 12 Performance-Based Measures Decrement in physical performance based on lower values at follow-up. The results o f the six regression models predicting change in balance, single leg stand, tandem stand-eyes open, tandem stand-eyes closed, tandem walk, and grip strength are shown in Table V I-10. Decrement on all six of these tasks is based on lower values (in seconds steps, or kilograms) at follow-up. Furthermore, recall that except for the case specific to balance, individuals who were ’'unable” to perform the activity were assigned a value of zero (in seconds, steps, or kilograms). Controlling for other covariates, individuals with declining cognitive performance show decrement on all six of these measures ranging from 0.10 seconds for tandem stand-eyes closed, (and 0.10 steps for tandem walk. 0.10 kilograms for grip strength) to 0.18 seconds for balance, per unit decrement on cognition. Baseline cognitive functioning is also predictive o f change on all six of these performance-based exercises. Per unit increase in baseline cognitive ability, individuals can perform these exercises for longer intervals of time (e.g.. balance. 0.12 seconds: leg stand. 0.06 seconds, tandem stand-eyes open, 0.13 seconds; and tandem stand-eyes closed. 0.09 seconds), take 0.07 more steps on the tandem walk, and grip a dynamometer with more strength (0.09 kilograms). Furthermore, across all six physical performance items, baseline levels o f each respective physical performance measure is predictive of change in performance. 222 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. For example, per second increase in the ability to balance in 1988, respondents declined by 0.60 seconds in 1995. T a b le V I-1 0 . P re d ic to rs o f c h a n g e in p e rfo rm a n c e -b a s e d m e a s u re s w h e re d e c re m e n t is b a s e d o n lo w e r v a lu e s a t fo llo w -u p V a r ia b le B a la n c e L e g S ta n d T a n d e m S ta n d , E y e s O p e n B ' (S E ): B (S E ) B (S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n -0 .1 8 (0 .0 4 ) c -0 .1 2 (0 .0 2 )c -0 .1 6 (0 .0 3 )c B a s e lin e C o g n itio n 0 .1 2 (0 .9 4 )b 0 .0 6 (0 .0 3 )* 0 .1 3 (0 .0 3 )c P h y s ic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e -0 .6 0 (0 .0 6 )c -0 .5 0 (0 .0 7 )c -0 .8 4 (0 .0 7 )c D e m o g r a p h ic C h a ra c te r is tic s A g e -0 .2 8 (0 .1 3 )* -0 .1 8 (0 .0 8 )* -0.17(0.09)“ F e m a le G e n d e r -1 .3 6 (0 .7 7 ) -0 .4 7 (0 .4 8 ) -1 .4 0 (0 .5 4 )b E d u c a tio n < 12 y e a rs 0 .7 6 (0 .7 0 ) 0 .9 0 (0 .4 4 )“ 0 .2 0 (0 .5 0 ) In c o m e < S I 0 ,0 0 0 -0 .7 0 (0 .7 3 ) -0 .3 1 (0 .4 6 ) -0 .6 5 (0 .5 2 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 0 .0 0 (0 .7 9 ) 0 .0 2 (0 .5 0 ) -0 .3 2 (0 .5 6 ) D ia b e te s -2 .0 8 (1 .0 4 )* -1 .8 5 (0 .6 6 )b - 1 .3 2 ( 0 .7 4 ) H ig h B lo o d P re s s u re - 0 .5 7 ( 0 .6 5 ) -0 .2 9 (0 .4 1 ) -0 .2 5 (0 .4 6 ) M y o c a rd ia l In fa rc tio n 0 .7 8 (0 .9 9 ) 0 .5 8 (0 .6 2 ) 0 .11 (0 .7 0 ) S tro k e -1 .0 6 (2 .5 2 ) - 0 .0 6 ( 1 .5 5 ) - 0 .4 7 ( 1 .7 6 ) H ip F ra c tu re -6 .1 5 (2 .4 5 )b -4 .0 7 (1 .6 5 )b -3.88(1.83)“ O th e r B o n e F ra c tu re s -1 .2 6 (0 .7 7 ) -0 .8 6 (0 .4 9 ) -0 .6 1 (0 .5 5 ) D e p re s s io n ( > 1 5 ) 0 .6 3 (0 .6 6 ) 0 .1 2 ( 0 .4 2 ) 0 .41 (0 .4 7 ) H e a lth B e h a v io rs B o d y M a s s In d e x - M e d iu m ( r e f / ) — — — L o w (< 2 4 .2 7 ) - 0 .1 0 ( 0 .8 0 ) - 0 .1 7 ( 0 .5 0 ) 0 .2 8 (0 .5 6 ) H ig h ( > 2 7 .7 7 ) -0 .0 4 (0 .7 8 ) -0 .4 1 (0 .4 9 ) 0 .3 0 ( 0 .5 5 ) S m o k in g S ta tu s - N e v e r (re f.) — — — C u r re n t S m o k e r -0 .0 9 (0 .9 0 ) -0.51 (0 .6 1 ) -0 . 1 2 (0 .68) P a st S m o k e r 0 . 1 1 (0 .6 9 ) -0 .0 5 (0 .4 4 ) -0 .5 4 (0 .5 0 ) A lc o h o l, < 1 g r a m /m o n th (re f.) — — — N o n e in p a s t y e a r 1.33 (0 .8 1 ) - 0 .1 7 ( 0 .5 1 ) -1 .3 0 (0 .5 8 )* > 1 g r a m /m o n th -0 .6 3 (0 .8 4 ) - 0 .1 0 ( 0 .5 3 ) -0 .7 7 (0 .6 0 ) N o r N o n - C e n s o r e d C a s e s ( L f t.4 C e n s o re d ) 3 6 9 3 0 1 ( 6 9 ) 313(61) A d ju s te d R : o r L o g L ik e lih o o o d 0 .2 3 -8 7 2 .7 2 -9 4 5 .3 8 Notes: 1 U n s ta n d a r d iz e d c o e ff ic ie n t. 1 S ta n d a rd E rro r. 3R e fe re n c e . 4 L e f t / p < .0 5 , b p <0 1. ‘ p < .001 223 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table VI-10 (cont.). Predictors o f change in performance-based measures where decrement is based o n lo w e r v a lu e s a t fo llo w -u p V a r ia b le T a n d e m S ta n d , E y e s C lo se d T a n d e m W a lk G r ip S tr e n g th B (S E ) B (S E ) B (S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n -0 .1 0 (0 .0 3 )c -0 .1 0 (0 .0 3 )' -0 .1 0 (0 .0 3 )' B a s e lin e C o g n itio n 0 .0 9 (0 .0 3 )b 0 .0 7 (0 .0 3 )’ 0 .0 9 (0 .0 3 )b P h y s ic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e -0 .7 1 (0 .0 8 )c -0 .6 2 (0 .0 8 )' -0 .4 4 (0 .0 4 )' D e m o g ra p h ic C h a ra c te ris tic s A g e -0.11 (0 . 10) - 0 .1 4 ( 0 .1 1 ) -0.20 (0 . 11) F e m a le G e n d e r -0 .5 9 (0 .5 9 ) -2 .0 3 (0 .6 2 )' -4 .8 1 (0 .9 3 )' E d u c a tio n < 12 y e a rs -0 .1 1 (0 .5 2 ) -0 .6 3 (0 .5 5 ) 0 .8 4 (0 .6 0 ) In c o m e < 5 1 0 ,0 0 0 0 .0 6 (0 .5 7 ) 0 .0 7 (0 .6 0 ) -1 .6 3 (0 .6 3 )b H e a lth C o n d itio n s /E v e n ts C a n c e r 0 .2 2 (0 .6 3 ) -0 .2 8 (0 .6 5 ) -0.21 (0 .68 ) D ia b e te s -0 .3 4 (0 .8 5 ) - 1 .1 4 ( 0 .8 7 ) -1 .4 5 (0 .8 8 ) H ig h B lo o d P re ss u re -0 .3 1 (0 .5 1 ) -1 .3 6 (0 .5 3 )b -0 .6 5 (0 .5 5 ) M y o c a rd ia l In fa rc tio n 1 .1 5 ( 0 .7 6 ) 0 .7 4 (0 .8 0 ) -0 .0 4 (0 .8 5 ) S tro k e 2 .0 4 (2 .6 0 ) - 0 .6 6 ( 2 .7 7 ) 4 .5 4 (2 .1 2 )" H ip F ra c tu re -4 .0 5 (2 .8 1 ) -2 .9 2 (2 .3 0 ) -1 .9 4 (2 .0 1 ) O th e r B o n e F ra c tu re s 0 .3 6 (0 .6 0 ) - 0 .1 5 ( 0 .6 3 ) 0 .9 9 (0 .6 7 ) D e p re s s io n ( > 1 5 ) 0 .6 9 (0 .5 1 ) 0 .3 9 ( 0 .5 4 ) 0 .6 2 (0 .6 0 ) H e a lth B e h a v io rs B o d y M a ss In d e x - M e d iu m ( r e f ? ) — — — L o w (< 2 4 .2 7 ) 0.11 (0 .6 2 ) -0 .2 3 (0 .6 5 ) 0 .5 0 ( 0 .7 0 ) H ig h ( > 2 7 .7 7 ) 0 .1 8 ( 0 .5 9 ) -1 .2 1 ( 0 .6 2 )" 0 .6 5 (0 .6 7 ) S m o k in g S ta tu s - N e v e r (re f.) — — — C u r re n t S m o k e r -0 .5 2 (0 .7 2 ) -0 .3 3 (0 .7 6 ) -2 .0 1 (0 .8 5 )* P a s t S m o k e r -0 .5 4 (0 .5 5 ) -0 .8 6 (0 .5 8 ) -0 .8 2 (0 .6 0 ) A lc o h o l, < 1 g ra m /m o n th (re f.) — — — N o n e in p a s t y e a r 0 .4 2 (0 .6 4 ) 0 .2 5 (0 .6 7 ) -0 .7 9 (0 .7 1 ) > 1 g ra m /m o n th 0 .2 6 (0 .6 4 ) -0 .4 7 (0 .6 7 ) 0 .3 7 ( 0 .7 3 ) N o r N o n - C e n s o r e d C a s e s ( L ft.4 C e n s o re d ) 185 (4 8 ) 192 (4 4 ) 3 7 7 (1 ) A d ju s te d IT1 o r L o g L ik e lih o o o d -5 2 8 .6 7 -5 6 0 .5 9 -1 1 5 3 .1 0 Notes: 1 U n s ta n d a rd iz e d c o e f f ic i e n t.: S ta n d a rd E rro r. R e f e r e n c e . J L eft.* p < . 0 5 . b p < . 0 1. ‘ p < . 0 0 1 These models also show that being older, female, and having lower levels o f education and annual income predict decrement on some of these physical performance exercises. Older individuals balance for 0.28 fewer seconds, stand on a single leg for 0.18 fewer seconds, and hold a tandem stand with their eyes open for 0.17 fewer seconds Compared to men, women hold a tandem stand-eyes open for 1.40 fewer seconds: take 2 2 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.03 fewer steps on the tandem walk exercise, and have a weaker grip strength by 4.81 kilograms. Individuals with less than a high school education are able to hold a single leg stand for 0.90 more seconds than those who have at least a college education. Finally, individuals of lower economic status show decrement on grip strength by 1.63 kilograms compared to those with annual incomes o f $10,000 or more. Specific health factors are also predictive o f declining performance on these exercises. For instance, having diabetes predicts decrement on balance (2.08 seconds) and the single leg stand (1.85 seconds) while having high blood pressure predicts decrement on the tandem walk exercise (1.36 steps). Hip fractures also predicted decrement on several exercises: balance (6.15 seconds), single leg stand (4.07 seconds), and the tandem stand-eyes open task (3.88 seconds). Interesting!}', individuals who had a stroke predicts improved grip strength. This finding may reflect the retraining of some physical skills that stroke patients may have received in rehabilitation programs. Surprisingly, only in a few cases are specific health behaviors predictive of declining physical performance. Compared to persons who drink moderately, those who have not had any alcohol in the past year, declined on their performance of the tandem stand-eyes open exercise, by 1.30 seconds. In addition, a high BMI is predictive of decrement on the tandem walk exercise. Compared to persons with moderate BMI levels, those with high BMI take 1.21 fewer steps on the tandem walk. Finally, compared to those who never smoke cigarettes, current smokers have poorer grip strength (by 2.01 kilograms) over time. Decrement in physical performance based on higher values at follow-up. The results o f these six '‘right” censored Tobit regression models predicting the change 2 2 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in the following six exercises are shown in Table VI-11: repeat chair stands, signing one's name, tapping one's feet, turning in a circle, and walking at a normal and fast pace. Here, decrement is defined as a higher performance score (in seconds) at follow-up for each o f these tasks. Furthermore, recall that individuals who are "unable" to perform the activity are assigned a value equal to 25% above the worst possible score among those who are "able." Individuals with declining cognitive performance show decrement on all of these physical tasks after controlling for other covariates. Per decrement on cognition, individuals take more time to complete the repeated chair stands (0.41 seconds), sign their name (0.24 seconds), tap their feet (0.25 seconds), turn in a circle (0.25 seconds), walk at a normal pace (0.72 seconds) and walk at a fast pace (0.43 seconds). Furthermore, although baseline cognitive ability was also predictive o f decrement across all six o f these exercises, baseline physical performance was only predictive o f decrement specific to chair stands, signing one's name, and foot tapping. These models also reveal that only a few demographic, health, and behavioral characteristics predict decrement on these tasks o f physical performance. O f the demographic characteristics, participants with less than 12 years o f education tap their feet for 1.29 fewer seconds, compared to persons with at least a high school education. O f the health conditions and events variables, having had cancer predicts decrement on completing the repeated chair stand exercise (3.97 seconds), foot tapping (1.26 seconds), and turning in a circle (2.23 seconds). In addition, having fractured a hip predicts decrement on completing the repeated chair stand exercise (23.83 seconds). Finally, compared to participants who moderately drink (<l gram per month), those who 226 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. drink heavily (> 1 gram per month) decline on the ability to sign their name (1.23 seconds). T a b le V l - 1 1. P re d ic to rs o f c h a n g e in p e rfo rm a n c e -b a s e d m e a s u re s w h e re d e c re m e n t is b a s e d o n h ig h e r v a lu e s a t fo llo w -u p C h a ir S ta n d s H a n d S ig .4 F o o t T a p p in g V a r ia b le B ‘ (S E )J B (S E ) B (S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n 0 .4 1 (0 .0 8 )' 0 .2 4 (0 .0 2 ) c 0 .2 5 (0 .0 3 )c B a s e lin e C o g n itio n -0 .1 9 (0 .0 9 )* -0 .1 1 (0 .0 2 )c -0 .1 6 (0 .0 3 )c P h y sic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e -0 .5 3 (0 .2 6 )* -0 .1 8 (0 .0 9 )* -0 .5 7 (0 .2 2 )b D e m o g r a p h ic C h a ra c te ris tic s A g e 0 .4 9 (0 .3 0 ) 0 .0 3 (0 .0 8 ) -0 .0 9 (0 .1 0 ) F e m a le G e n d e r -0 .3 2 ( 1 .7 4 ) 0 .2 5 (0 .4 5 ) 0 .5 8 ( 0 .6 0 ) E d u c a tio n < 12 y e a rs -1 .8 3 (1 .6 3 ) -0 .5 1 (0 .4 3 ) -1 .2 9 (0 .5 7 )“ In c o m e < 5 1 0 ,0 0 0 2 .9 4 ( 1 .7 0 ) 0 .6 0 ( 0 .4 5 ) -0 .3 0 (0 .5 9 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 3 .9 7 (1 .8 3 )* -0 .0 0 (0 .4 8 ) 1 .2 6 (0 .6 3 )* D ia b e te s 0 .9 6 (2 .3 6 ) 0 .7 0 ( 0 .6 3 ) - 0 .5 2 ( 0 .8 2 ) H ig h B lo o d P re ss u re 1.30 (1 .5 2 ) 0 .5 2 ( 0 .4 0 ) - 0 .1 5 ( 0 .5 3 ) M y o c a rd ia l In fa rc tio n -2 .4 2 (2 .2 9 ) -0 .2 4 (0 .6 0 ) -0 .7 0 (0 .7 9 ) S tro k e 1 .8 6 ( 5 .9 6 ) 0 .8 3 ( 1 .6 1 ) 0.71 (2 .0 5 ) H ip F ra c tu re 2 3 .8 3 (5 .8 2 )c 0 .7 2 ( 1 .4 3 ) 0.31 (2 .0 5 ) O th e r B o n e F ra c tu re s 3 .3 0 ( 1 .8 1 ) 0.11 (0 .4 7 ) -0.68 (0 .6 2 ) D e p r e s s io n (> 15 ) 0 .2 4 ( 1 .5 3 ) - 0 .1 9 ( 0 .4 1 ) - 0 .1 4 ( 0 .5 3 ) H e a lth B e h a v io rs B o d y M a s s In d e x - M e d iu m ( r e f .5) — — — L o w (< 2 4 .2 7 ) - 0 .4 9 ( 1 .8 7 ) 0 .0 8 (0 .4 9 ) 0.31 (0 .6 5 ) H ig h ( > 2 7 . 7 7 ) 0 .2 5 ( 1 .8 0 ) 0 .2 5 (0 .4 7 ) 0 .0 4 ( 0 .6 2 ) S m o k in g S ta tu s - N e v e r (re f.) — — — C u r re n t S m o k e r 3 .0 7 (2 .2 6 ) -0 .0 3 (0 .5 9 ) -0 .3 0 (0 .7 8 ) P a s t S m o k e r - 0 .1 0 ( 1 .6 2 ) - 0 .5 8 ( 0 .4 3 ) -1 .0 4 (0 .5 6 ) A lc o h o l, < 1 g ra m /m o n th (re f.) — . . . . . . N o n e in p a s t y e a r 2 .1 3 ( 1 .9 1 ) 0 .7 8 (0 .5 1 ) 1.0 0 (0 .66 ) > I g ra m /m o n th 3 .0 8 ( 1 .9 7 ) 1 .2 3 (0 .5 2 )* 1.22 (0 .6 9 ) N o n - C e n s o r e d C a s e s (R tJ C e n s o re d ) 3 2 7 (5 4 ) 3 7 7 (3 ) 3 6 8 (1 3 ) L o g L ik e lih o o o d -1 3 9 3 .5 4 -1 0 .2 9 .5 3 -1 1 2 3 .5 1 Notes: U n s ta n d a r d iz e d c o e ffic ie n t. " S ta n d a rd E rro r. R ig h t. S ig n a tu re . R e fe re n c e . * p < . 0 5 . b p < 0 1, cp < .0 0 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T a b le V I - 11 (c o n t.). P re d ic to rs o f c h a n g e in p e rfo rm a n c e -b a s e d m e a s u re s w h e re d e c re m e n t is b a s e d o n h ig h e r v a lu e s a t fo llo w -u p _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ T u r n in C ir c le W a lk /N o r m a l W a lk /F a st V a ria b le B (S E ) B (S E ) B (S E ) C o g n itiv e Im p a irm e n t C h a n g e in C o g n itio n 0 .2 5 (0 .0 5 )c 0 .7 2 (0 .1 2 )c 0 .4 3 (0 .0 6 )c B a s e lin e C o g n itio n -0 ,1 5 (0 .0 5 )b -0 .3 7 (0 .1 4 )b -0 .2 8 (0 .0 7 )1 P h y sic a l P e rfo rm a n c e B a s e lin e P e rfo rm a n c e -0 .2 2 (0 .4 5 ) 0 .3 3 (0 .5 4 ) - 0 .1 8 ( 0 .4 1 ) D e m o g ra p h ic C h a ra c te r is tic s A g e 0 .0 2 ( 0 .1 6 ) -0 .1 1 (0 .4 2 ) 0.02 (0 .2 2 ) F e m a le G e n d e r -0 .7 1 (0 .9 3 ) - 3 .6 2 ( 2 .5 0 ) - 0 .5 0 ( 1 .3 6 ) E d u c a tio n < 12 y e a rs -1 .2 4 (0 .8 8 ) -2 .7 9 (2 .3 3 ) - 1 .5 6 ( 1 .2 2 ) In c o m e < 5 1 0 ,0 0 0 1.40 (0 .9 1 ) 2 .3 7 (2 .4 3 ) 0 .5 5 (1 .2 8 ) H e a lth C o n d itio n s /E v e n ts C a n c e r 2 .2 3 (0 .9 7 )1 * 3 .5 2 (2 .5 7 ) 2 .4 7 (1 .3 6 ) D ia b e te s 0 .5 9 ( 1 .2 8 ) 1 .3 4 ( 3 .3 8 ) 1 .5 7 ( 1 .8 1 ) H ig h B lo o d P re s s u re -0 .6 0 (0 .8 2 ) - 2 .1 9 ( 2 .1 6 ) - 0 .9 9 ( 1 .1 4 ) M y o c a rd ia l In fa rc tio n -0 .5 8 (1 .2 4 ) - 2 .4 4 ( 3 .2 8 ) -0 .6 5 (1 .7 4 ) S tro k e - 1 .3 5 ( 3 .1 7 ) - 2 .7 4 ( 8 .4 0 ) - 0 .1 4 ( 4 .3 8 ) H ip F ra c tu re - 0 .0 7 ( 3 .1 3 ) 0 .8 4 ( 7 .7 4 ) -0 .0 6 (4 .0 4 ) O th e r B o n e F ra c tu re s 1.02 (0 .9 6 ) -1 .3 3 (2 .5 4 ) -0 .3 8 (1 .3 5 ) D e p re s s io n (> 15 ) -0 .2 5 (0 .8 2 ) 1 .7 4 ( 2 .1 7 ) 0 .7 7 ( 1 .1 5 ) H e a lth B e h a v io rs B o d y M a ss In d e x - M e d iu m ( r e f .5) — — — L o w (< 2 4 .2 7 ) - 0 .9 7 ( 1 .0 0 ) - 3 .6 6 ( 2 .6 7 ) 1 .5 0 ( 1 .3 9 ) H ig h ( > 2 7 .7 7 ) -0 .0 7 (0 .9 7 ) -0 .2 3 (2 .5 6 ) 0 .0 4 ( 1 .3 6 ) S m o k in g S ta tu s - N e v e r (r e f .) — — — C u rre n t S m o k e r 0 .4 0 ( 1 .2 0 ) 2 .3 8 ( 3 .2 3 ) 1 .2 9 ( 1 .6 9 ) P a st S m o k e r 0 .1 7 ( 0 .8 8 ) - 1 .4 2 ( 2 .3 1 ) - 0 .0 2 ( 1 .2 ) A lc o h o l, < 1 g r a m /m o n th (re f.) — — — N o n e in p a s t y e a r 0 .9 3 (1 .0 3 ) 3 .5 7 ( 2 .7 2 ) 0 .3 4 (1 .4 3 ) > 1 g ra m /m o n th 0 .7 7 ( 1 .0 6 ) 1.83 (2 .7 9 ) 0 .5 6 ( 1 .4 6 ) N o n - C e n s o r e d C a s e s (R tJ C e n s o r e d ) 3 5 4 (2 3 ) 361 (1 5 ) 3 4 8 (1 8 ) L o g L ik e lih o o o d .. w--r~----- :-----— -- r ------ r -1 2 5 0 .7 3 — ---: --. _ s . — : — -1 6 1 5 .1 0 “ 1 ------- “ P ---T — -1 3 3 5 .3 4 Notes: U n s ta n d a r d iz e d c o e ffic ie n t. " S ta n d a rd E rro r. R ig h t. S ig n a tu re . R e fe re n c e . ap < .05. bp <01, cp < .001 D. Discussion Together these findings suggest that change in cognitive performance predicts change on physical performance based measures, irrespective o f the type o f functioning, and after controlling for baseline physical performance, cognitive performance demographic, health, and behavioral characteristics. Contrary to the classification scheme proposed of Barberger-Gateau and Fabrigoule (1997), this study found that for both novel 228 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. or demanding physical exercises (e.g., single leg stand) and routine physical activities (e.g., walking at a normal pace), cognitive impairment causes decrement on performance- based measures o f physical functioning. Despite the fact that evaluators are directing individuals on how to complete an exercise, physical performance exercises continue to demand the cognitive processes of the individual. The finding that health behaviors only influence change on four o f the physical performance exercises (e.g.. tandem stand-eyes open, tandem walk, grip strength, and signing one's name) is also intriguing. Most o f the research on the effect o f health behaviors on physical functioning have relied on self-reported measures of physical functioning (Guralnik and Kaplan. 1989; LaCroix et al., 1993; Clark et al., 1998; Visser. Langlois et al., 1998 ). These studies have found that poorer health behaviors contribute to poorer or declining physical functioning. Specific to performance-based measures o f physical functioning, the findings presented here contradict those o f Nelson et al. (1994). Among community residing women 65 and older. Nelson and colleagues (1994) found that higher BMI was positively associated with performance on grip strength and negatively associated with performance on tapping one's feet and walking speed. Here, high BMI was only found to predict decrement on the tandem walk exercise. In addition, although Nelson et al. (1994) did not find a significant association between smoking behavior and grip strength, this study found that compared to those who have never smoked, current smokers have a weaker grip strength over time. Finally, in regard to alcohol consumption, although Nelson et al. (1994) showed that nondrinkers (including former and never drinkers) have poorer performed physical functioning on grip strength, foot tapping, walking 6 meters, tandem stand (assumed with eyes open), tandem 2 2 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. walk, and five repeated chair stands, this study only finds an effect for tandem stand-eyes open and signing one's name. For these two performance measures, compared to those who drink moderately, participants who have not had any alcohol in the past year, declined on their ability to perform the tandem stand-eyes open exercise and sign their name. A possible explanation for the discrepancies between this study and that o f Nelson and colleagues (1994) could be due to the fact that this study includes men in the sample. Furthermore, while Nelson controlled for age. history' o f stroke, and physical activity, this study controlled for additional demographic and health characteristics that may have reduced the effect o f these health behaviors on physical performance. A limitation of this study is that these older persons are not necessarily comparable to the elderly population, in general, because they were originally selected on their high cognitive and physical functioning. Despite this fact, this study granted the opportunity to explore physical and cognitive decrement in parallel. In so doing, this study extends information to health care professionals and researchers who seek to assess the physical functioning o f cognitively impaired older persons. Guralnik. Branch et al.. (1989) have suggested that performance-based measures are superior to self-reported measures when assessing the physical functioning of cognitively impaired persons because cognitively impaired persons may have difficulty documenting how well they are able to function on specific tasks (e.g.. walking across a room). This investigation has shown that when assessing the physical functioning o f cognitively impaired persons, performance-based measures o f physical functioning are appropriate. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter VII Conclusion The number o f elderly in the United States are increasing. According to the most recent United States Census (U.S. Census Bureau, 2000) the number o f Americans 65 and older is expected to grow from 13.2 % in 2010 to 20.5 % in 2040. The projected increase in the size of our aging population has implications for health and social services; particularly to the needs o f older women who have higher life expectancy than men and may spend a longer time physically disabled. With increased life expectancy, the number o f people with handicaps and disabilities has increased such that Americans now spend 15 % o f their life span in an ’'unhealthy" state, impaired by diseases, disabilities or injuries (NCHS. 1993). The dependency imposed by functional limitations and disability is an important problem among older persons and its magnitude will most likely increase, with the expected increase in size o f our older population. Consequently it is important to examine the methods by which researchers choose to assess the physical functioning o f older persons. The proliferation o f functional assessment measures has lead to a growing interest in comparing self-reported and performance-based measures of physical functioning. Although researchers have already begun to examine the association o f these measures, the majority o f the studies have been in the cross-section. It is anticipated that the information extended here on the longitudinal association between self-reported and performance-based measures o f physical functioning will improve our ability to communicate among professional researchers, practitioners, and program directors. In this study, community-based longitudinal data, from MacArthur Study o f persons 231 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 70 to 79 at baseline, provide information on the relationship between self-reported and performance-based measures o f physical functioning, both in the cross-section and longitudinally. This study has examined the association between individual and summary measures o f self-reported and performance-based measures. Although the MacArthur Study was not originally designed to examine the concordance between self-reported and performance-based measures, the performance-based measures can be considered as representative indicators o f physical functioning for the comparisons made. This study has also provided information on the demographic, health (conditions/events, depressive symptomatology, and cognitive functioning), psychological, and behavioral characteristics that contribute to the observed discrepancies between self-reported and performance-based measures o f physical functioning both in the cross-section and longitudinally. Finally, this study tested whether change in cognitive performance is related to decrement on more physically demanding or novel tasks. A combination o f approaches suggests that: a) a relatively weak longitudinal association exists between the individual items selected on an a priori basis; b) a relatively weak association exists, in the cross-section, between summary measures o f three domains o f self-reported (ADL, gross mobility, and physical activity) and performance-based measures; c) longitudinally, a moderate association exists between summary measures o f three domains o f reported (ADL, gross mobility, and physical activity) and performance-based measures o f physical functioning; d) different determinants influence the observed discrepancies within different domains o f reported functioning (both in the cross-section and longitudinally); and e) change in cognitive Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. performance mirrors change on physical performance-based measures, irrespective o f the type o f functioning. The question as to which type o f measurement is most appropriate in which circumstance arises. Recently, it was shown that performance-based measures should not be viewed as being "superior’* (i.e.. more acceptable to patients, clinically feasible, reproducible, sensitive to change) compared to self-reported measures (Myers et al.. 1993). Because self-reported measures are easier to implement in most data collection endeavors (reduced expense, no special equipment or training needed) than performance- based measures, self-reported measures will continue to be used. The findings from this investigation suggest that researchers who rely on self-reported measures to assess the physical functioning o f older persons should consider the demographic, health, psychological and behavioral characteristics of their subjects because these characteristics may influence self-reported measures. Findings also suggest that should investigators choose performance-based measures to assess the physical decrement of their subjects, consideration must be given to the change in cognitive performance of their subject's, as well. Findings from the study also call for the development of an alternative conceptual model to describe the process by which individuals move toward a state o f “disability.” Specific findings are discussed below in a summary organized around the questions that have guided this study. Furthermore, the author suggests an alternative conceptual model, based on the well accepted models o f Wood (ICIDH, 1980). Nagi (1965, 1966, 1976. 1977) and Verbrugge and Jette (1994) for future researchers to 2 3 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. consider when describing the process by which individuals become disabled. This chapter concludes with recommendations for future research. A. Summary and Discussion o f Findings A "gold standard" for measurement o f physical functioning does not exist. Historically, studies evaluating functional disability have relied on individual and proxy self-reported measures to assess physical functioning due to their relative ease and reduced expense in implementation. Recently, however, there has been a growing interest examine the relationship between self-reported and performance-based measures of physical functioning. A good deal o f controversy exists in the literature on how to measure physical functioning (Mungall and Hainsworth. 1979; Linne et al.. 1980: Branch et al.. 1983; Tinetti et al.. 1986; Guralnik, Branch et al.. 1989: Law and Letts et al.. 1989; Dorevitch et al.. 1992; Cress et al.. 1995). Although studies examining the relationship between self-reported and performance-based measures o f physical functioning have found a moderate association between these measures (Jette and Branch, 1985; Kelly-Hayes et al., 1992; Sager et al.. 1992; Myers et al., 1993: Cress et al., 1995; Reuben et al.. 1995; Kempen, Van Heuvelen et al., 1996; Kempen, Steverink et al.. 1996: Mendes de Leon et al.. 1996; Merrill et al.. 1997; and Ferrer et al., 1999), research on their association over time, is limited. O f the three studies with longitudinal components (Myers et al.. 1993; Mendes de Leon et al.. 1996; Kempen. Sullivan et al., 1999), only Kempen. Sullivan et al. (1999) investigated how the observed discordance between reported and performed measures is influenced by other factors. Furthermore, although some studies have used summary reported 2 3 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. measures, when comparing reported and performed measures o f physical functioning, (Cress et al., 1995; Reuben et al., 1995; Kemen Van Heuvelen et al.. 1996; Kempen. Steverink et al.. 1996; Kempen, Sullivan et al., 1999). none have examined how specific factors influence the observed discrepancies within different domains of reported functioning. Consequently, the following three questions were asked: 1) What is the relationship between self-reported and performance-based measures over time? 2) What demographic, health (conditions/events, cognitive, depressive symptomatology), health behavior, and psychological characteristics influence the observed discrepancy between self-reported and performance-based measures over time? and 3) How do specific factors influence discrepancies between self-reported and performance-based measures within different domains o f reported physical functioning (in the cross section and over time)? The results showed that when comparing a priori matched pairs of reported and performed measures over time, the sensitivity o f the individual items selected showed lower rates o f sensitivity (the ability to correctly identify those who become unable to function) than earlier studies have shown in the cross-section (Merrill et a!.. 1997. Ferrer et al.. 1999). For instance, using data from the EPESE. Merrill et al. (1997) found that, when comparing the reported difficulty to stoop, crouch or kneel with the performed ability to complete five chair stands, the measures for both men (53%) and women (58%) had moderate rates o f sensitivity. Here, however, comparing these same exercises over time, overall (for both men and women) sensitivity of the measures was 30%. This 2 3 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. value is approximately half o f that found by Ferrer et al. (1999) when they compared the reported difficulty to stand up from a chair with the observed performance of five chair stands (63%). The lower association found is probably due to the increased measurement error in change variables. Furthermore, contrary to other cross-sectional studies (Elam et al., 1991; Dorevitch et al., 1992; Kelly Hayes et al., 1992; Sager et al.. 1992; Myers et al., 1993; Cress et al., 1995; Reuben et al.. 1995: Kempen. Steverink et al., 1996; Kempen. Van Heuvelen et al.. 1996; Merril et al.. 1997; and Ferrer et al.. 1999) this investigation suggests that there is a relatively weak association between self-reported and performance-based measures o f physical functioning in the cross-section. Within the three domains examined (ADL. gross motor functioning, and physical activity) a weak association was found specific to the domain of gross motor functioning and no association was found specific to the domains o f ADL or physical activity. This finding may be attributable to the high functioning nature of the participants. Participants o f the MacArthur Study were originally selected based on their high physical functioning. Consequently, this sample may not adequately represent the older population, in general, because even at the follow-up assessment (seven years after baseline measures were taken), a high percentage of individuals reported no limitations on the ADL (81.8%). physical activity (73.2%), and on the Rosow-Breslau scale o f gross mobility (55.4%). Recall, that cross-sectional analyses were conducted based on the data collected during this follow-up, in 1995. The longitudinal association between reported and performed measures of physical functioning, however, is moderate. This finding supports the findings by 2 3 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Myers et al. (1993), Mendes de Leon et al. (1996) and Kempen, Sullivan et al. (1999). A stronger association exists within the domain o f ADL than within the domain of gross motor functioning or physical activity. Although this is the first study to compare the association between different reported domains physical functioning and performance- based measures, others have found moderate longitudinal associations within the domain o f ADL (Mendes de Leon et al.. 1996: Kempen. Sullivan et al.. 1999) and IADL (Myers et al., 1993). The results also suggest that demographic, health, behavioral, and psychological characteristics differentially influence discrepancies among the three domains both in the cross-section and longitudinally. In the cross-section, reasonable contributions from demographic, cognitive, health, and behavioral characteristics are made to reported levels o f functioning within all three domains (ADL, gross mobility, and physical activity). This study supports the findings from Kempen, Steverink et al. (1996) that age. gender, and education do not influence the discrepancies within the domain o f ADL. In addition, the domain o f physical activity is particularly affected by health conditions or events and health behaviors. While heavy drinkers under-report dysfunction within the domain of physical activity, those who have not had any alcohol in the past year over-report dysfunction. Longitudinally, discrepancies within the domain o f ADL and gross motor functioning are particularly affected by psychological attributes, while those within the domain o f gross motor functioning and physical activity are influenced by health behaviors. While individuals with higher levels o f mastery under-report decrement within the domain o f ADL, those with higher levels o f self-efficacy under-report 2 3 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. decrement within the domain o f gross motor functioning, as compared to performance. In addition, within the domains o f gross motor functioning and physical activity, those who have not had any alcohol within the past year over-report decrement in physical functioning, as do those with higher BMI. Furthermore, change in cognitive functioning and levels o f depressive symptomatology influence discrepancies within the domain of ADL. alone. Those with declining cognitive functioning and higher levels o f depressive symptomatology over-report decrement with the domain o f ADL. Although self-reported measures are easier to administer than performance-based measures, the effects o f cognitive functioning, depressive symptomatology, psychological attributes (mastery, self-efficacy), and health behaviors (BMI. alcohol consumption) should be considered in the application of specific self-reported measures to assess physical decrement. As demonstrated, within different domains o f reported measures o f physical functioning, different factors contribute to the discrepancies between reported and performed measures of physical functioning. Researchers who choose to use self-reported measures within the domain of ADL should pay particular attention to participant's cognitive functioning, depressive symptomatology, BMI. and levels o f mastery because these characteristics are likely to influence this type o f reported measure. Although BMI is also likely to influence reported measures within the domain o f gross motor functioning, so too are levels o f self-efficacy and alcohol consumption. Finally, when applying measures from the domain o f physical activity, gender, economic status, BMI. and alcohol consumption are likely to confound one's results. The final purpose o f this investigation was to examine the relationship between cognitive impairment and physical functioning. This aspect o f the investigation 2 3 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. advances our knowledge concerning the characteristics of older people in the United States who are at greater risk of decrement on specific physical tasks. Based on the findings o f previous researchers, the author asked: 4) Do individuals with declining cognitive abilities manifest decrement on more physically demanding or novel tasks than more routine physical activities when baseline physical performance, cognitive performance, demographic, health, and behavioral characteristics are controlled for? The findings suggest that change in cognitive performance mirror change on physical performance based measures, irrespective o f the type o f functioning and after controlling for baseline physical performance, cognitive performance, demographic, health, and behavioral characteristics. Contrary to the classification proposed by Barberger-Tateau and Fabrigoule (1997), this study found that for both novel or demanding physical exercises (e.g.. single leg stand) and routine physical activities (e.g.. walking at a normal pace), cognitive functioning parallels physical functioning. Despite the fact that evaluators are directing individuals on how to complete an exercise, physical performance exercises continue to demand the cognitive processes of the individual. This information should be particularly helpful to health care professionals who implement preventive programs aimed at maintaining current levels o f physical independence. As mentioned, specific preventive strategies directed toward functional limitations that are associated with cognitive impairment may serve to improve an individual’s intrinsic ability to function. For instance, from a public health perspective, primary preventive programs that target specific health conditions, such as hypertension 2 3 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Guo et al.. 1997; Palombo et al.. 1997) or diabetes (Dey. 1997) associated with cognitive impairment may reverse or slow the progression toward disability. B. Implications for Future Research The findings presented point to the need for further research on the patterns o f loss o f physical and cognitive functioning related to environmental factors and subsequent disability. Unfortunately, there is little research on the effect o f the environment on physical dysfunction. With the continued growth, fragmentation, and demands on our current health care systems, the development o f emerging “Super Networks" consisting o f providers, managed care organizations, designers, and developers is foreseen to address the physical (environmental and rehabilitative) needs of our population. As the demand for aesthetically appealing innovations, that permit disabled individuals to remain in their homes and promote their sense o f independence and subsequent "quality o f life." increases, it is expected that an understanding o f the effect o f the environment on various stages o f physical dysfunction will also grow. To more carefully examine the relationship between the physical and cognitive functioning and the environment, a improved conceptual understanding is needed. The lack o f a well accepted conceptual scheme and the use o f confusing terminology in the scientific literature specific to the field of disability has resulted in the implementation of concepts from different models in research studies. As Verbrugge and Jette (1994) have noted, the use of the terms impairment, functional limitations, and disability have alternative and overlapping meanings and their use together are confusing. For instance, the combined use o f these terms has confused the conceptual understanding o f 2 4 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. research studies: '‘functional disabilities” (Leon and Lair, 1990), or "impaired physical functioning in daily living activities” (NIH, 1990). This confusion may be due to the fact that most instruments used to assess physical functioning pre-date the International Classification o f Impairments, Disabilities, and Handicaps (ICIDH) (WHO, 1980). Consequently, researchers have tried to operationalize their studies by forcing specific measures within the concepts defined by the ICIDH or by other subsequent theoretical models such as that proposed by Nagi (1965, 1969, 1976, 1977) or Verbrugge and Jette (1994). As discussed in Chapter II, all three o f these theoretical models (ICIDH, by Wood, 1980; Nagi framework. 1965. 1969, 1976, 1977; and the Disablement Process by Verbrugge and Jette, 1994) are limited because they assume a single linear causal process beginning with a pathology which leads to an impairment, functional limitations (or disability), and culminating in the expression o f a disability (or handicap). This causal path is problematic because there are: 1) pathologies that lead directly to a functional limitation (for instance in the case o f obsessive-compulsive disorder disabilities are part o f the pathology), and 2) one could hypothesize that physical dysfunction could lead to impaired circulation or "impairment” o f an organ or physiological system. Furthermore, the author suggests that instead of distinguishing "functional limitations” and "disability” by the terms "action” and "activity” (as Verbrugge and Jette. 1994 suggest) the concepts should be distinguished by incorporating the role o f the environment, as Wood (ICIDH,1980) originally suggested. Recall, that Wood defined handicap as a disadvantage placed on the individual by the environment. This "disadvantage” is what should be called "disability.” Unfortunately, however. 24 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wood did not explicitly detail the role o f the environment in the original ICIDH. The idea explicitly incorporating the role o f the environment has recently been shown in the draft revision o f the ICIDH, the ICIDH-2. Although the draft version of the ICIDH-2 attempts to propose a simpler and more inclusive model for describing the components that contribute to greater physical dysfunction and eventual disability, this model is hampered by it’s complexity and redundancy. For instance, although the ICIDH-2 explicitly includes an environmental dimension to the model, the distinction between the dimensions o f "activities," and "participation.” Consequently, the ICIDH-2 is difficult to operationalize. For instance, o f the four dimensions of the ICIDH-2 (e.g.. body structure, participation, activities, and environment) reported measures o f physical functioning could fall within the dimension o f "activities” or "participation." If the ICIDH-2 is to be used for scientific investigation, the definitions and distinctions of these two dimensions (activities and participation) is needed to better operationalize these constructs. The author provides an alternative theoretical model aimed at addressing these limitations, see Figure V II-1. First, at the center of the model, it is suggested that risk factors (demographic, behavioral, psychological) predispose an individual to certain pathologies. This idea is based on that of Verbrugge and Jette (1994) who suggest that these factors are pre-disposing in that the exist at or before the outset of the disablement process. These pathologies can lead to either "impairment" or "functional limitations." Although the concepts o f “impairment" and "functional limitations" are defined according to Nagi (1965, 1969. 1976, 1977), functional limitations should also include self-care activities. The decision to define ADL as “functional limitations." instead of " ‘disability” stems from the author’s disagreement with others (Nagi, 1965. 1969, 2 4 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. F ig u r e V I M . A n a lte r n a tiv e th e o r e tic a l m o d e l o f p h y s ic a l d y s f u n c tio n to Disability Environment Risk Factors Pathology ^ Functional Limitations Impairment Environment Disability 1976. 1977; Verbrugge and Jette. 1994). The findings from this investigation lend support to the theoretical conceptualization that self-reported measures o f ADL should be incorporated into the definition o f "functional limitations" because a stronger association was found between this domain o f reported functioning and performance-based measures o f functional limitations than between other domains of functional limitations (Nagi measures o f physical activity, 1976; and Rosow and Breslau measures o f gross mobility. 1966) and performed measures. Consequently, measures of ADL are classified as "functional limitations" and not "disability." An additional difference in the proposed model is the recognition that the relationship between functional limitations and impairments can proceed in either direction. Impairments can lead to functional limitations and functional limitations can cause impairments. Surrounding this central framework is the environment (e.g.. social, structural, political). "Disability" then results from the interaction o f any o f the three central concepts (pathology, impairment, functional limitations) with the environment. Clearly, research is needed on testing this theoretical model. Although this model is grounded on the accepted models o f Wood (ICIDH. 1980). Nagi (1965.1969, 1976. 1977). and Verbrugge and Jette (1994), it is hoped that this model offers additional conceptual and operational clarity. Researchers interested in testing this theoretical model may consider asking the following questions: 1) What is the effect o f the environment on physical dysfunction?; 2) What pattern o f loss o f physical and cognitive functioning is related to environmental factors and subsequent disability?; and 3) What functional limitations are related to specific pathologies? 2 4 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Information gained from the investigations exploring these questions would provide insight on how the conceptual stages o f physical dysfunction interact with each other and the environment. A better understanding o f the pathology, impairment, and functional characteristics that are associated with oppressive environments is of growing importance. Such information would serve to guide public health professionals in developing preventive programs aimed at preventing, postponing, or reducing the levels o f disability in the older population. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. References Abalan. F. Alzheimer's disease and malnutrition: a new etiological hypothesis. Medical Hypotheses. ]_5, 385-393. Ahto, M., Isaho, R., Puolijoki, H., Laippala, P., Romo, M.. and Kivela, S.L. (1989). Functional abilities o f elderly coronary heart-disease patients. Aging-Clinical and Experimental Research. 10 (2), 127-136. Alexopoulos, G.S., Vrontou, C., Kakuma. T.. Meyers, B.S.. Young, R.C.. Klausner, E., Clarkin, J. (1996). Disability in geriatric depression. American Journal of Psychiatry, 153, 877-885. Allaire, S.H.. Lavalley, M.P., Evans, S.R., O'Connor, G.T.. Kelly-Hayes, M., Meenan, R.F.. Levy, D., and Felson, D.T. (1999). Evidence for decline in disability and improved health among persons 55 to 70 years: The Framingham Heart Study. American Journal o f Public Health. 89 (11). 1678-1683. Allen, J.K.. (1987). Factors related to functional status following coronary bypass surgery. Unpublished doctoral dissertation. Johns Hopkins University School of Hygiene and Public Health, Baltimore, MD. Allison, P.D. (1995). Survival Analysis Using the SAS System. SAS Institute Inc., Cary. North Carolina. Anderson. K.O.. Bradley, L.A., Young, L.D.. McDaniel, L.K.. and Wise. C.M. (1985). Rheumatoid arthritis; review of psychological factors related to etiology, effects and treatment. Psychological Bulletin. 98, 358-387. Andrews, G., Tennant. C., Hewson, D.M., and Vaillant. G.E. (1978). Life event stress, social support, coping style and risk o f psychological impairment. Journal of Nervous and Mental Disease. 166, 307-315. Antonucci, T.C. (1990). Social support and social relationships. In R.H. Binstock and L.K. George (Eds.). Handbook o f aging and the social sciences (pp. 205-226). Sand Diego, CA: Academic Press. Atchley, R.C. (1988). Social Forces and Aging (5th edition), Belmont, CA: Wadsworth. Avlund, K., Schroll, M., Davidsen, M., Lovborg, B.. and Rantanen, T. (1994). Maximal isometric muscle strength and functional ability in daily activities among 75 year old men and women. Scandinavian Journal of Medical Science, 4, 32-40. Ayanian, J.Z., Guadagnoli, E., and Cleary, P.D. (1995). Physical and psychosocial functioning o f women and men after coronary-artery bypass-surgery. Journal of the American Medical Association. 274 (22), 1767-1770. 2 4 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Badley, E.M. (1987). The ICIDH: Format, application in different settings and distinction between disability and handicap. International Disability Studies. 9 (3). Badley, E.M. (1993). An introduction to the concepts and classifications o f the International Classification of Impairments. Disabilities, and Handicaps. Disability Rehabilitation, 15, 161-178. Baltes. M.M., Wahl. H.W.. and Schmid-Furstoss. U. (1990). The daily life of elderly Germans: Activity patterns, personal control and functional health. Journal of Gerontology: Psychological Sciences. 45, PI 73-PI 79. Bandura, A. (1977). Self-efficacy: Toward a unifying theory o f behavioral change. Psychological Review. 84, 191-215. Bandura, A. (1986). Social Foundations o f Thought and Actions: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall Inc. Bandura, A. (1989). Human agency in social cognitive theory. American Psychologist. 44,1175-1184. Barberger-Gateau, P.. Chaslerie, A.. Dartigues, J.F., Commenges. D.. Gagnon, M., and Salamon. R. (1992). Health measures correlates in a French elderly community population: the PAQUID study. Journal o f Gerontology: Social Sciences. 47, S88- S95. Barberger-Gateau, P, Commenges, D., Gagnon, M.. Letenneur. L., Sauvel, C., and Dartigues, J.F. (1992). Instrumental Activities o f Daily Living as a Screening Tool for Cognitive Impairment and Dementia in Elderly Community Dwellers. Journal o f American Geriatrics Society. 40, 1129-1134. Barberger-Gateau, P.and Fabrigoule, C. (1997). Disability and cognitive impairment in the elderly. Disability and Rehabilitation. 19(5), 175-193. Barrera, M. (1981). Social support in the adjustment of pregnant adolescents: Assessment issues. In B.H. Gottlieb (Ed.) Social Networks and Social Support (pp. 69-96). Beverly Hills, CA: Sage. Baumgartner, R.N., Stauber, P.M., McHugh, D., Koehler, K.M., and Garry, P.J. (1995). Cross-sectional age differences in body composition in persons 60+ years o f age. Journal o f Gerontology: Medical Sciences. 50, M307-M316. Baylor. A.M., and Spirduso, W.W. (1988). Systematic aerobic exercise and components o f reaction time in older women. Journal o f Gerontology: Psychological Sciences. 43, P121-P126. 2 4 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Beckett, L.A., Brock. D.B., Lemke, J.H., Mendes de Leon, C.F., Guralnik, J.M.. Fillenbaum, G.G., Branch, L.G., Wetle, T.T.. and Evans, D.A. (1996). Analysis of change in self-reported physical function among older persons in four population studies. American Journal of Epidemiology. 143. 766-78. Befante, R., Reed, D., and Brody, J. (1985). Biological and social predictors o f health in an aging cohort. Journal o f Chronic Disease, 38, 385-395. Berg. K.. Wood-Dauphinee, S., Williams, J.I., and Gayton, D. (1989). Measuring balance in the elderly: Preliminary development o f an instrument. Physiotherapy Canada. 41, 304-311. Berg, K. (1992). Berg Balance Scale. Unpublished doctoral dissertation. McGill University, Montreal. Berg. K... Maki, B., Williams, J .I., Holliday, P.. Wood-Dauphinee, and S.L. (1992). Clinical and Laboratory measures o f postural balance in an elderly population. Archives o f Physical Medicine Rehabilitation. 73. 1073-1080. Bergner, M.. Babbitt, R.A.. Carter W.B. and Gilson, B.S. (1981). The sickness impact profile development and final revision o f a health status measure. Medical Care. J9, 787-805. Berkman, L.F. and Syme, S.L. (1979). Social networks, host resistance and mortality: A nine year follow-up study o f Alameda County residents. American Journal o f Epidemiology. 109. 186-204. Berkman, L.F. and Breslow. L. (1983). Health and Wavs o f Living. The Alameda Countv Study. Oxford University Press. New York, 237. Berkman, L.F., (1985). The relationship o f social networks and social support to morbidity and motility . In S.Cohen and S.L. Syme (Eds.), Social Support and Health (pp. 241-262). Orlando. FL: Academic Press. Berkman. L.F., Berkman. C.S., Kasl, S., Freeman, D.H. Jr., Leo, L., Ostfeld, A.M., Comoni-Huntley, J., and Broody, J.A. (1986). Depressive symptoms in relation to physical health and functioning in the elderly. American Journal o f Epidemiology. 124. 372-388. Berkman, L.F., Leo-Summers, L., and Horwitz. R.I. (1992). Emotional support and survival after myocardial infarction: a prospective, population-based study of the elderly. Annals o f Internal Medicine. 117, 1003-1009. 248 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Berkman, L.F.. Seeman, T.E., Albert. M. Blazer, D., Kalin, R., Mohs, R., Finch, C., Schneider, E., Cotman, C., McCleam, G., Nesselroade, J.. Featherman, D., Garmezy, N., McKhann, G.. Brim, G., Prager, D., and Rowe, J. (1993). High, usual and impaired functioning in community dwelling older men and women: Findings from the MacArthur Foundation Research Network on Successful Aging, Journal o f Clinical Epidemiology, 46, 1129-1140. Bettinghaus, C.O. (1980). International Standards for a System of Disability Classification. Paper presented at the Annual Meeting o f the American Psychological Association, Montreal. 1980. Bickenback, J.E., Chatteiji, S.. Badley, E.M., and Ustun, T.B. (1999). Models of disablement, universalism and the international classification o f impairments, disabilities and handicaps. Social Science and Medicine. 48, 1173-1187. Billings, A.G., and Moos, R.H. (1981). The role o f coping responses and social resources in attenuating the stress of life events. Journal o f Behavioral Medicine. 4, 139- 157. Binder, E.F.. Storandt, M., and Birge, S. (1999). The relationship between psychometric test performance and physical performance in older adults. Journal of Gerontology: Medical Sciences. 54A (8), M428-M432. Blanke. D.. and Hatgeman. P. (1989). Comparison o f gait o f young and elderly men. Physical Therapy. 69. 144-148. Blazer. D.G. (1982). Social support and mortality in an elderly community population. American Journal o f Epidemiology. 115.684-694. Borson, S.. Barnes, R.A., Kukull, W.A., Okimoto, J.T.,Veith. R.C., Inui, T.S., Carter, W., and Raskind. M.A. (1986). Symptomatic depression in elderly medical outpatients. Prevalence, demography, and health service utilization. Journal o f the American Geriatric Society, 34, 341-347. Boult. C.. Kane, R.L., Louis, T.A.. Boult, L., and McCaffey, D. (1994). Chronic conditions that lead to functional limitations in the elderly. Journal of Gerontology: Medical Sciences. 49, M28-M36. Bowling, A., and Brundy, E. (1997). Activities o f daily living: changes in functional ability in three samples o f elderly and very elderly people. Age and Ageing. 26 (2). 107-8. Branch, L.G., Katz, S., Kniepman, K., and Papsider, J.A. (1984). A prospective study of functional status among community elders. American Journal of Public Health. 74,266-8. 2 4 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Branch, LG. (1985). Health practices and incident disability among the elderly. American Journal o f Public Health. 75. 1436-1439. Branch. L.G., Kohout.F., Foley, D., Guralnik, J., Kasl, and S.. Fillenbaum, G. (1986). A prospective panel study o f physical function among community dwelling elders in East Boston, Iowa and New Haven. Presented at the 19lh Annual Meeting o f the Society for Epidemiologic Research, Pittsburgh. PA. Branch, L.G., and Meyers, A.R. (1987). Assessment o f physical functioning in the elderly. Clinics in Geriatric Medicine. 3, 29-51. Breen, R. (1996). Regression Models: Censored. Sample Selected, or Truncated Data. Thousand Oaks, CA: Sage. Brennan, T.A., Leape. L., Laird, N., Herbert, L., and Localio, A.R. (1991). Incidence of adverse events and negligence in hospitalized patients results o f the Harvard Medical Practice Study. New England Journal of Medicine, 324, 370-376. Broadhead. W.E., Kaplan, B.H., James, S.A.. Wagner. E.H.. Schoenback. V.J.. Grimson. R.. Heyden. S., Tibblin, G.. and Gehlbach, S.H. (1983). The epidemiologic evidence for a relationship between social support and health. American Journal of Epidemiology. 117. 521-537. Brody. J.A.and Schneider. E.L. (1986). Diseases and disorders o f aging: an hypothesis. Journal o f Chronic Diseases. 39, 871-876. Brown, G.K., Wallston. K.A.. and Nicassio, P.M. (1989). Social support and depression in rheumatoid arthritis: a one-year prospective study. Journal o f Applied Social Psychology, .19, 1164-1181. Brown, S.C. (1993). Revitalizing "Handicap’’ for disability research. Journal of Disability Policy Studies. 4, 57-76. Bruett, T.L., and Overs, R.P. (1968). A critical review of 12 ADL scales. Physical Therapy. 49, 859-861. Buchner, D.M., Beresford, S.A., Larson, E.B., LaCroix, A.Z., and Wagner, E.H. (1992). Effects o f physical activity on health status in older adults. II Intervention studies. Annual Review o f Public Health. 13,469-488. Buchner, D.M. and Wagner, E.H. (1992). Preventing frail health. Clinical Geriatric Medicine, 8, 1-17. Bureau o f Social Science Research, Inc. (1981). Toward a comprehensive data system on the demographic and epidemiological characteristics o f the handicapped population. Washington, DC: Bureau o f Social Science Research, Inc. 2 5 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Campbell, A.J.. Reinken, J., Allen, B.C., and Martinez, G.S. (1981). Falls in old age: A study o f frequent and related factors. Age and Aging. 10. 264-270. Campbell, A.J.. Borrie, M.J., and Spears, G.F. (1989). Risk factors for falls in a community based prospective study o f people 70 years and older. Journal of Gerontology: Medical Sciences. 44, M l 12-MI 17. Carter. R., Coast, J.R., and Idell, S. (1992). Exercise training in patients with chronic obstructive pulmonary disease. Medicine and Science in Sports and Exercise. 24(3), 281-291. Cassel. J. (1976). The contribution o f social environment to host resistance. The Fourth Wade Hampton Frost Lecture. American Journal o f Epidemiology, 104, 107-123. Cassileth, B.R., Lusk, E.J.. Strouse, T.B.. Miller, D.S., Brown, L.L.and Cross, P.A. (1985). A psychological analysis o f cancer patients and their next o f kin. Cancer, 55, 72-76. Castillo, C.S., Starkstein. S.E., Federoff, J.P., Price, T.R., and Robinson. R.G. (1993). Generalized anxiety disorder after stroke. Journal of Nervous and Mental Diseases. 181. 100-106. Cattell, R.B. (1943). The measurement o f adult intelligence. Psychological Bulletin. 40, 153-193. Cerella, J. (1985). Information-processing rates in the elderly. Psychological Bulletin. 98. 67-83. Chapireau, F. (1994). The Conceptual Framework of the International Classification o f Impairments, Disabilities and Handicaps ( ICIDH). Strasbourg Council o f Europe Press. Chapireau, F. and Colvez, A. (1998). Social disadvantage in the international classification o f impairments, disabilities, and handicap. Social Science and Medicine, 47 (1), 59-66. Christiansen, C. (1993). Continuing challenges of functional assessment in rehabilitation: recommended changes. American Journal of Occupational Therapy, 47, 258-259. Clark, D.O., Stump. T.E.. and Wolinsky, F.D. (1998). Predictors o f onset o f and recovery from mobility difficulty among adults aged 51-61 years. American Journal o f Epidemiology, 148, 63-71. Cobb, S. (1976). Social support as a moderator o f life stress. Psychosomatic M edicine. 38,300-314. 25 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cody, R.B., and Smith, J.K. (1991). Applied Statistics and the SAS Programming Language, third edition (p. 136-159), Prentice Hall. New Jersey. Cohen. S.. and Wills, T. (1985). Stress, social support and the buffering hypothesis. Psychology Bulletin. 98, 310-357. Colsher, P.L., Wallace, R.B. (1990). Elderly men with histories of heavy drinking: correlates and consequences. Journal of Studies on Alcohol. 5L 528-535. Colverz, A., Robine, J.M. (1986). Problems encountered in using the concepts of impairment, disability and handicap in a health assessment survey o f the elderly in upper Normandy. International Rehabilitative Medicine. 8. 18. Cook, N.R.. Evans. D.A.. Scherr, P.A., Speizer. F.E.. Taylor. J.O., and Hennekens, C.H. (1991). Peak expiratory flow rate and 5-year mortality in an elderly population. American Journal of Epidemiology. 133. 784-794. Cooper, J.E. (1980). The description and classification o f social disability by means o f a taxonomic hierarchy. Acta Psychiatric Scandinavia, 27. 140-146. Cooper. R.S. (1993). Health and the social status of blacks in the United States. Annals of Epidemiology. 3, 137-44. Creditor, M.C. (1993). Hazards of hospitalization o f elderly. Annals of Internal Medicine. 118.219-223. Cress. M.E.. Schechtman, K.B.. Mulrovv, C.D.. Fiatarone, M.A., Gerety. M.B.. and Buchner, D.M. (1995). Relationship between physical performance and self perceived physical function. Journal of the American Geriatric Society. 43,93- 1 0 1 . Crimmins, E.M.. Saito. Y.. and Ingegneri D.L. (1989). Changes in life expectancy and disability-free life expectancy in the United States. Population and Development Review. L5, 235-267. Crimmins. E.M., and Saito, Y. (1993). Getting better and getting worse: Transitions in functional status among older Americans, Journal o f Aging Health. 5, 3-36. Crimmins, E.M., Saito, Y., and Reynolds, S.L. (1997). Trends in the Prevalence and Incidence o f Disability Among Older Americans From Two Sources: The LSOA and the NHIS, Journal of Gerontology. Social Sciences, 52B. S59-S71. Croog, S., Lipson, A., and Levine, S. (1972). Helping patterns in severe illness: The roles o f kin network, non-family resources and institutions. Journal o f Marriage and the Family. 34. 3241. 2 5 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dartigues, J.F., Gagnon, M., Letenneur, L., Barberger-Gateau, P., Commenges, D., Evaldre. M.. and Salamon. R. (1992). Principal lifetime occupation and cognitive impairment in a French elderly cohort (PAQUID). American Journal of Epidemiology, 135, 981-988. Deaux, K. (1984). From individual differences to social categories: Analysis of a decade’s research on gender. American Psychologist. 39, 105-116. Deaux. K.. and Lewis, L.L. (1984). Structure of gender stereotypes: Interrelationships among components and gender labels. Journal o f Personality and Social Psychology. 46. 991-1004. De Boer, M.F., McCormick, L.K.. Pruyn, J.F.A.. Ryckman. R.M.. and Van den Borne, B.W. (1999). Physical and psychosocial correlates o f head and neck cancer: A review o f the literature. Otolarvngologv-Head and Neck Surgery, 120 (3), 427- 436. Deeg, D.J.H. (1993). Sex differences in IADL in the Netherlands: Functional and situational disability. In J.M. Robine, C.D. Mathers. M.R. Bone, and I. Romieu (Eds.), Calculation o f health expectancies: Harmonization, consensus achieved and future perspectives (pp. 203-213). Montrouge/Lond: Colloque INSERM/John Libbey Eurotext Ltd.. vol. 226). Deeg, D.J.H. (1994). Performance tests of physical disability. In D.J.H. Deeg and M. Westendorp-de Seriere (Eds.). Autonomy and well-being in the aging population. Report for the Longitudinal Aging Study. Amsterdam 1992-1993 (pp. 21-29). Amsterdam, The Netherlands: VU University Press. Dey, J.. Misra. A.. Desai. N.G., Mahapatra. A.K., and Padma, M.V.(1997). Cognitive function in younger type II diabetes. Diabetes Care. 20. 32-35. DiMatteo. M.R.. and Hays. R. (1981). Social support and serious Illness. In B.H. Gottlieb (Ed.). Social Networks and Social Support. Beverly Hills. CA: Sage. Donaldson, S.W.. Wagner, C.C.. and Gresham, G.E. (1973). A unified ADL evaluation form. Archives o f Physical Medicine and Rehabilitation, 54, 175-185. Dorevitch, M.I., Cossar. R.M., Bailey. F.J.. Bisset. T.. Lewis.S.J., and Wise, L.A., MacLennan, W.J. (1992). The accuracy o f self and informant ratings o f physical functional capacity in the elderly. Journal o f Clinical Epidemiology, 45(7). 791 - 798. Duncan. P.W. (1994). Stroke Disability. Physical Therapy. 74 (5). 399-407. 253 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Duncan. P.W.. and Studenski, S.. (1994). Balance and gait measures. In M.P. Lawton. J.A. Teresi (Eds.). Annual Review of Gerontology and Geriatrics, (pp.76-93) Springer Publishing Company. Inc.. New York. Duncan, P.W., Samsa, G.P., Weinberger, M., Goldstein, L.B., Bonito, A., Witter, D.M.. Enarson, C.. Matchar, D. (1997). Health status o f individuals with mild stroke. Stroke. 28. 740-745. Dunlop, D., Hughes, S.L.. and Manheim, L.M. (1997). Disability in Activities o f Daily Living: Patterns o f change in a hierarchy o f disability. American Journal o f Public Health. 81(3), 378-383. Eagly. A.H. (1987). Sex difference in social behavior: A social-role interpretation. Hillsdale, NJ: Lawrence Erlbaum. Elam. J.T., Graney, M.J., Beaver. T., El Derwi, D.. Applegate. W.B., and Miller. S.T.(1991). Comparison o f subjective ratings o f function with observed functional ability of frail older persons. American Journal of Public Health, 8L 1127-1130. Ell, K.. Nishimoto, R., Morvay, F.. Mantell, J., and Hamovitch, M. (1989). A longitudinal analysis o f psychological adaptation among survivors of cancer. Cancer. 63,406- 413. Endler. N.S.. and Parker, J.D.A. (1990). Multidimensional assessment o f coping: A critical evaluation. Journal o f Personality and Social Psychology, 58 (5). 844-854. Ensrud. K.E.. Nevitt. M.C.. Yunis, C.. Cauley. J.A.. Seeley. D.G.. Fox. K.M., and Cummings. S.R. (1994). Correlates o f impaired function in older women. Journal o f the American Geriatric Society. 42,481-489. Established Populations for Epidemiologic Studies of the Elderly data book. Washington, DC: GPO. 1990. (National Institute o f Health publication no (NIH)90-495). Ettinger. W.H., Fried. L.P.. Harris. T.. Shemanski. L.. Schulz. R.. Robbins. J. (1994). Self-reported causes o f physical disability in older people: The Cardiovascular Health Study. Journal o f the American Geriatric Society. 42, 1035-1044. Evans, D.A. (1993). Methodological issues in population studies o f physical function. In D.J.H. Deeg, C.P.M. Knipscheer and Van Tilburg (Eds.) Autonomy and well being in the agins population. Concepts and design o f the Longitudinal Aging Study Amsterdam (pp. 151-150). Bunnik, The Netherlands: Netherlands Institute o f Gerontology. Feinstein, A.R., Josephy, B.R., and Wells, C.K. (1986). Scientific and clinical problems in indexes o f functional disability. Annals o f Internal Medicine, 105.413-420. 254 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Femia, E.E. (1997). Predicting change in activities o f daily living: a longitudinal study of the oldest old in Sweden. Journal o f Gerontology: Psychological Sciences. 52B (6). P294-P302. Ferrer. M.. Lamarca. R... Orfila, F., and Alonso, J. (1999). Comparison o f performance- based and self-rated functional capacity in Spanish elderly. American Journal of Epidemiology. 149 (3). 228-235. Ferrucci. L.. Guralnik, J.M., Baroni, A., and Tesi, G. (1991). Value o f combined assessment o f physical health and functional status in community-dwelling aged: A prospective study in Florence, Italy. Journal o f Gerontology: Medical Sciences. 46. M52-56. Ferrucci, L., Guralnik, J.M., Pahor. M., Chiara.-Corti, M.C., and Havlik, R.J. (1997). Hospital diagnoses. Medicare charges, and nursing jhom e admissions in the year when older persons become severely disabled. Journal o f the American Medical Association. 277, 728-734. Fiatarone, M.A.. O'Neil, E.F., Ryan, N.D., Clements. K.M., Solares. G.R.. Nelson. M.E.. Roberts, S.B.. Kehayias. J.J., Lipsitz, L.A., and Evans. W.J. (1994). Exercise training and nutritional supplementation for physical frailty in very elderly people. New England Journal of Medicine, 330. 1769-1375. Fillenbaum. G.G. (1985). Screening the elderly: A brief instrumental activities o f daily living measure. Journal of the American Geriatrics Society. 33, 698-706. Foley. D.J., Berkman, L.F.. Branch, L.G.. Farmer. M.E.. and Wallace. R.B. (1986). Physical functioning. In: Comoni-Huntley J, Brock D.B., Ostefeld A.M., Taylor J., and Wallace. R.B. (eds) Established Populations for Epidemiologic Studies of the Elderly. Resource Data Book. No. 86-2443. Bethesda. MD: National Institutes of Health, 56-94. Folkman, S.. and Lazarus, R.S. (1980). An analysis o f coping in a middle-aged community sample. Journal of Health and Social Behavior, 22, 219-239. Folstein, M.F.. Folstein, S., and McHugh, P.R. (1975). Mini Mental State Exam: a practical method for grading the cognitive state o f patients for clinicians. Journal o f Psychiatric Research. 12, 189-198. Ford, A.B., Folmar, S.J., Salmon, R.B., Medalie, J.H., Rov, A.W., and Galazka, S.S. (1988). Health and function in the old and very old. Journal of the American Geriatric Society. 36. 187-97. Forette, F., Boiler, F. (1991). Hypertension and the risk o f dementia in the elderly. American Journal o f Medicine. 90, 14S-19S. 2 5 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fougeyrollas, P. (1993a). Applications o f the Concept of Handicap o f the ICIDH and its Nomenclature. Council o f Europe Press. Fougeyrollas. P. (1993b). Documenting environmental factors for preventing the handicap creation process: Anthropological studies in Quebec Rehabilitation Programs. In Partners for Independence: Models That Work, pp. 550-9. Proceedings o f the First North American Regional Conference o f Rehabilitation International, Atlanta Georgia, October 1993. Fougeyrollas, P. (1995). Documenting environmental factors for preventing the handicap creation process: Quebec contributions relating to ICIDH and social participation o f people with functional differences. Disability and Rehabilitation. L7, 145-153. Freedman. V.A.. and Martin, L.G. (1998). Understanding trends in functional limitations among older Americans. American Journal of Public Health, 88. 1457-1462. Freedman. V.A.. and Martin. L.G. (1999). The role of education in explaining and forecasting trends in functional limitations among older Americans. Demography. 36(4). 461-473. Frey. W.D. (1984). Functional assessment in the 80*s. In Functional Assessment in Rehabilitation (eds.) A.S. Halpem, M.J. Fuhrer. Paul H. Brookes Publishing Co.. p. 11-44. Fried. L.P., Ettinger. W.H.. Lind, B., Newman, A.B.. and Gardin. J. (1994). Physical disability in older adults: A physiological approach. Journal o f Clinical Epidemiology. 47(7). 747-760. Friedland. J.. and McCool, M. (1987). Social support and psychosocial dysfunction after stroke: Buffering effects in a community sample. Archives o f Physical and Medical Rehabilitation. 68: 475-480. Fries. J.F., Spitz, P.. Kraines. R.G., and Holman, H.R. (1980). Measurement o f patient outcome in arthritis. Arthritis and Rheumatology. 23, 137-145 Fuch, D.P., and Mettlin, C. (1982). The role o f support in relation to recovery form breast surgery. Social Science and Medicine. .16, 91-98. Fuhrer, R., Antonucci, T.C., Gagnon, M., Dartigues, J.F.. Barberger-Gateau, P., and Alpervotich, A. (1992). Depressive symptomatology and cognitive functioning: an epidemiological survey in an elderly community sample in France, Psychological Medicine. 22,159-172. Fulton, J.P.. Katz. S., Jack. S.S., and Jack, S.S. (1989). Physical functioning o f the aged: United States, 1984. National Center for Health Statistics. Vital Health Statistics, series 10 Data From the National Health Survey. 167. 1-48. 2 5 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Galanos, A.N., Fillenbaum, G.G., Cohen, H.J.. and Burchett, B.M. (1994). The comprehensive assessment o f community dwelling elderly: why functional status is not enough. Aging Clinical and Experimental Research. 6, 343-352. Ganguli. M.. Seaberg, E.. Belle, S. Fischer, L.. and Kuller, L.H. (1993). Cognitive impairment and the use o f health services in an elderly rural population: to mo VIES project. Journal o f the American Geriatric Society, 41., 1065-1070. Garrity. T.F. (1973). Vocational adjustment after first myocardial infarction: Comparative assessment of several variables suggested in the literature. Social Science and Medicine. 7. 705-717. Gerety. M.A.B., Mulrow, C.D., Tuley, M.R.. Hazuda, H.P., Lichtenstein. M.J.. Bohannon. R., Kanten, D.N.. O ’Neil. M.B.. and Gorton. A. (1993). Develment and validation o f physical performance instrument for the functionally impaired elderly: The physical disability index (PDI). Journal of Gerontology: Medical Sciences. 49, M33-M28. Gill, T.M., Williams. C.S., and Tinetti. M.E. (1995). Assessing risk for the onset of functional dependence among older adults: the role o f physical performance. Journal o f American Geriatrics Society. 43. 603-609. Gill, T.M., Willimans, C.S.. Richardson, E.D. and Tinetti. M.E. (1996). Impairments in physical performance and cognitive status as predisposing factors for functional dependence among non-disabled older persons. Journal of Gerontology: Medical Sciences. 5L M283-M288. Glass. T., and Maddox. G.L. (1992). The quality and quantity o f social support: Stroke recovery as psycho-social transition. Social Science and Medicine. 4, 1249-1261. Gortner, S.R.. Jenkins, L.S. (1990). Self-efficacy and activity level following cardiac surgery. Journal of Advanced Nursing. 15. 1132-1138. Granger, C.V., Albercht. G.. and Hamilton, B.B. (1979). Outcome of comprehensive medical rehabilitation: Measurement by PULSES Profiles and Barthel Index. Archives o f Physical Medical Rehabilitation. 60. 145-154. Granger, C.V., Hamilton. B.B., Keith, R.A., Zielezny, M., Sherwin, F.S. (1986). Advances in functional assessment for medical rehabilitation. Topics in Geriatric Rehabilitation. i(3). 59-74. Grasel, E. (1994). Non-pharmacological intervention strategies on aging process: Empirical data on mental training in ‘normal’ older people and patients with mental impairment. Archives o f Gerontology and Geriatrics. Supplement. 4 , 91- 98. 2 5 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Graves. A.B., Van Duijn. C.M., Chandra. V.. Fratiglioni, L.. Heyman, A.. Jorm, A.F., Kokmen, E., Kondo. K.. Mortemer. J.A.. and Rocca, W.A. (1991). Alcohol and tobacco consumption as risk factors for Alzheimer’s disease: a collaborative re analysis o f case-control studies. International Journal o f Epidemiology. 20 (Supplement 2). S48-S57. Greiner, P. A., Snowdon, D.A., and Greiner. L.H. (1996). The relationship of self-rated function and self-rated health to concurrent functional ability, functional decline, and mortality: Findings from the Nun Study. Journal of Gerontology: Social Sciences. 51B (5), S234-S241. Griffiths, R.A., Good, W.R., Watson, N.P., O ’Donnell. H.F., Fell, P.J., Shakespeare. J.M. (1987). Depression. Dementia and Disability in the Elderly. British Journal of Psychiatry. 150.482-493. Grimby. G.. Finnstam. J., and Jette, A. (1988). On the application o f the WHO handicap classification in rehabilitation. Scandinavian Journal o f Rehabilitative Medicine. 20. 93-98. Gryfe, C.I.. Amies, A., and Ashley. M.J. (1977). A longitudinal study o f falls in the elderly population. I. Incidence and morbidity. Age and Aging. 6. 201-210. Guccione, A.A.. Felson D.T.. Anderson, J.J., Anthony, J.M.. Zhang. Y.Q., Wilson, P.W.F., Kelly-Hayes, M., Wolf, P.A.. Kreger. B.E., and Kannel. W.B. (1994). The effects o f specific medical conditions on the functional limitations o f elders in the Framingham-Study. American Journal of Public Health. 84 (3), 351-358. Guo. Z., Fratiglioni, L.. Winblad. B.. and Vinanen. M. (1997). Blood pressure and performance in the mini-mental state examination in the very old. Cross-sectional and longitudinal data from the Kungsholment Project. American Journal o f Epidemiology. 145 (12). 1106-1113. Guralnik, J.M.. Branch. L.G.. Cummiongs, S.R., and Curb. J.D. (1989). Physical performance measures in aging research. Journal o f Gerontology: Medical Sciences. 44, M 141-M 146. Guralnik, J.M., and Kaplan, G.A. (1989). Predictors of healthy aging: Prospective evidence from the Alameda County Study. American Journal o f Public Health, 79, 703-708. Guralnik, J.M., and LaCroix. A.Z. (1992). Assessing physical function in older populations. In Wallace, R.B. and Woolson. R.F. (Eds.), The Epidemiological Study o f the Elderly. Oxford University Press, New York, 159-181. 258 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Guralnik, J.M., LaCroix, A.Z.. Abbott, R.D., Berkman, L.F., Satterfield, S., Evans, D.A.. and Wallace, R.B. (1993). Maintaining mobility in late life. I. Demographic characteristics and conditions. American Journal of Epidemiology, 137, 845-57. Guralnik. J., Simonsick, E.M.. Ferrucci, L., Glynn. R.J., Berkman, L.F.. Blazer, D.G., Scherr, P.A., and Wallace, R.B. (1994). A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction o f mortality and nursing home admission. Journal o f Gerontology: Medical Sciences. 49. M85-M94. Guralnik, J.M., Ferrucci, L.. Simonsick. E.M. Salive, M.E.. and Wallace. R.B. (1995). Lower-extremity function in persons over the age o f 70 years as a predictor of subsequent disability. New England Journal of Medicine. 332. 556-561. Guyatt. G.H., Pugsley. S.O., Sullivan. M.J.. Thompson, P.J.. Berman, L.B., Jones. N.L., Fallen, E.L.. and Taylor. D.W. (1984). Effect of encouragement on walking test performance. Thorax. 39. 818-822. Haan, M.N., and Weldon. M. (1996). The influence o f diabetes, hypertension and stroke on ethnic-differences in physical and cognitive-functioning in an ethnically diverse older population. Annals o f Epidemiology, 6 (5). 392-398. Haber. L.D. (1985). Trends and demographic studies on programs for disabled persons. In Perlaman, LG.. Austin, G.F. (Eds.). Social Influences in Rehabilitation Planning: Blueprint for the 2 P t Century. National Rehabilitation Association. Alexandria. VA. p.27. Haber. L.D. (1990). Issues in the definition of disability and the use o f disability survey data. In Disability Statistics. An Assessment: Report o f a Workshop. Washington. D.C.: National Academy Press. Halbertsma, J. (1992a). Classification o f health problems: The case o f the ICIDH. In: Proceedings of the Fifth International Conference on System Science in Health Care. Prague, Czechoslovakia: Omnipress Publishers. 1105-1108. Halbertsma. J. (1992b). Classification o f disabilities: State o f affairs and proposals. Presented at the Second Meeting of the Updating o f the ICIDH, Zoetermeer. The Netherlands. Hall, R.G., and Charming, D.M. (1989). Age, pattern o f consultation, and functional disability in elderly patients in one general practice. British Medical Journal. 301, 424-428. Hamilton, S., and Fagot, B.l. (1988). Chronic stress and coping styles: A comparison o f male and female undergraduates. Journal o f Personality and Social Psychology, 55(5), 819-823. 2 5 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hanis, T., Kovar, M.G., Suzman, R.. Kleinman, J.C., and Fledman, J.J. (1989). Longitudinal study o f physical ability in the oldest old. American Journal of Public Health. 79. 698-702. Harper. C.H.. and Lyles, Y.M. (1988). Physiology and complications o f bed rest. Journal of The American Geriatric Society. 36. 1047-1054. Harris, B.A.. Jette. A.M.. Campion E.W., and Cleary, P.D. (1986). Validity of self-report measures of functional disability. Topics in Geriatric Rehabilitation. !(3), 31-41. Harris, T., Kovar, M.G., Suzman, R.. Kleinman, J.C., and Feldman, J.J. (1989). Longitudinal study o f physical ability in the oldest-old. American Journal of Public Health. 79 (6). 698-702. Haworth. R.J.. and Hollings. E.M. (1979). Are hospital assessments of daily living valid? International Journal of Rehabilitative Medicine. L 59-62. HDFP-Hypertentions detection and follow-up program cooperative group (1978). Variability of blood pressure and results o f screening in HDFP program. Journal o f Chronic Disease. 31, 651 -657. Heeren. T.J.. Lagady. A.M.. Beck. W.C.. Rooymans. H.J.. and Hijmans. W. (1990). Reference values for the Mini-Mental State Exam (MMSE) in octo an nonagenarians. Journal of American Geriatric Society. 38, 1093-1096. Hedrick, S.C.. Katz. S., and Stroud. M.W. (1981). Patient assessment in long term care: Is there a common language? Aged Care and Services Review. 2 (4), 3-19. Heerkens. Y.F.. Brandsma. J.W., Lakerveld-Heyl, K., and Van Ravensberg, C.D. (1994). Impairments and disabilities— the difference: Proposal for adjustment o f the international classification o f impairments, disabilities, and handicaps. Physical Therapy. 74 (5). 430-445. Herbert. R„ Brayne. C.. and Spiegelhalter. D. (1999). Factors associated with functional decline and improvement in a very elderly community dwelling population. American Journal of Epidemiology. 150 (5), 501-510. Hirsch, B.J. (1980). Natural support systems and coping with major life changes. American Journal of Community Psychology, 8,159-172. Hirsch, C.H., Sommers, L., Olsen A., Mullen L.. and Winograd. C.H. (1990). The natural history o f functional morbidity in hospitalized older patients. Journal o f the American Geriatric Society. 3 8 ,1296-1303. Ho, S.C. (1991). Health and social predictors o f mortality in an elderly Chinese cohort. American Journal o f Epidemiology. 133. 907-921. 260 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hoenig, H.M., and Rubenstein, L.Z. (1991). Hospital associated de-conditioning and dysfunction. Journal of the American Geriatric Society. 39. 220-221. Hogue, C.C., Studenshki, S., and Duncan P.W. (1990). Assessing mobility: The first step in falls prevention. In S.G. Funk, E.M. Tomquist, M.P. Chapagne, L.A. Copp and R. A. Weise, (Eds.). Key aspects of recovery: Improving nutrition, rest, and mobility, (pp. 275-280) New York: Springer. Horak, F.B., Shupert, C.L.. and Mirka, A. (1989). Components o f postural dyscontrol in the elderly: A review. Neurology of Agine. 10, 727-738. House, J.S. (1981). Work stress and social support. Reading, MA: Addison-Wesley Publishing. House. J.S. (1990). Americans’ Changing Lives: Wave I. 1986 (computer file). Ann Arbor. MI: Survey Research Center (producer); 1989. Ann Arbor, MI: Inter- University Consortium for Political and Social Research (distributor); 1990. Houx. P.J.. and Jolles, J. (1993). Age-related decline o f psychomotor speed: Effects of age. brain health, sex and education. Perceptual Motor Skills. 76. 195-211. Hurwicz. M. and Berkanovic. E. (1993). The stress process in rheumatoid arthritis. Journal o f Rheumatology. 20, 1836-1844. Hyman. M.D. (1971). Disability and patients' perceptions o f preferential treatment: Some preliminary findings. Journal of Chronic Disease. 24. 329-342. Hyman. M.D. (1975). Social psychological factors affecting disability among ambulatory patients. Journal o f Chronic Disease, 28, 199-216. Inouye. S.K., Albert, M.S., Mohs, R.. Sun K., and Berkman, L.F. (1993). Cognitive performance in a high functioning community-dwelling elderly population. Journal o f Gerontology: Medical Sciences. Medical Sciences. 48, M 146-151. Institute o f Medicine (1991). Disability in America: Toward a National Agenda for Prevention. Washington. DC National Academy Press. Jagger, C., Clarke, M., and Cook, A.J. (1989). Mental and physical health o f elderly people: five-year follow-up o f a total population. Age and Aging. 18, 77-82. Jebsen .R.H., Taylor, N, Trieschmann, R.B., Trotter, M.J.. and Howard, L.A. (1969). An objective and standardized test o f hand function. Archives o f Physical Medicine and Rehabilitation. 50. 311-319. Jette, A.M. (1980a). Functional capacity evaluation: An empirical approach. Archives of Physical Medicine and Rehabilitation, o i (2), 85-89. 261 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Jette, A.M. (1980b). Functional status index: Reliability o f a chronic disease evaluation instrument. Archives o f Physical Medicine and Rehabilitation. 6 ! (9). 395-401. Jette, A.M., and Branch, L.G. (1981). The Framingham Disability Study: II. Physical disability among the aging. American Journal o f Public Health. 7L 1211-1216. Jette. A.M.. and Branch. L.G. (1985). Impairment and disability in the aged. Journal of Chronic Diseases. 38, 59-65. Jette A.M., Davies. A.R., Cleary, P.D., Calkins, D.R.. Rubenstein, L.V.. Fink, A., Kosecoff. J., Young, R.J.. Brook, R.H., and Delbanco, T.L. (1986). The functional status questionnaire reliability and validity when used in primary care. Journal of General Internal Medicine, i , 143-149. Jette. A.M. (1987). The Functional Status Index: reliability and validity of a self-report functional disability measure. Journal o f Rheumatology. \4 (Suppl. 15). 15-19. Johnson, R.J., and Wolinsky, F.D. (1993). The structure of health status among older adults: disease, disability, functional limitations, and perceived health. Journal of Health and Social Behavior. 34. 105-21. Kay. D.W.K., Henderson. A.S.. Scott. R.. Wilson, J.. Rickwod, D.. and Grayson. D.A. (1985). Dementia and depression among the elderly living in the Hobart community: the effect of the diagnostic criteria on the prevalence rates. Psychological Medicine. 15. 771-788. Kaplan. E.. Goodglass, H„ and Weintraub. S. (1983). Boston Naming Test. Philadelphia. PA: Lea and Febiger. Kaplan, G.A., Seeman. T.E., Cohen. R.D., Knudsen. L.P., and Guralnik. J. (1987). Mortality among the elderly in the Alameda County study: Behavioral and demographic risk factors. American Journal o f Public Health. 77, 307-312. Kaplan. G.A., Strawbridge. W.J.. Camacho. T.. and Cohen. R.D. (1993). Factors associated with change in physical functioning in the elderly: A six year prospective study. Journal o f Aging and Health. 5 (1), 140-153. Kaplan, G.A.. Wilson, T.W., Cohen R.D. Kauhanen, J.. Wu, M., and Salonen, J.T. (1994). Social functioning and overall mortality : prospective evidence form the Kuopio Ischemic Heart Disease Risk Factor Study. Epidemiology. 5,495-500. Kaplan. R.M., Riese, A.L.. Prewitt, L.M.. and Eakin, E. (1994). Self efficacy expectations predict survival for patients with chronic obstructive pulmonary disease. Health Psychology. 13, 366-368, 262 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Katz. S., and Staff o f the Benjamin Rose Hospital (1959). Multidisciplinary studies o f illness in aged persons: II. A New classification o f functional status in activities o f daily living. Journal o f Chronic Disability. 9, 55-62. Katz, S.. Ford. A.B.. Moskowitz, R.W., Jackson, B.A., and Jaffe. M.W. (1963). Studies o f illness in the aged. Journal o f the American Medical Association. 185. 914- 919. Katz. S., Downs, T.D., Cash, H.R., and Grotz. P.C. (1970). Progress in the development o f an index o f Activities o f Daily Living. Gerontologist, 10, 20-30. Katz. S., Downs, T.D., Cash, H.R.. and Grotz. R.C. (1979). Progress in the development o f an index o f ADL. Gerontologist, 10, 20-30. Katz. S., and Stroud, M.W. (1989). Functional assessment in geriatrics. A review of progress and directions. Journal of American Geriatric Society. 37. 267-271. Kaufert. J.M.. Green, S., Dunt. D.R., Corkhill, R., Creese. A.L., and Locker, D. (1979). Assessing functional status among elderly patients: a comparison o f questionnaire and service provider ratings. Medical Care. 17, 807-817. Kaufmann. P.G.. McMahon. R.P.. Becker. L.C.. Bertolet, B., Bonsall. R.. Chaitman. B.. Cohen. J.D.. Forman. S., Goldberg, A.D.. Freedland, Kl, Ketterer, M.W.. Krantz. D.S.. Pepine. C.J., Raczynsko, J.. Stone, P.H.. Taylor. H.. Knatterud. G.L., and Sheps. D.S. (1998). The psychophysiological investigations o f myocardial- ischemia (PIMI) study - objective, methods, and variability o f measures. Psychosomatic Medicine. 60 (1). 56-63. Keefover, R.W. (1998). Aging and Cognition. Neurologic Clinics o f North America. ]6 (3), 635-648. Keil. J.E., Gazes, P.C.. Sutherland, S.E., Rust. P.F., Branch, L.G., and Tyrolder. H.A. (1989). Predictors o f physical disability in elderly Blacks and Whites o f the Charleston Heart Study. Journal o f Clinical Epidemiology. 42, 521-529. Keith. R.A. (1981). The status o f functional assessment— instruments in medical rehabilitation. San Diego, CA: American Congress o f Rehabilitation. Kelly-Hayes, M„ Jette, A.M., Wolf, P.A., D’Agostino. R.B., and Odell, P.M. (1992). Functional limitations and disability among elders in the Framingham Study. American Journal o f Public Health. 82, 841-845. Kempen, G.I.J.M., and Suurmeijer, T.P. (1990). The development o f a hierarchical polychotomous ADL-IADL scale for non-institutionalized elders. Gerontologist. 30,497-502. 2 6 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Kempen. and Suurmeijer, T.P. (1991). Professional Home Care for the Elderly: An Application of the Andersen-Newman Model in the Netherlands. Social Science and Medicine. 33 (9). 1081-1089. Kempen. G.I.J.M., Steverink, N., Ormel, J., and Deeg, D.J.H. (1996). The assessment of ADL among frail elderly in an interview survey: Self-report versus performance based tests and determinants of discrepancies. Journal o f Gerontology: Psychological Sciences. 51B(5). P254-P260. Kempen. G.I.J.M., Van Heuvelen, M.J.G., Van Den Brink. R.H.S.. Kooijman. A.C., Klein. M.. and Houx. P.J.. Ormel. J. (1996). Factors affecting contrasting results between self-reported and performance based levels o f physical limitations. Age and Aging. 25,458-464. Kempen, G.I.J.M., Sullivan, M„ Van Sonderen, E., and Ormel, J. (1999). Performance- Based and Self-Reported Physical Functioning in Low-Functioning Older Persons: Congruence of Change and the Impact o f Depressive Symptoms. Journal o f Gerontology: Psychological Sciences. 54B (6). P380-P386. Kempen, G.I.J.M., Van Sonderen, E.. and Ormel, J. (1999). The impact o f psychological attributes on changes in disability among low functioning older persons. Journal o f Gerontology: Psychological Sciences, 54B (1). P23-P29. Kivela. S.L. (1984). Measuring disability— do self-ratings and service provider ratings compare? Journal o f Chronic Disease. 37. 1 15-123. Kessler. H.H. (1931). Accidental injuries. Philadelphia: Lea and Febiger. Kivinen, P., Sulkava, R., Halonen. P., and Nissinen. A. (1998). Self-reported and performance-based functional status and associated factors among elderly men: The Finnish cohorts o f the seven countries study. Journal o f Clinical Epidemiology. 5J. ( 12), 1243-1252. Klein. R.M.. and Bell, B. (1982). Self-care skills: Behavioral measurement with Klein- Bell scale. Archives of Physical Medicine and Rehabilitation, 63, 335-338. Kostreski, F.. and Zwillich, T. (1999). No disability for obesity. Family Practice News. 29, 57. Kovar, M.G. (1988). Aging in the eighties, people living alone - two years later. Advance Data, 146, April 4. Koyano, W., Shibata, H., Haga, H., and Suyama, Y. (1986). Prevalence and outcome o f low ADL and incontinence among the elderly: Five years follow-up in a Japanese urban community. Archives o f Gerontology and Geriatrics. 5, 197-206. 2 6 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Kretchmer, N., Beard, J.L., and Carlson, S. (1996). The role o f nutrition in the development of normal cognition. American Journal o f Clinical Nutrition. 63, 997S-1001S. Krol, B., Sanderman, R.. Suurmeijer, T.P.B.M., Doeglas, D.. Van Sonderen. E., Van Rijswijk, M., Van Leeuvven. M., and Van den Heuvel, W.J.A. (1998). Early rheumatoid arthritis, personality and psychological status: A follow-up study. Psychology and Health. 13, 35-48. Kubzansky, L.D., Berkman, L.F.. Glass, T.A., and Seeman. T.E. (1998). Is educational- attainment associated with shared determinants of health in the elderly : Findings from the MacArthur studies of successful aging. Psychosomatic Medicine. 60 (5). 578-585. Kuriansky, J. and Gurland, B. (1976). The performance test o f activities o f daily living. International Journal o f Aging and Human Development, 7. 343-352. Kurtz. M.E., Kurtz, J.C.. Given. C.W.. Given, B.A. (1993). Loss o f physical functioning among patients with cancer: A longitudinal view. Cancer Practice, i (4). 275- 281. Kurtz. M.E.. Given. B.. Kurtz. J.C., and Given, C.W. (1994). The interaction o f age. symptoms, and survival status on physical and mental health o f patients with cancer and their families. Cancer. 74 (7), 2071-2078. Kurtz. M.E., Kurtz. J.C.. Stommel, M.. Given. C.W.. Given, B. (1999). The influence of symptoms, age. comorbidity and cancer site on physical functioning and mental health o f geriatric women patients. Women and Health, 29 (3). 1-12. LaCroix A.Z.. Guralnick, J.M., Berkman. L.F., Wallace.R.B., and Satterfield.S. (1993). Maintaining mobility in late life. II: Smoking, alcohol consumption, physical activity, and body mass index. American Journal of Epidemiology, 137. 858-869. Lammi, U.K., Kivela. S.L.. Nissenen. A.. Pekkanen. J., and Punsar. S. (1989). Functional capacity and associated factors in elderly Finnish men. Scandinavian Journal of Social Medicine. 17. 67-75. Lammi, U.K., Kivela, S.L.. Nissenen, A.. Punsar. S.. Pushk, P, and Karvonen. M. (1989). Predictors o f disability in elderly Finish men: A longitudinal study. Journal of Clinical Epidemiology. 42, 1215-1225. Lamont, C.T., Sampson, S., Matthias, R., and Kane. R. (1983). The outcome of hospitalization for acute illness in the elderly. Journal o f American Geriatric Society. 31., 282-288. 2 6 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Landerman, R.. George, L.K., Campbell, R.T.. and Blazer, D.G. (1989). Alternative models o f the stress buffering hypothesis. American Journal o f Community Psychology. 17, 625-642. LaPlante. M.P. (1990). Who counts as having a disability? Musings on the meaning and prevalence o f disability. Disability Studies Quarterly. 10, 15-17. Laukkanen. P., Era. P., Heikkinen, R.L., Suutama, T.. Kauppinen. M.. and Heikkinen. E. (1994). Factors related to carrying out everyday activities among elderly people aged 80. Aging Clinical Experimental Research. 6, 433-443. Launer, L.J.. Harris, T.. Rumpel. C., and Madans. J. (1994). Body mass index, weight change and risk of mobility disability in middle age and older women. Journal of the American Medical Association. 271. 1093-1098. Law. M., and Letts. I. (1989). A critical review of scales of activities o f daily living. American Journal of Occupational Therapy, 43. 522-528. Lawrence. P.S. (1958). Chronic illness and socioeconomic status (n E.G. Jaco (Ed.). Patients. Physicians, and Illness (pp. 37-49). New York: Free Press. Lawton. M.P. and Broody, E.M. (1969). Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist. 9, 179-186. Lawton. M.P. (1990). Aging and performance o f home tasks. Human Factors. 32, 527- 536. Lazarus, R.S. and Folkman. S. (1984). Stress. Appraisal and Coping. New York: Springer Publishing. Leavy, R.L. (1983). Social support and psychological disorder: A review. Journal of Community Psychology. J_L , 3-21. Letenneur. L., Jacqmin, H., Barberger-Gateau. P. Dartigues. J.F.. and Salamon. R. (1993). Cerebral and functional aging: first results on prevalence and incidence of the PAQUID cohort. Methods of Information in Medicine. 32. 249-251. Letenneur, L.. Commenges, D.. Dartigues, J.F.D., and Barberger-Gateau. P. (1994). Incidence o f dementia and Alzheimer’s disease in elderly community residents of south-western France. International Journal o f Epidemiology, 23, 1256-1261. Leveille, S.G., Guralnik. J.M., Ferrucci, L., Corti, M.C., Kasper. J., and Fried, L.P. (1998). Black/white differences in the relationship between MMSE scores and disability: The Women’s Health and Aging Study. Journal o f Gerontology: Psychological Sciences. Psychological Sciences. 53B. P201-P204. 266 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Lewis, C.E. (1966). Factors influencing the return to work o f men with congestive heart failure. Journal o f chronic disease. 19, 1193-1209. Lezak, M.D. (1983). Neuropsychological assessment. New York: Oxford University Press. Liao. A., McGee, D.L., Cao. G.. and Cooper. R.S. (1999). Black-White differences in disability and morbidity in the last years of life. American Journal of Epidemiology. 149. 1097-1103. Lichtenberg, P.A., Ross. T., Millis. S.R.. and Manning, C.A. (1995). The relationship between depression and cognition in older adults: A cross-validation study. Journal o f Gerontology: Psychological Sciences. SOB. P25-P32. Lindholm. L. (1990). Hypertension and ageing. Clinical Experiments Hypertension. L2 (5), 745-749. Linn. M.W., Hunter. K.I.. and Linn, B.S. (1980). Self-assessed health, impairment and disability in Anglo, Black and Cuban elderly. Medical Care. J_8. 282-288. Livingston, G., Hawkins, A.. Graham. N., Blizard, B.. and Mann, A. (1990). The Gospel Oak Study: prevalence rates of dementia, depression and activity limitation among elderly residents in inner London. Psychological Medicine. 20, 137-46. Loge. J.H.. Abrahamse. A.F.. Ekeberg, O.. and Kaasa, S. (1999). Reduced health-related quality o f life among Hodgkin's disease survivors: A comparative study with general population norms. Annals of Oncology. K) (I), 71-77. Lohman, T.G., Roche. A.F., and Martorell, R. (1988). Anthropometric Standardization Reference Manual. Champaign, IL: Human Kinetics. Long, J.D. (1999). Supreme Court issues several key ADA decisions. HR Focus, p. 3. Luteijn. F.. and Vanderploeg. F.A.E. (1983). GIT: Groninger Intelligence Test: Handleiding. Lisse. The Netherlands: Swets and Zeitlinger. MacFarlane. A.C. (1988). The International Classification o f Impairments. Disabilities, and Handicaps: Its usefulness in classifying and under-standing biopsychosocial phenomena. Australial and New Zealand Journal o f Psychiatry, 22, 31-42. Mahoney, F.J., and Barthel, D.W. (1965). Functional evaluation: The Barthel Index, Maryland Medical Journal. 14,61-65. 267 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mahoney, J.E., Sager, M.A., and Jalaluddin, M. (1998). New walking dependence associated with hospitalization for acute medical illness: Incidence and significance. Journal of Gerontology: Medical Sciences. 53A (4). M307-M312. Mahurin. R.K.. DeBettignies, B.H., and Pirozzolo. F.J. (1991). Staictured assessment of independent living skills: Preliminary report o f a performance-based measure of functional abilities in dementia. Journal of Gerontology: Psvchologica Sciences. 46, P58-P66. Manton, K.G., Corder, L.. and Stallard.E. (1997). Chronic Disability Trends in the Elderly U.S. Populations: 1982-1994. Proceedings o f the National Academy of Science. 94, 2593-2598. Mathias. S.. Mayak. U.S.L.. and Isaacs, B. (1986). Balance in elderly patients: The Get Up and Go Test. Archives o f Physical Medicine and Rehabilitation. 67, 387-389. Mausner, J.S., and Bahn. A.K. (1974). Epidemiology: An Introductory Text. W.B. Saunders Company, Philadelphia, pp. 242-247. Mazariegos, M., Wang, Z.M., Gallagher, D., Baumgartner. R.N., Allison. D.B.. Wang. J.. Pierson. R.N.. and Heymsfield. S.B.(1994). Difference between young and old females in the five levels of body composition and their relevance to the two- compartment medical model. Journal o f Gerontology: Medical Sciences. 49. M 201-208. Mazaux, J.M., Dartigues. J.F.. and Letenneur, L. (1995). Visuo-spacial attention and psychomotor performance in elderly community residents: Effects of age. gender and education. Journal of Clinical Experimental Neuropsychology. J_7. 71-81. McDowell, I. (1996). Measuring Health: A Guide to Rating Scales and Questionnaires, second edition. New York: Oxford University Press. McLeroy. K.R.. DeVellis. R.. DeVellis. B., Kaplan. B., and Toole. J. (1984). Social support and physical recovery in a stroke population. Journal of Social and Personal Relationships. L 395-413. McVey. L.J., Becker, P.M., Saltz, C.C., Feussner, J.R., and Cohen. H.J. (1989). Effects of geriatric consultation team on functional status o f elderly hospitalized patents. A randomized controlled clinical study. Annals o f Internal Medicine. 110, 79-84. Mendes de Leon. C.F., Seeman, T.E., Baker, D.I., Richardson, E.D., and Tinetti. M.E. (1996). Self-efficacy, physical decline, and change in functioning in community- living elders: A prospective study. Journal o f Gerontology: Social Sciences, 51B (4), S183-S190. 268 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Menec, V.H., and Chipperfield, J.G. (1997). Remaining active in later life— the role of locus o f control in senior’s leisure activity participation, health, and life satisfaction. Journal o f Aging and Health. 9, 105-125. Merrill, S.S., Seeman, T.E., Kasl, S.V., and Berkman, L.F. (1997). Gender differences in the comparison o f self-reported disability and performance measures. Journal o f Gerontology: Medical Sciences. 52, M19-M26. McAuley. E. (1993). Self-efficacv and the maintenance of exercise participation in older adults. Journal o f Behavioral Medicine, 16, 103-113. McBride. E.D. (1963). Disability evaluation and principles o f treatment o f compensable injuries (6lh ed.) Philadelphia: J.B. Lippincott. Molander. U. (1993). Urinary incontinence and related urogenital symptoms in elderly women. Acta Qbstet Gynecology Scandinavia Supplem ent, 158. 1-22. Mor. V.. Muiphy. J.. Masterson-Allen, S.. Willey. C.. Razmpour. A., Jackson, M.E., Greer, D., and Katz, S. (1989). Risk of functional decline among well elders. Journal o f Clinical Epidemiology. 49 (9), 895-904. Mor. V.. Murphy, J., Masterson-Allen, S., Willey. C.. Razmpour. A., Jackson, M.E., Greer, D., and Katz, S. (1989). Risk of functional decline among well elders. Journal o f Clinical Epidemiology, 42, 895-904. Mor. V., Wilcox. V.. Rakowski. W„ and Hiris, J. (1994). Functional transitions among the elderly: Patterns, predictions, and related hospital use. American Journal of Public Health. 84. 1274-1280. Morbidity and Mortality Weekly Report. (1994). Prevalence o f disabilities and associated health conditions- United States, 1991-1992.43, 730-735. Morris, J.C., Heyman, A., Mohs, R.C., Hughes, J.P., van Belle. G., Fillenbaum, G., Mellits, E.D., and Clark, C. (1989). The consortium to establish a registry for Alzheimer’s disease (CERAD). Part I: Clinical and neuropsychological assessment o f Alzheimer's disease. Neurology. 39, 1159-1165. Moss, M.. Albert, M., Butters, N., and Payne. M. (1986). Differential patterns o f memory loss among patients with Alzheimer’s disease, Huntington's disease, alcoholic Korsakoff s syndrome, Archives o f Neurology. 43. 239-246. Mossey, J.M., Mutran, E., Knott. K.. and Craik, R. (1989). Determinants o f recovery 12 months after hip fracture: The importance o f psychosocial factors. American Journal o f Public Health. 79, 279-286. 2 6 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Mulrow. C.D., Gerety, M.B., Cornell, J.E., Lawrence, V.A., Kanten, D.N. (1994). The relationship between disease and function and perceived health in very frail elders. Journal o f the American Geriatric Society, 42, 374-380. Mungall. I.P.F., and Hainsworth, R. (1979). Assessment o f respiratory function in patients with chronic obstructive airways disease. Thorax. 34, 254-258. Myers. A.M., Holliday, P.J., Harvey, K.A., and Hutchinson. K.S. (1993). Functional performance measures: Are they superior to self assessments? Journal of Gerontology: Medical Sciences, 48, M196-M206. Nagi, S.Z. (1965). Some conceptual issues in disability and rehabilitation . In M.B. Sussman (Eds.) Sociology and Rehabilitation, pp. 100-113. American Sociological Association. Washington. D.C. Nagi. S.Z. (1969). Disability and rehabilitation: Legal, clinical and self concepts and measurement. Columbus: Ohio State University Press. Nagi. S.Z.(1976). An epidemiology o f disability among adults in the United States. Milbank Memorial Fund Quarterly, 54> 439-468. Nagi. S.Z. (1977). The disabled and rehabilitation services: A national overview. American Rehabilitation. 2(5), 26-33. Nagi. S.Z. (1991). Disability concepts revisited: Implications for prevention. In A.M. Pope and A.R. Tarlov (Eds.) Disability in America: Toward a national agenda for prevention (pp. 309-327). Washington, D.C.: National Academy Press. Nakayama. H., Jorgensen. H.S.. Raaschou. H.O.. and Olsen, T.S. (1994). The influence o f age on stroke outcome: The Copenhagen Stroke Study. Stroke 25 (4), 808-813. National Center for Health Statistics. Health, United States. 1992. and Healthy People 2000 review. Hyattsville, MD. Public Health Service. 1993. (DHHS publication no (PHS) 93-1232). Nelson, H.D.. Nevitt. M.C.. Scott, J.C., Stone. K.L.. and Cummings. S.R. (1994). Smoking, alcohol and neuromuscular and physical function of older women. Journal o f the American Medical Association. 272. 1852-1831. Nevitt. M.C.. Cummings, S.R., Kidd, S., and Black, D. (1989). Risk factors for recurrent nonsyncopal falls, Journal of the American Medical Association, 261, 2663-2668. Newman, S.P., Fitzpatrick, R., Lamb, R., and Shipley. M. (1989). The origins of depressed mood in rheumatoid arthritis. Journal o f Rheumatology. 16, 740-744. 2 7 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Newsom. J.T., and Schulz, R. (1996). Social support as a mediator in the relation between functional status and quality o f life in older adults. Psychology and Aging, i ! (1), 34-44. Nickel. J.T. and Chirikos. T.N. (1990). Functional disability o f elderly patients with long term coronary heart disease: a sex stratified analysis. Journal of Gerontology. 45. S60-S68. Nikolaus. T.. Bach, M., Ostcr, P., and Schlierf. G. (1995). The timed test o f money counting: A simple method o f recognizing geriatric patients at risk for increased health care. Aging Clinical Experimental Research. 7, 179-183. Oberg, T„ Karsznia. A., and Oberg, K. (1993). Basic gait parameters: Reference data for normal subjects. 10-79 years. Journal o f Rehabilitation Research and Development. 30. 210-223. Office of Human Development Services, US Department o f Health and Human Services: The Need for Long-Term Care. A Chartbook of the Federal Council on Aging. Washington, DC. 1981. O'Loughlin J.L., Robitaille, Y.. Boivin. J.F.. and Suissa. S. (1993). Incidence o f and risk factors for falls and injurious falls among the community-dwelling elderly. American Journal o f Epidemiology. 137, 342-354. Olsen. O. (1993). Impact o f social network on cardiovascular mortality in middle aged Danish men. Journal of Epidemiology and Community Health. 47. 176-180. Olson. D.A.. and Shultz. K.S. (1994). Gender differences in the dimensionality o f social support. Journal o f Applied Social Psychology, 24 (14). 1221-1232. Orgogozo. J.M. (1994). The concepts o f impairment, disability and handicap. Cerebrovascular Disease. 4, 2. Ormel, J., Kempen. G.I.J.M., Penninx, B.W.J.W.. Brilman. E.I., Beekman. A.T.F.. and Van Sonderen. E. (1997). Chronic medical conditions and mental health status in older people: Disability and psychosocial resources medicate specific mental health effects. Psychological Medicine. 27. 1065-1077. Ormel. J.. Kempen. G.I.J.M.. Deeg, D.J.H., Brilman. I.. Van Sonderen. E.. Relyveld. J. (1998). Functioning, well-being, and health perception in late middle-aged and older people: Comparing the effects o f depressive symptoms and chronic medical conditions. Journal o f the American Geriatric Society. 46. 39-48. Overstall. P.W., Smith, A.N.. Imms. F.J., and Johnson. A.L. (1974). Falls in the elderly related to postural imbalance. British Medical Journal. I , 261-264. 271 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Palombo. V., Scurti, R.. Muscari, A., Pudder. G.M.. Dilorio. A., Zito, M.. and Abate. G. (1997). Blood pressure and intellectual function in elderly subjects. Age and Ageing. 26 (2), 91-98. Parker. M.G., Thorslund, M., and Nordstrom. M.L. (1992). Predictors of mortality for the oldest old. A 4-year follow-up o f community-based elderly in Sweden. Archives of Gerontology and Geriatrics. 14. 227-237. Parker. M.G.; Thorslund. M., and Lundberg, O. (1994). Physical function and social class among Swedish oldest old. Journal of Gerontology: Social Sciences. 49. SI 96- S201. Pattie. A.H., and Gilleard. C.J. (1979). Manual of the Clifton Assessment Procedures for the Elder (CAPE). Sevenoaks. Kent: Hodder and Stoughton Educational. Palmore. E.B., Nowlin.J.B.. and Wang, H.S. (1985). Predictors o f function among the old-old: A 10 year follow up. Journal of Gerontology. 40. 244-250. Pearlin. L.I.. and Schooler. C. (1978). The structure o f coping. Journal of Health and Social Behavior. J9, 2-21. Penninx, B.W.J.H., van Tilburg, T.G.. Kriegsman. D.M.W., Deeg. D.J.H., Boeke, A.J.P.. and Van Eijk, J.T.M. (1997). Effects of social support and personal coping resources on mortality in older age: The longitudinal aging study Amsterdam. American Journal o f Epidemiology. 146 (6), 510-19. Perros. P.. Deary. I.J.. Sellar. R.J. Best. J.J.. and Frier. B.M. (1997). Brain abnormalities demonstrated by magnetic resonance imaging in adult 1DDM patients with and without a history of recurrent severe hypoglycemia. Diabetes Care. 20, 1013- 1018. Pinsky. J.L.. Branch, L.G.. and Jette. A.M. (1985). Framingham disability study: relationship o f disability to cardiovascular risk factors among persons free of diagnosed cardiovascular disease. American Journal of Epidemiology. 122. 644- 656. Pinsky, J.L., Leaverton. P.E., and Stokes, J. (1987). Predictors o f good function: The Framingham Study. Journal o f Chronic Diseases. 40 (supplement 1), 159S-167S. Pinsky, J.L., Leaverton. P.E., and Stokes, J (1989). Predictors o f good (unction: the Framingham study. Journal o f Chronic Disease, 40 (supplement 1), 159S-167S. Podsiadlo, D., and Richardson, S. (1991). The timed “up and go,” a test of basic mobility for frail elderly persons. Journal o f the American Geriatrics Society. 39, 142-148. 2 7 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pope A.M., and Tarlov, A.R. (Eds.) 1991. Disability in America: Toward a National Agenda for Prevention. Division o f Health Promotion and Disease Prevention, Institute o f Medicine. National Academy Press, Washington, D.C., 1991. Porter. L.S., and Stone. A.A. (1995). Are there really gender differences in coping? A reconsideration o f previous data and results from a daily study. Journal o f Social and Clinical Psychology. 14 (2), 184-202. Ptacek, J.T., Smith. R.E., and Zanas. J. (1992). Gender, appraisal, and coping: A longitudinal analysis. Journal o f Personality. 60 (4), 747-770. Reuben, D.B., Laliberte, L.. Hiris, J., and Mor. V. (1990). A hierarchical scale to measure function at the advanced activities of daily living (AADL) level. Journal of American Geriatric Society. 38, 885-861. Reuben. D.B.. and Siu, A.L. (1990). An objective measure o f physical function o f the elderly outpatients: the physical performance test. Journal o f the American Geriatric Society. 38, 1105-1112. Reuben, D.B.. Valle, L.A., Hays. R.D., and Siu, A. L. (1995). Measuring physical function in community dwelling older persons: A comparison o f self administered, interviewer-administered and performance based measures. Journal of the American Geriatrics Society, 43. 17-23. Rikli. R.E., and Jones C.J. (1997). Assessing physical performance in independent older adults: Issues and guidelines. Journal of Aging and Physical Activity, 5,244- 261. Rissanen, A.M., Heliovaara. M.. Knekt, P.. Reunanen. A., and Aromaa. A. (1991). Determinants o f weight gain and overweight in adult Finns. European Journal of Clinical Nutrition, 45, 419-430. Ritchie. K. (1995). Is senile dementia ‘age-related' or ’ageing-related'? Evidence from meta-analysis o f dementia prevalence in the oldest old. Lancet. 346.931 -934. Robbins. A.S., Rubenstein, L.Z.. Josephson, K.R., Schulman, B.L.. Osterweil. D., and Fine G. (1989). Predictors o f falls among elderly people: Results o f two population-based studies. Archives o f Internal Medicine, 149, 1628-1633. Roberts, B. L., Dunkle. R., and Haug, M. (1994). Physical, psychological and social resources as moderators o f the relationship o f stress to mental health o f the very old. Journal o f Gerontology: Social Sciences. 49, S35-S43. Rodin. J. (1986). Aging and Health: Effects of the sense o f control. Science, 233, 1271- 1276. 2 7 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rodin, J. (1990). Control by any other name: Definitions, concepts, and processes. In J. Rodin, C. Schooler, and K.W. Schaie (Eds.), Self-directedness: Cause and effects throughout the life course (pp. 1-17). Hillsdale. NJ: Erlbaum. Rodin. J.. and McAvay (1993). Determinants o f change in perceived health in a longitudinal study o f older adults. Journal o f Gerontology: Psychological Sciences, 47, P373-P384. Rogers, R.G., Rogers, A., and Belanger, A. (1992). Disability-free life among the elderly in the United States. Journal o f Aging and Health, 4(1). 19-42. Roos. N.P., and Havens B. (1991). Predictors of successful aging: a 12 year study of Manitoba elderly. American Journal of Public Health. 81. 63-68. Rose, G. and Barker. D.J.P. (1978). Repeatability and validity. British Medical Journal. 2, 1070-1071. Rosen W.. Mohs. R.. and Davis. R. (1984). A new rating scale for Alzheimer's disease. American Journal o f Psychiatry. 141. 1356-1364. Rosow, I., and Breslau, N. (1966). A Guttman health scale for the aged. Journal of Gerontology. 21, 556-559. Rozzini. R., Frisoni, G.B., Bianchetti, A., Zanetti. 0 ., and Trabucchi, M. (1993). Physical performance test and activities of daily living scales in the assessment o f health status in elderly people. Journal o f the American Geriatrics Society, 4JL 1109- 1113. Sager. M.A.. Dunham. N.C.. Schvvanters. A.. Mecum. L.. Halverson. K. and Harlowe. D. (1992). Measurement o f activities o f daily living in hospitalized elderly: A comparison o f self-report and performance based methods. Journal o f the American Geriatric Society. 40,447-462. Sager, M.A., Rudberg, M.A.. Jalaluddin, M., Franke, T., Inouye. S.K., Landefeld, S., Siebens, H., and Winogard, C.H. (1996). Hospital admission risk profile (HARP): Identifying older patents at risk for functional decline following acute medical illness and hospitalization. American Geriatric Society. 44. 251-57. Salthouse, T.A. (1994). The aging o f working memory. Neuropsychology. 8, 535-543. Sanderman, R., and Ranchor, A.V. (1994). Stability o f personality traits and psychological distress over six years. Perceptual and Motor Skills. 78, 89-90. 274 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Santariano, W.A., Ragheb, N.E., Branch, L.G., Swanson, G.N. (1990). Difficulties in physical functioning reported by middle aged and elderly women with breast cancer: A case controlled comparison. Journal o f Gerontology. 45, M 3-M 11. SAS Institute, Inc. (1990). SAS/STAT User’s Guide. Version 6, Fourth Edition. 2. 893-996 and 1070-1126. Scherr. P.A.. Albert. M.S.. Funkenstein. H.H.. Cook, N.R.. Hennekens. C.H., Branch. L.G., White. L.R.. Taylor. J.O., and Evans, D.A. (1988). Correlates of cognitive function in an elderly community population. American Journal o f Epidemiology. 128,1084-1101. Schiaffino. K.M.. Revenson. T.A.. and Gibofsky. A. (1991). Assessing the impact of self- efficacy beliefs on adaptation to rheumatoid arthritis. Arthritis Care and Research. 4, 150-157. Schiaffino, K.M.. and Revenson, T. A. (1992). The role o f perceived self-efficacy. perceived control, and causal attributions in adaptation to rheumatoid arthritis: Distinguishing mediator from moderator effects. Personality and Social Psychology Bulletin. 18. 709-718. Schoenfeld, D.E.. Malmrose, L.C.. Blazer. D.G.. Gold. D.T.. and Seeman. T.E. (1994). Self-rated health and mortality in the high-functioning elderly— a closer look at healthy individuals: MacArthur field study o f successful aging. Journal of Gerontology: Medical Sciences. 49. M 109-M115. Schuntermann, M.F. (1996). The international classification of impairments, disabilities and handicaps (ICIDH) - results and problems. International Journal of Rehabilitation Research. ]9, 1-11. Seeley. D.G.. Kelsey, J., Jergas, M., and Nevitt. M.C. (1996). Predictors o f ankle and foot fractures in older women. Journal o f Bone and Mineral Research. U (9). 1347- 1355. Seeman, T.E., Kaplan, G.A., Knudsen, L., Cohen, R., and Guralnik. J. (1987). Social network ties and mortality among the elderly in the Alameda County Study. American Journal o f Epidemiology. 126, 714-23. Seeman, T.E., and Berkman. L.F. (1988). Characteristics o f social networks and the provision o f social support in the elderly: who provides support? Social Science Medicine. 26, 737-749. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Seeman, T.E., Charpentier, P.A., Berkman, L.F.. Tinetti, M.E., Guralnik, J.M., Albert, M.. Blazer, D., and Rowe, J.W. (1994). Predicting changes in physical performance in a high-functioning elderly cohort: MacArthur Studies of Successful Aging, Journal of Gerontology: Medical Sciences. 49, M97-M108. Seeman. T .E .,, Berkman, L.F.. Charpentiener, P.A., Blazer, D.G.. Albert, M.S.. and Tinetti, M.E. (1995). Behavioral and psychosocial predictors of physical performance: MacArthur studies o f successful aging, Journal o f Gerontology: Medical Sciences. 50A (4), M177-M183. Seeman, T.E.. Bruce, M.L., and McAvay, G.J. (1996). Social network characteristics and onset o f ADL disability: MacArthur studies o f successful aging. Journal of Gerontology: Social Sciences. 51B (4), S191-S200. Selby. P.. and Robertson. B. (1987). Measurement of quality of life in patients with cancer. Cancer Surveys. 6, 521-543. Sheikh. K.. Smith. D.S., Meade, T.W.. Goldenberg, E.. Brennan. P.J.. and Kinsella. G. (1979). Repeatbility and validity o f a modified Activities o f Daily Living (ADL) index in studies o f chronic disability. International Journal of Rehabilitative Medicine. L 51-58. Sherer, M.. Maddux, J.E., Mercandante, B., Prentice-Dunn. S.. Jacobs. B.. and Rogers, R.W. (1992). The self-efficacy scale: Construction and validation. Psychological Reports, 51,663-671. Simeonsson, R.J., Lollar. D.. Hollowell. J., Adams. M. (2000). Revision o f the international classification of impairments, disabilities, and handicaps developmental issues. Journal o f Clinical Epidemiology. 53. 113-124. Smith. D.S.. and Clark, M.S. (1995). Competence and performance in Activities o f Daily Living of patients following rehabilitation from stroke. Disability and Rehabilitation. 17(1). 15-23. Snowdon, D.A.. Ostwalk. S.K., Kane, R.L.. and Kennan. N.L. (1989). Years o f life with good, poor mental and physical function in the elderly. Journal o f Clinical Epidemiology, 42, 1055-1066. Spitzer, W.O. (1987). State o f science 1986: Quality o f life and functional status as target variables for research. Journal o f Clinical Disease. 40. 465-471. Spirduso, W.W., and MacRae, P.G. (1990). Motor performance and aging. In Birren JD. Schaie, K.W., eds. The Handbook o f the Psychology o f Aging. 3r ed. San Diego. CA: Academic Press, 184-197. 276 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Stein, R.E.K. (2000). Can one size fit all? Journal o f Clinical Epidemiology. 53. 111-112. Steine, S.. Laerum, E., Eritsland. J., and Amesen. H. (1996). Predictors of enhanced well being after coronary-artery bypass-surgery. Journal o f Internal Medicine, 239 (1). 69-73. Stelmach, G.E.. and Worringham. C.J. (1985). Sensorimotor deficits related to postural stability. Clinics in Geriatric Medicine. \, 679-692. Stephens, M.A. P.. Kinney. J.M.. Norris, V.K.. and Ritchie. S.W. (1987). Social networks as assets and liabilities in recovery from stroke by geriatric patients. Psychology and Aging. 2. 125-129. Stewart, A.L.. and Walre, J.E. (1992). Measuring functioning and well-being: The Medical Outcomes Study Approach. Durham. N.C.: Duke University Press. Stokes. J.P., and Wilson. D.G. (1984). The inventory o f socially supportive behaviors: Dimensionality, prediction, and gender differences. American Journal of Community Psychology, ]_2, 53-69. Stone, A. A., and Neale, J.M. (1984). New measure o f daily coping: Development and preliminary results. Journal o f Personality and Social Psychology. 46 (4). 892- 906. Strawbridge. W.J., Kaplan, G .A .. Camacho. T.. and Cohen. R.D. (1992). The dynamics o f disability and functional change in an elderly cohort: Results from the Alameda County Study. Journal o f the American Geriatrics Society, 40 (8). 799-806. Stump. T.E., Clark. D.O.. Johnson. R.J.. and Wolinsky F.D. (1997). The structure of health status among Hispanic, African American, and white older adults. Journal o f Gerontology: Psychological Sciences. B Psychological Social Sciences. 52, 49- 60. Taal. E., Rasker, J.J., Seydel, E.R.. and Wiegman, O. (1993). Health status, adherence with health recommendations, self-efficacy and social support in patients with rheumatoid arthritis. Patient Education and Counseling. 20. 63-76. Teresi, J.A., Golden, R.R., Gurland, B.J., Wilder, D.E., and Bennett. R.G. (1984). Construct validity o f indicator scales developed from the comprehensive interview schedule. Journal o f Gerontology. 39, 147-157. Thompson. S.C., Sobolew. A., Graham, M.A., and Janigian, A.S. (1989). Psychosocial adjustment following a stroke. Social Science and Medicine. 28. 239-247. Tinetti, M.E. (1986). Performance-oriented assessment o f mobility problems in elderly patients. Journal o f the American Geriatrics Society, 34, 119-126. 2 7 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tinetti, M.E., Williams, T.F., and Mayewski, R. (1986). Falls risk index for elderly patients based on number o f chronic disabilities. American Journal o f Medicine. 80. 428-434. Tinetti. M.E., Speechley, M., and Gimnter, S.F. (1988). Risk factors for falls among elderly persons living in the community. New England Journal o f Medicine. 319. 1701-1707. Tinetti. M.E., Mendes de Leon, C.F.. Doucete, J.T., and Baker, D.I. (1994). Fear of falling and fall-related efficacy in relationship to functioning among community- living elders. Journals of Gerontology: Medical Sciences. 49, M140-M147. Tinetti, M.E.. Inouye, S.K., Gill. T.M., and Dourcette. J.T. (1995). Shared risk factors for falls, incontinence and functional dependence: unifying the approach to geriatric syndromes. Journal o f the American Medical Association. 273. 1348-1353. Tobin. J. (1958). Estimation of Relationships for Limited Dependent Variables. Econometrica. 26. 24-36. Tondat-Carter. L., Olsen-Oliveira, D.. Duponte. J.. and Lynch, S.V. (1988). The relationship of cognitive skills performance to activities of daily living in stroke patients. American Journal o f Occupational Therapy. 42< 449-455. Uhlmann. R.F., and Larson. E.B. (1991). Effect of education on the Mini-Mental State Examination as a screening test for dementia. Journal of American Geriatric Society. 39, 876-880. United States Census Bureau, United States Department of Commerce (January 2000). Projections o f the Resident Population bv Age. Sex. Race, and Hispanic Origin: 1999 to 2040. Population Projections Program Population Division, Washington. D.C. United States Congress (1982). Technology and handicapped people. Washington D.C.: Office o f Technology Assessment. United States Department o f Education (March 1982). Annual report o f the National Council on the Handicapped. Washington, DC: Institute o f Handicapped Research. Vanschaardenburg, D., Vandenbrande, K.J.S., Ligthart, G.J., Breedveld, F.C., and Hazes. J.M.W. (1994). Musculoskeletal disorders and disability in persons aged85 and over: A community survey. Annals of Rheumatic Diseases, 53(12). 807-811. 278 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Verbrugge, L.M. (1983). Women and men: Mortality and health of older people. In Aging in Society. (Eds.) M.W. Riley, B.B. Hess, K. Bond. Lawrence Erbaum Associates. New Jersey, 139-174. Verbrugge, L.M. (1985). Gender and health: an update on hypotheses and evidence. Journal o f Health and Social Behavior, 26, 156-182. Verbrugge. L.M.. and Wingard D.L., (1987). Sex differentials in health and mortality. Women Health. 12, 103-145. Verbrugge. L. M. (1989). Recent, present and future health of American adults. Annual Review of Public Health. 10. 333-361. Verbrugge. L.M. (1992). Disability transitions for older persons with arthritis. Journal of Aging and Health. 4 (2). 212-243. Verbrugge. L.M.. and Jette, A.M. (1994). The disablement process. Social Science Medicine. 38 (1). 1-14. Victor, C.R. (1989). Inequalities in health in later life. Age and Ageing. ]_8, 387-391. Viney. L.L.. and Westbrook. M.T. (1982). Coping with chronic illness: The mediating role of biographic and illness-related factors. Journal of Psychosomatic Research. 26, 595-605. Vingerhoets. A.J.. and VanHeck. G.L. (1990). Gender, coping, and psychosomatic symptoms. Psychological Medicine. 20. 125-135. Visser. M.. Harris, T.B.. Langlois. J.. Hannan, M.T., Roubenoff. R., Felson. D.. Wilson. P.W.. and Kiel, D.P. (1998). Body fat and skeletal muscle mass in relation to physical disability in very old men and women o f the Framingham Heart Study. Journal o f Gerontology: Medical Sciences. 53, M214-221. Visser, M„ Langlois, J., Guralnik. J.M., Cauley, J.A., Kronmal, R.A.. Robbins. J., Williamson, J.D., and Harris. T.B. (1998b). High Body Fatness, but not low fatO- free mass, predicts disability in older men and women: The Cardiovascular Health Study. American Journal o f Clinical Nutrition. 68, 584-590. Vogt. T.M., Mullooly, J.P., Ernst, D.. Pope, C.R.. and Hollis. J.F. (1992). Social networks as predictors o f ischemic heart disease, cancer, stroke and hypertension: incidence, survival and mortality. Journal o f Clinical Epidemiology. 45, 659-666. Voorwerk 12 (1993). WCC newsletter on the ICIDH. 16. 3. 279 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wagner E.H., LaCroix, A.Z., Grothaus. L., Leveille, S.G, Hecht, J.A., Artz, K., Odle, K., and Buchner D.M. (1994). Preventing disability and falls in older adults: A population-based randomized trial, American Journal o f Public Health. 84 (11). 1800-1806. Ware. J.E., and Sherboume, C.D. (1992). The MOS 36-Item Short-Form Health Survey (SF-36) Conceptual Framework and Item Selection. Medical Care. 30.473-483. Warren. R.C. (1993). The morbidity/mortality gap: what is the problem? Annals of Epidemiology. 3. 127-9. Warshaw. G.A., Moore, J.T., Friedman, W„ Currie, C.T., Kennie. D.C.. Kane, W.J., and Mears, P.A. (1982). Functional disability in the hospitalized elderly, Journal of the American Medical Association. 248: 847-850. Wechsler, D. (1981). Adult Intelligence Scale-Revised. New York: Psychological Corporation. Wiener. J.. Hanley. R.. Clark. R.. and Van Nostrand. J. (1990). Measuring the activities o f daily living: comparisons across national surveys. Journal o f Gerontology: Social Sciences, 45, S229-S237. Welin, L.. Tibblin. G.. Svardsudd. K.. Tibblin, B.. Ander-Peciva, S.. Larsson, B., and Wilhelmsen. L. (1985). Prospective study of social influences on mortality. Lancet, i , 915-918. Welin. L., Larsson, B.. Svardsudd. K., Tibblin, B.. and Tibblin, G. (1992). Social network and activities in relation to mortality from cardiovascular diseases, cancer and other causes: a 12 year follow up o f the study o f men bom in 1913 and 1923. Journal Epidemiology Community Health. 46. 127-132. Wernicke. T.F.. and Reischies, F.M. (1994). Prevalence o f dementia in old age clinical diagnosis in subjects 95 years and older. Neurology. 44, 250-253. Wessen, A.F. (1965). The rehabilitation apparatus and organization theory. In M.B. Sussman (Ed.) Sociology and rehabilitation. American Sociological Association, in cooperation with the Vocational Rehabilitation Administration. Washington. D.C.: U.S. Department o f Health, Education, and Welfare. West, L.J., Maxwell, D.S., Noble, E.P., and Solomon, D.H. (1984). Alcoholism. Annals o f Internal Medicines. 100.405-416. Wetle, T. Scherr, P., Branch, L.G., Resnick, N.M., Harris. T., Evans, D., and Taylor, J.O. (1995). Difficulty with holding urine among older persons in a geographically defined community: Prevalence and correlates. Journal o f American Geriatric Society, 43. 349-355. 280 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Whiteneck, G., and Fougeyrollas, P. (1996). Environmental factors task force position paper. Environmental factors and ICIDH. International 1CIDH Network. 8 (Sept.), 14-18. Wiersma. D. (1986). Psychological impairments and social disabilities: On the applicability o f the ICIDH to psychiatry. International Rehabilitv Medicine, 8. 3- 7. Wiersma, D. (1996). Measuring social disabilities in mental health. Social Psychiatry Epidemiology. 31. 101-108. Wilcox, B.L. (1981). Social support, life stress, and psychological adjustment: A test of the buffering hypothesis. American Journal of Community Psychology. 9. 371- 386. Wild. D., Mayak, U.S.L., and Isaacs, B. (1981). Prognosis o f falls in old people at home. Journal o f Epidemiology and Community Health. 35, 200-204. Wilson. L.A., Grant, K.. Witney. P.M., Kerridge, D.F. (1973). Mental status o f elderly hospital patients related to occupational therapist's assessment of activities o f daily living. Gerontologia Clinica. 15, 197-202. Williams. S.J. (1989). Chronic respiratory illness and disability, a critical review o f the psychosocial literature. Social Science and Medicine. 28, 791-803. Winograd, C.H. (1984). Mental status tests and capacity for self care. Journal o f the American Geriatrics Society. 32,49-55. Winograd. C.H., Lemski, C.M.. Nevitt, M.C.. Nordstrom.T.M. Stewart. A.L., Miller. C. J.. and Bloch. P. A. (1994). Development o f a Physical Performance and Mobility Examination (PPME). Journal o f the American Geriatric Society. 42(7). 743-749. Winter. D. (1991). The biomechanics and motor control of human gait: Normal, elderlv and pathological. (2n d ed.). Waterloo, Ontario: University of Waterloo Press. Wolf, P.A., D'Agostino, R.B., O'Neal, M.A., Sytkowski, P.. Kase, C.S., Belanger. A., Kannel. W.B. (1992). Secular trends in stroke incidence and mortality: the Framingham Study. Stroke. 23, 1551-1555. Wolinsky, F.D.. Stump. T.E., Callahan. C.M., and Johnson, R.J. (1996). Consistency and change in functional status among older adults over time. Journal o f Aging and Health. 8.155-82. Woo. J., Ho, S.C., Lau, J., Ckhan, S.G., and Yuen Y.K. (1995). Age associated gait changes in the elderly: Pathological or physiological? Neuroepidemiologv, 14. 65-71. 281 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Woo. J.. Ho. S.C.. Yu, L.M., Lau. J., and Yuen, Y.K. (1998). Impact of chronic diseases on functional limitations in elderly Chinese aged 70 years and over - a cross- sectional and longitudinal survey. Journal o f Gerontology: Medical Sciences. 53 (2), M102-M106. Wood. P.H.N. and Badley. E.M. (1978). An epidemiological appraisal o f disablement. In Recent advances in Community Medicine, ed. A.E. Bennett. Churchill Livingstone, Edinburgh. Wood. P.H.N. (1980). The language of disablement: A glossary relating to disease and its consequences. International Rehabilitv Medicine. 2, 86-92. Wood. P.H.N., and Badley. E.M. (1981). People with Disabilities—Toward Acquiring Information which Reflects More Sensitively Their Problems and Needs. New York: World Rehabilitation Fund. Woollacott, M.H. (1990). Changes in posture and voluntary control in the elderly. Research findings and rehabilitation. Topics in Geriatric Rehabilitation. 5. l-l 1. World Health Organization (1980). International Classification of Impairments. Disabilities and Handicaps: a manual o f classification relating to the consequences of disease. Geneva. Switzerland, World Health Organization. World Health Organization (1993). International Classification of Diseases. ICD-10. Geneva, Switzerland. World Health Organization (1995). Technical Report Series. Physical status: the use and interpretation o f anthropometry: report o f a WHO expert committee. 854. 1-452. Zimberg, S. (1987). Alcohol abuse among the elderly. In Cartenson B. Edelstein B (eds.): Handbook o f Clinical Gerontology. New York. Pergamon. p. 57. Zimmer. Z.. Lieu, X., Hermalin, A., and Chuang, Y.L. (1998). Educational-attainment and transitions in functional status among older Taiwanese. Demography. 35 (3). 361-375. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Processes of loss and recovery of physical abilities in a longitudinal study of Americans 70 years of age and older

PDF

Out -of -pocket health expenditures by older adults in relation to age, race, and insurance

PDF

The relationship between depressed mood and cognitive decline in older Americans: A longitudinal analysis by gender

PDF

Narrative practices of intersubjectivity: An ethnography of children with autism in a sensory integration based occupational therapy clinic

PDF

The extent and effectiveness of nursing home regulation in the 50 states

PDF

Heart disease among middle -aged and elderly persons in the United States: Trends and a multistate model

PDF

Monoamine oxidase A: (1) As one of many candidate genes for preeclampsia and (2) as a candidate gene for tobacco addiction

PDF

The effects of East Asian values, coping styles, social support on physical health among Korean American caregivers

PDF

Using network perspective to examine the organization of community -based elder care systems across four communities

PDF

Marriage and health: The effect of marital status on health and the mechanisms by which marital status affects health among elderly people

PDF

The relationships among five sensory systems in children with sensory integration problems as measured by the Evaluation of Sensory Processing (ESP)

PDF

Psychologists' perceptions of older clients: The effect of age, gender, knowledge, and experience

PDF

Age, SES, and race/ethnic differences in biological risk: Allostatic load and metabolic syndrome

PDF

Physical activity and adenomatous polyps: Measures of association and impact

PDF

Evaluation of the Care Advocate Program: Bridging managed care and home community -based services

PDF

Social support, social change, and psychological well-being of the elderly in China: Does the type and source of support matter?

PDF

Sources of aging anxiety in the non -elderly population: The importance of images, perceptions, and social resources

PDF

Survey of employers in the field of aging: Implications for gerontological education

PDF

Mothering occupations: An ethnographic study of Thai women with physical disabilities in the north of Thailand

PDF

Self -care strategies as coping mechanisms: Older adults' proactive role in managing functional limitations and achieving adaptational outcomes

Asset Metadata

Creator Tabbarah, Melissa (author)

Core Title The measurement of physical functioning in a longitudinal study of Americans 70 years of age and older

Degree Doctor of Philosophy

Degree Program Gerontology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Gerontology,health sciences, human development,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-87501

Unique identifier UC11327951

Identifier 3018133.pdf (filename),usctheses-c16-87501 (legacy record id)

Legacy Identifier 3018133.pdf

Dmrecord 87501

Document Type Dissertation

Rights Tabbarah, Melissa

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA