Effectiveness of a heart plus brain health-focused nutrition intervention among Latinas: a randomized controlled trial of Buenos Hábitos Alimenticios Para Una Buena Salud (Good Eating Habits for ...

PDF

Download

Open document

Flip pages

Summarize

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content EFFECTIVENESS OF A HEART PLUS BRAIN HEALTH-FOCUSED NUTRITION
INTERVENTION AMONG LATINAS: A RANDOMIZED CONTROLLED TRIAL OF
BUENOS HÁBITOS ALIMENTICIOS PARA UNA BUENA SALUD
(GOOD EATING HABITS FOR GOOD HEALTH)

by

Poorni G. Otilingam

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)

December 2010

Copyright 2010 Poorni G. Otilingam
ii

Epigraph

Se necesita una aldea.

iii

Dedication

I dedicate this dissertation:

To my Maata, Rani Bai G. Otilingam,
my first teacher, for your malai of unconditional love, inspiring kindness, constant
patience, as well as your boundless support of my academic pursuits.

To both my Maata and Pitaa, Ganapathi-Raj M. Otilingam,
for all of your sacrifices, for instilling the significance of education, and for modeling a
passion for lifelong learning and intellectual curiosity.

To my Guru, my dearest Baba,
for always walking alongside me.

and

In memory of my late and beloved Paati, Visalakshi Mangadu.

iv

Acknowledgments

I gratefully acknowledge Dr. Margaret Gatz for tirelessly lending her expertise
and guidance as my advisor, mentor, and dissertation committee chair, and for pouring
her heart and brain into this study. I thank my committee members, Dr. Richard John
for his statistical support, and Drs. Beth Meyerowitz and Maryalice Jordan-Marsh for
their involvement. I thank Drs. Helena Chui and Wendy Mack, and Mr. Michael Adkins
of the USC Alzheimer Disease Research Center, for their support of this study. I thank
Dr. Rohit Varma, Ms. Mina Torres, and the rest of the USC Los Angeles Latino Eye
Study (LALES) team, for allowing us access to the study population, and for permitting
us use of the LALES clinic. This study would not have been possible without the
dedicated involvement of the participants, who shared details of their lives in hopes
that it would help other women. I also gratefully acknowledge a dedicated research
study team who tirelessly spent many days, evenings, and weekends providing
research support in the lab or in the field. Eleven of these students worked as research
assistants: Mr. Henry Cook, Mr. Fabian Corona, Ms. Jillion Crawford, Ms. Cristina
Huizar, Ms. Kyanne O’Bryant, Ms. Joanna Palma, Ms. Ashley Phillips, Mr. Chris Saito,
Ms. Deidra Smith, Ms. Giulia Suro, and Ms. Jessica Zetino. Additionally, three students
served as project coordinators extraordinaire, namely: Mr. Antonio Escobar, Ms. Aviva
Goldstein, and Ms. Elizabeth Tello. I offer each of them my sincerest gratitude.
Additionally, I thank the project coordinators for fiercely embodying the “Fight On” spirit,
and for Ms. Tello’s dedication well above the call of duty. I thank past and present
fellow lab members in the Section on Clinical Research in Aging and Psychology,
including Ms. Barbara Yuen, Drs. Margaret-Anne Mackintosh, Jessica Brommelhoff,
v

Amber Hall, Kecia Watari Knoell, Michael Crowe, Ross Andel, and Amy Fiske, Ms.
Randi Jones, Ms. Priya Kapoor, and soon-to-be-Drs. Lewina Lee, Patricia George,
Carlos Rodriguez, and Emily Schoenhofen Sharp, for their steady support. My heartfelt
gratitude to Ms. Cecilia Fuentes, Ms. Sandy Medearis, and Ms. Twyla Ponton for their
incredible support (administrative and otherwise). I thank several informal consultants
at USC who offered invaluable input: Drs. Carol McCleary, Donna Spruijt-Metz, Susan
C. Harris, and Rand Wilcox, and employees from USC Housekeeping Services who
served as pilot testers. I am grateful to several consultants in the community who
generously lent their expertise. I thank Ms. Georgina Serrano of Accelerated Healthy
Eating Active Communities and Ms. Marie Mayen-Cho of the Alzheimer’s Association
for their invaluable feedback of the curriculum. I thank the pilot testers affiliated with the
following organizations: Norwood Elementary School’s Healthy Start Parents’ Group,
Accelerated Healthy Eating Active Communities, Coalition for Community Health,
Sustainable Economic Enterprise of Los Angeles, California Hospital Medical Center –
Catholic Healthcare West, Healthy Homes Project Promotores Group, Esperanza
Housing Coalition, and Sycamore-Hathaway Child and Family Services Center. I thank
Ms. Virginia Céspedes and Ms. Edna Smith for their translational services. I am
grateful to Dr. Debra Cherry of the Alzheimer’s Association and Dr. Ramón Valle of
San Diego State University for their invaluable input. This study was supported in part
by NIH Grant No. P50 AG005142 (PI: Dr. Helena Chui; Director of Education Core: Dr.
Margaret Gatz); a USC Wallis Annenberg Fellowship; and a USC College of Letters,
Arts, and Sciences Summer Undergraduate Research Fund (awarded to Ms.

vi

Goldstein). The pilot test was supported by NIH Grant No. R25 MH071544-01 (PI: Dr.
Barry Lebowitz; co-PI: Dr. Jilip Jeste).
Several individuals supported me during this study, and more generally, in
graduate school. I thank Drs. Susan Corban Harris, Joseph Hellige, Randa Issa,
Julena Lind, Bradley Stoner, Nancy Thompson, and Mark Todd, for their unwavering
support. I appreciate the dissertation support offered by Drs. Sheila Bienenfeld, Glenn
Callaghan, Tat Fu, Elena Klaw, and Annabel Prins. Additionally, I thank Dr. Jennifer
Gregg for her support and friendship over the years, and especially as the study
concluded. I gratefully acknowledge Drs. John McQuaid and Stephen Rao for their
patience, understanding, and encouragement as the study neared completion. I am
indebted to Drs. William Howell, Robert Mocharnuk, and Lisa Richardson for skillfully
demonstrating the science and art of applied clinical research with such humanity,
humility, and hope, and for inspiring me to never give up.
I thank my friends who generously fortified me with love, support, hope, and
comic relief during graduate school: Heidi, Benoy, and Elina Ojha, and their families;
Helen Chow; Lisa, Glen, and Evan Aquino, and the Lindvall family; Che-Chin Lie;
Bhagya Wickrama; Rachel Wing DiMatteo, Brian DiMatteo and their families;
Cassandra Folan; Julie Vassallo; Geetha Govindarajan; Nilima Abrams; Gayatri
Yadavalli; Siva Nadarajah, and many individuals at the Peninsula Sai Center. A cadre
of dear friends and colleagues similarly enriched my life in countless ways during
graduate school, and for that I gratefully acknowledge Jennifer Dave, Margaret-Anne
Mackintosh, Yolanda Céspedes, Lewina Lee, Kevin Ennis, Randa Issa, Ashley
Borders, Lara Heflin, and Kysa Christie, for walking this path with me. I also thank

vii

Swati, Christopher, and Aasha Jacobs for lending life perspective when the going got
tough. My circle of friends made the long days and nights spent working in graduate
school brighter than ever imaginable.
And last, but certainly not least, I thank my family for their tremendous love,
support, and prayers that sustained me throughout my schooling. Without them, none
of this would have been possible. My parents, Rani Bai G. Otilingam and Ganapathi-
Raj M. Otilingam, unceasingly inspired me to work hard and persevere in the midst of
adversity. My brother, Venkat-Raj G. Otilingam, kindly offered priceless pep talks, an
alarmingly quick wit, and fervent support (especially of the cardinal and gold). My
goddaughter, Sahana Sai Narayan, literally breathed new life into my doctoral studies
by simply being. Her parents, Gayathri Narayan and Sankara Lakshmi Narayan,
inspired me to continually see past seemingly insurmountable odds. And yes, Sahana,
I finally “finished my homework”.
viii

Table of Contents

Epigraph ii

Dedication iii

Acknowledgments iv

List of Tables xi

List of Figures xiii

Abstract xiv

A. Background and Significance 1

A.1. Cardiovascular risk and Latinos 2
A.2. Vascular disease as a risk factor for vascular dementia and Alzheimer’s
disease
3
A.3. Cognitive impairment and Latinos 5
A.4. Diet and vascular risk 7
A.5. Latinas as target audience 9

A.6. Theoretical foundation 11
A.6.a. Risk communication 11
A.6.b. Health behavior theory 12

A.7. Dietary fat outcomes 14
A.7.a. Dietary fat knowledge 14
A.7.b. Dietary fat intentions 15
A.7.c. Self-reported dietary fat behaviors 16

A.8. Previous dietary fat outcome research among Latinas 16

A.9. Intervening variables 18
A.9.a. Susceptibility as a moderator 18
A.9.b. Familism as a moderator 21
A.9.c. Covariates 22

A.10. Feedback and motivation 24
A.11. Treatment integrity variables 25
A.12. Study hypotheses 26

ix

B. Intervention 31

B.1. Overview of Buenos Habitos Alimenticios para una Buena Salud 31
B.2. Application of theory 32
B.3. Pilot testing 35
B.4. Intervention development 37
B.5. Intervention procedure 41

C. Methodology 45

C.1. Design 46

C.2. Participants 49
C.2.a. Estimated power and sample size 49
C.2.b. Recruitment from LALES base study 49
C.2.c. Recruitment from La Puente community 50
C.2.d. Yoked participants 53

C.3. Procedure 54

C.4. Measures 58
C.4.a. Dietary fat outcomes 61
C.4.b. Moderators 65
C.4.c. Covariates 67
C.4.d. Treatment integrity 70
C.4.e. Descriptive information 72

C.5. Data analysis 74
C.5.a. Preliminary analyses 74
C.5.b. Treatment integrity 76
C.5.c. Primary outcomes 78
C.5.d. Intervening variable analyses 80
C.5.e. Effect size 83

D. Results 84

D.1. Achieved sample size and actual power 84

D.2. Baseline characteristics 85
D.2.a. Sample characteristics 85
D.2.b. Test of group equivalence at pretest 91
D.2.c. Posttest only condition 95
D.2.d. Time lag 95
D.2.e. Test of selective attrition 96

D.3. Treatment integrity 100
D.3.a. Manipulation check 100
D.3.b. Horizontal diffusion 104
x

D.3.c. Self-reported facilitator adherence 104
D.3.d. Self-reported facilitator competence 104
D.3.e. Participant satisfaction 108
D.3.f. Motivational boost 108

D.4. Primary outcomes 110
D.4.a. Completers analysis 110
D.4.b. Post hoc analyses for completers 117
D.4.c. Intent-to-treat analysis 118

D.5. Intervening analyses 125
D.5.a. Tests for moderators 125
D.5.b. Covariates 128

E. Discussion 131

E.1. Findings 131
E.1.a. Overview 131
E.1.b. Dietary fat knowledge 133
E.1.c. Dietary fat intentions 134
E.1.d. Dietary fat behaviors 135
E.1.e. Brain health focus 140
E.1.f. Latinas as nutrition gatekeepers in their families 143

E.2. Limitations 145
E.3. Strengths 147
E.4. Lessons learned 149
E.5. Conclusion 151

References 152

Appendices 172
Appendix A: Certificate of Completion 172
Appendix B: Pilot Test Findings 173
Appendix C: Study Logos 175
Appendix D: Study Timeline 176
Appendix E: Permission to Recruit Letter 177
Appendix F: Motivational Telephone Call Script 178
Appendix G: Waiting List Letter 184
Appendix H: Recruitment Letter 186
Appendix I: Telephone Screening and Intake 190
Appendix J: Information Brochure 196
Appendix K: Information Sheet for Non-Medical Research 198
Appendix L: Interview Packet 207
Appendix M: Self-reported Facilitator Adherence 240
Appendix N: Self-reported Facilitator Competence 241
Appendix O: Participant Satisfaction 242

xi

List of Tables
Table 1: Intervention Activities by Program Area and Intervention Objectives 40
Table 2: Study Design 48
Table 3: Sample Size Projections 52
Table 4: Interview Schedule 60
Table 5: Staging Algorithm for Dietary Fat Intentions for a Low-Fat Diet Measure 64
Table 6: Participant Ascertainment from Screening to Completion of One-Month
Follow-up Assessment

86
Table 7: Achieved Sample Size and Power 87
Table 8: Baseline Characteristics of Participants by Study Condition 88
Table 9: Test of Group Equivalence for Completers at Pretest 92
Table 10: Number of Responses per Time Point by Item among Completers 93
Table 11: Mean Outcome Scores for Pretest-Posttest Condition at Pretest and
Posttest and for Posttest-Only Condition at Posttest

97
Table 12: Stage of and Reason for Attrition 98
Table 13: Test for Selective Attrition for Completers versus Non-completers at
Pretest

99
Table 14: Manipulation Check Mean Scores for Completers 102
Table 15: Percentage of Workshops That Covered Each BHA Curriculum Module 106
Table 16: Test of Motivational Boost for Intervention Conditions at Follow-up 109
Table 17: Mean Scores (and Standard Deviations) for Outcome Measures for
Intervention and Pre-Post Waitlist Conditions, Completers Only

112
Table 18: Time, Condition, and Time by Condition Mean Score Differences for the
Main Dependent Variables, Completers Only

113

xii
Table 19: 3X3 Contingency Table for Dietary Fat Intentions Assessed as Discrete
Stage Score, Pretest to Posttest

116
Table 20: 3X3 Contingency Table for Dietary Fat Intentions Assessed as Discrete
Stage Score, Posttest to Follow-up

116
Table 21: Mean Scores (and Standard Deviations) after Imputing Scores for
Individuals who have Attrited for Outcome Measures for Intervention and
Pre-Post Waitlist Conditions, All Participants

119
Table 22: Time, Condition, and Time by Condition Mean Score Differences for the
Main Dependent Variables, All Participants

120
Table 23: Tests for Moderators 126
Table 24: Tests for Covariates 129

xiii
List of Figures
Figure 1: Yoked participants

54
Figure 2: Dietary fat knowledge mean scores for all participants across three time
points

122
Figure 3: Dietary fat intentions mean scores for all participants across three time
points

122
Figure 4: Self-reported total fat behavior mean scores for all participants across
three time points

123
Figure 5: Self-reported avoiding fat behavior mean scores for all participants across
three time points

123
Figure 6: Self-reported modifying fat behavior mean scores for all participants
across three time points

124
Figure 7: Dietary fat knowledge mean scores for completers from pretest to follow-
up by self vascular risk

127
Figure 8: Self-reported dietary fat avoidance behavior mean scores for completers
from pretest to posttest by young children living at home

130
Figure 9: Self-reported dietary fat avoidance behavior mean scores for completers
from pretest to follow-up by young children living at home

130
xiv

Abstract

This study assessed the effectiveness of a nutrition intervention that
emphasized the connection between heart health and brain health among Latinas.
Public health researchers and practitioners have noted the rise in the number of
Latinos being diagnosed with diseases involving the cardiovascular system and with
risk factors that might portend future vascular disease. Vascular risk factors are
important modifiable conditions for both vascular dementia and Alzheimer’s disease.
As diets high in saturated fatty acids contribute to vascular risk, change in diet could
lower risk for vascular disease and dementia. A randomized controlled study with four
conditions was conducted: two intervention conditions (nutrition workshops that
emphasized heart health and brain health, or “heart plus brain”, and workshops that
emphasized heart health, or “heart only”) and two wait list conditions (“pretest-posttest”
and “posttest only”). Intervention conditions and the first wait list condition were
assessed at pretest, posttest, and one-month follow-up. The heart plus brain condition
received materials that emphasized the importance of vascular health in reducing
dementia risk as well as information about the relationship of nutrition to heart
health.The heart only condition received identical material about the relationship of
nutrition to heart health; the relevance of vascular health to dementia was not
introduced. The objective was to assess whether the two intervention conditions
demonstrated improvements relative to the pretest-posttest wait list condition, and
whether the heart plus brain condition demonstrated greater improvements in dietary
fat outcomes compared to the heart only condition. The culturally-tailored curriculum
developed for this study drew on the Health Belief Model, Social Cognitive Theory, and
xv

materials now in use by health educators. A convenience sample of 100 Latinas were
randomly assigned to condition, with 82 in the three repeated time conditions. Attrition
for these three conditions was 10%. There was significant improvement in self-reported
total fat and fat avoidance behaviors from pretest to follow-up in both of the intervention
conditions compared to the pretest-posttest condition. There were not significant
effects for dietary fat knowledge or dietary fat intentions. The completers analysis was
corroborated with findings from an intent-to-treat analysis. The heart plus brain
condition had significantly higher levels of perceived susceptibility to Alzheimer’s
disease at posttest compared to the heart only condition. However, there was no
indication of greater gains for a heart plus brain intervention compared to a heart only
intervention. Greater self vascular risk moderated greater dietary fat knowledge from
pretest to follow-up for both intervention conditions compared to the pretest-posttest
condition. Unexpectedly, the presence of young children living at home was
significantly related to greater improvement in self-reported fat avoidance behaviors
from pretest to posttest and pretest to follow-up. These results suggest that a culturally-
tailored nutrition intervention can modestly improve self-reported total fat and fat
avoidance behaviors, but that the addition of a brain health component may not
increase effectiveness.

1

A. BACKGROUND AND SIGNIFICANCE

The focus of this study was to test aspects of the behavioral science behind
brain health-focused education campaigns currently in existence that promote lifestyle
behaviors such as healthy diets, as included in the Alzheimer’s Association’s “Maintain
Your Brain” campaign (Alzheimer's Association, 2007) and the AARP’s “Staying Sharp”
campaign (American Association of Retired Persons, 2007). These older-adult-focused
organizations have begun to initiate mass communication campaigns as a means to
increase awareness of and decrease the risk for certain neurodegenerative disorders,
including the dementias (Alzheimer's Association, 2007; American Association of
Retired Persons, 2007). However, the impact of a brain health focus on actual
behaviors being targeted in these campaigns, such as maintaining a healthy diet, had
not been studied ("Safeguarding cognitive health in an ageing population", 2006).
Additionally, the effect of a brain health focus had not been studied with respect to
specific ethnic groups, an important area of growing interest (Alzheimer's Association,
2005; Centers for Disease Control and Prevention and the Alzheimer's Association,
2007). This study asked: whether a heart plus brain health-focused nutrition
intervention (“heart plus brain”) and a heart health-focused nutrition intervention (“heart
only”) demonstrated greater improvements in dietary fat knowledge, intentions, and
behaviors versus a pretest-posttest control condition, as well as whether the heart plus
brain condition demonstrated greater improvements in dietary fat knowledge,
intentions, and behaviors versus a heart only condition; and if so, what specific factors
contributed to this difference?

2

A.1. Cardiovascular risk and Latinos

The study emerged in part from growing epidemiological concern. Latinos
comprise 12.5% of the U.S. population (U.S. Census Bureau, 2004b), and are
estimated to increase to a full quarter of the population by year 2050 (U.S. Census
Bureau, 2004a). In Southern California, individuals of Latino origin constitute nearly
47% of Los Angeles County’s population (U.S. Census Bureau, 2007). These
demographic changes have increased public health interest in the number of Latinos
diagnosed with diseases involving the cardiovascular system – angina pectoris (Ford &
Giles, 2003), myocardial infarctions (Ford & Giles, 2003; Goff, Nichaman, & Chan,
1997), coronary artery disease (Aronow & Ahn, 2001), and stroke (Muntner, Garrett,
Klag, & Coresh, 2002) – and with risk factors that might portend future vascular
disease –Type II diabetes, obesity, hypertension (Sundquist & Winkleby, 1999) and
microalbuminuria (Ford & Giles, 2003). Additionally, research findings suggested major
shifts in cardiovascular risk factors among older Mexican Americans. For example,
longitudinal data of older adults (i.e., ages 65 to 74 and 65 and older, respectively)
from the 1982-1984 Hispanic Health And Nutrition Examination Survey and the 1993-
1994 Hispanic Established Populations for Epidemiologic Studies of the Elderly
revealed that older Mexican Americans’ rates of obesity, diastolic blood pressure, and
Type II diabetes increased over a 10 year period (Stroup-Benham, Markides, Espino, &
Goodwin, 1999).

3

A.2. Vascular disease as a risk factor for vascular dementia and Alzheimer’s
disease

The growing number of Latinos diagnosed with vascular diseases warrants
attention from a cognitive health perspective. A mounting body of evidence indicates
that vascular disease is an important – yet modifiable – risk factor for
neurodegenerative disorders.
A scientific review of large-scale, longitudinal cohort studies conducted by the
NIH Cognitive and Emotional Health Project revealed strong scientific evidence that
cardiovascular risk factors (i.e., diabetes, heart disease, hypertension, or
stroke/transient ischemic attack) predicted neurodegenerative outcomes such as
vascular dementia and AD (Hendrie et al., 2006). Notably, research has considered not
only the mechanism by which cardiovascular risk portends vascular dementia but also
on its contributing role in the development and potentiation of AD (Launer, 2006).
Several studies suggest a metabolic syndromal pathway – particularly at midlife and
one that has been marked by well-known cardiovascular risk factors such as obesity
(Hughes & Ganguli, 2009; Whitmer et al., 2008), centralized adiposity (Whitmer et al.,
2008), diabetes (Hughes & Ganguli, 2009; Xu et al., 2009), hypertension (Hughes &
Ganguli, 2009; Kloppenborg, van den Berg, Kappelle, & Biessels, 2008; Stewart &
Liolitsa, 1999), hypercholesterolemia (Hughes & Ganguli, 2009), hyperlipidemia
(Hughes & Ganguli, 2009; Kloppenborg, van den Berg, Kappelle, & Biessels, 2008),

4

and atherosclerosis and hypo- and hyperglycemia (Kalmijn, Fesken, Launer, Stijnen, &
Kromhout, 1995; Stewart & Liolitsa, 1999) – that contribute to the later development of
dementia.
Cross-sectional findings from the 1988-1994 Third National Health and Nutrition
Examination Survey (NHANES III) indicated that key markers for metabolic syndrome
such as abdominal adiposity, higher body mass index, obesity and diabetes were
disproportionate among Latinos of Mexican descent compared to non-Latinos in the
U.S. (Ford, Giles, & Dietz, 2002; Park et al., 2003). Another cross-sectional study with
a community-based sampling of Latinos of Mexican descent in Los Angeles revealed
that hypertension and cardiovascular disease were the strongest diagnostic predictors
for vascular dementia (Fitten, Ortiz, & Pontón, 2001).
Four longitudinal studies offer insight into long-term adverse cognitive
outcomes for a few risk factors, i.e., abdominal adiposity, higher body mass index,
obesity, diabetes, and hypertension (Kloppenborg, van den Berg, Kappelle, & Biessels,
2008; Whitmer et al., 2008; Xu et al., 2009). First, a retrospective cohort study that
followed patients in a large health maintenance organization for an average of 36 years
revealed a threefold increased risk for dementia among those at midlife with the
greatest centralized distribution of adipose (fat) tissue; moreover, obesity coupled with
high centralized adiposity revealed almost a fourfold increased risk of dementia
(Whitmer et al., 2008). Second, a retrospective case-control study of Swedish twin
pairs discordant for dementia found that diabetes increased risk for dementia, with the
risk slightly higher for vascular dementia than Alzheimer’s disease (AD) (Xu et al.,
2009). Additionally, Xu and colleagues (2009) indirectly underscored the importance of
lifestyle modification at mid-life when they revealed that pairs with a twin with mid-life
5

diabetes had nearly a two-and-a-half times increased risk for later dementia compared
to pairs with a twin with late-life diabetes (defined as disease onset at 65 years of age
or older) – a finding that was not explained by genetic or early-life environmental
factors. Third, a systematic review of vascular risk factors in population-based
longitudinal studies suggested that at midlife, the risk for later life dementia was highest
for individuals with hypertension (Kloppenborg, van den Berg, Kappelle, & Biessels,
2008). And fourth, a population-based study spanning seven countries found that
cardiovascular risk factors were associated with dementia as long as four decades
prior to the outcome (Alonso et al., 2009). Although there has been some variation
among longitudinal population-based studies as to which particular symptom was
considered the largest vascular risk factor for dementia (Kloppenborg, van den Berg,
Kappelle, & Biessels, 2008), the point to underscore for the purpose of the present
study was the comorbidity found among many of these risk factors – that may have
been potentially explained by a pleiotropic effect (Launer, 2006) – and to be particularly
mindful of the need to target modifiable vascular risk in its totality.

A.3. Cognitive impairment and Latinos

Reviewing epidemiological studies conducted in predominantly Mexican
heritage populations in the Western U.S. (i.e., in Texas, Colorado, Arizona, New
Mexico, and in California, the latter including large community samples from Los
Angeles and the Sacramento Valley) indicated that the prevalence of dementia was
3.8%; the proportion of dementia of the AD type varied from 38.5% to 47%; and the

6

proportion of vascular dementia varied from 18% to 38.5% (Fitten, Ortiz, & Pontón,
2001; Haan et al., 2003; Yeo & Gallagher-Thompson, 1996).
Of those in Fitten et al.’s (2001) study, the proportion of individuals with AD was
lower and the proportion of those with vascular dementia was higher (Fitten, Ortiz, &
Pontón, 2001) than in White non-Latino populations (Brayne, Gill, & Huppert, 1995;
Evans et al., 1989; Hendrie, 1998). The presence of the apolipoprotein E epsilon4
(APOE ε4) allele has been long established as a risk factor for AD (Hendrie et al.,
2006; Hughes & Ganguli, 2009). However, the genetic evidence is mixed. That is, the
low frequency of the APOE ε4 allele, coupled with the strong presence of other
vascular risk factors for dementia (i.e., Type II diabetes, stroke, hypertension) – a
finding replicated in a few studies (Farrer et al., 1997; Maestre et al., 1995; Mayeux et
al., 2004) – suggests that the etiology of dementia in Latinos of Mexican descent differs
from that of other populations (Fitten, Ortiz, & Pontón, 2001; Haan et al., 2003).
Although some studies have belied this finding, and instead have found the allele to be
more robust and comparable to the risk ratios for the APOE ε4 allele in Anglo
populations (Duara et al., 1996; Harwood et al., 1999; Sevush, Peruyera, Crawford, &
Mullan, 2000; Tang et al., 1998), recent research has found that genetic factors are
associated with insulin metabolism, supporting the idea that there are genetic, vascular
and metabolic channels by which dementia arise (Watson et al., 2003). In sum, the
current research suggests that Latinos are cumulatively at increased risk for
developing dementia, and that risk may stem from vascular factors.

7
A.4. Diet and vascular risk

Recent research supporting the role of vascular risk in incident dementia has
led to encouraging lifestyle behavior modification as a strategy to offset this risk. One
such behavior change is dietary modification. Established long ago as a viable pathway
by which to reduce cardiovascular disease risk, dietary modification is receiving
newfound attention as a possible mechanism by which to reduce cerebrovascular
disease risk (Alzheimer's Association, 2007; American Association of Retired Persons,
2007; Gray, 1989; Hu & Willett, 2002). The goal of cerebrovascular-focused
interventions is to reduce risk for several vascular risk factors that are at least in part,
related to nutrition (Gonzales-Gross, Marcos, & Pietrzik, 2001), e.g., Type II diabetes,
atherosclerosis, hypo- and hyperglycemia, mid-life hypertension, and
hypercholesterolemia (Kalmijn, Fesken, Launer, Stijnen, & Kromhout, 1995; Stewart &
Liolitsa, 1999).
Although the mechanisms involved in the role of diet in cognitive decline have
yet to be fully understood, one focus is fat-based pathways. Fatty acids are
biochemical compounds that occur naturally (either singly or combined) and consist of
strongly linked carbon and hydrogen atoms in a chain structure (Berg, Tymoczko, &
Stryer, 2006). Fatty acids appear in two forms: saturated and unsaturated (Berg,
Tymoczko, & Stryer, 2006). Unsaturated fatty acids are further categorized as either
monounsaturated fatty acids or polyunsaturated fatty acids, also known as
monounsaturated and polyunsaturated fats, respectively (Berg, Tymoczko, & Stryer,
2006). Trans fatty acids, or trans fats, are positioned between unsaturated and
saturated fats; although they begin as unsaturated fats, a process known as partial

8

hydrogenation chemically alters the fatty acids to more closely resemble saturated fats
in structure and property (for a detailed review, see Mozaffarian, Katan, Ascherio,
Stampfer, & Willett, 2006). Essential fatty acids, or EFAs, have been identified as forms
of fatty acids that physiologically cannot be sufficiently synthesized and thus,
necessitate consumption by dietary intake (Yehuda, Rabinovitz, & Mostofsky, 1999).
Diets high in certain types of lipids (including saturated fats and trans fats) have
long been known to contribute to vascular risk (Hu & Willett, 2002). Dietary fat
modification has been a standard mechanism used to reduce the risk of, delay, or even
prevent the onset of cardiovascular disease and other related cardiovascular
conditions such as diabetes and hypertension (Bowen & Beresford, 2002; Hooper et
al., 2001; Yancey et al., 2004). Several areas of emerging research have suggested
potentially promising associations between dietary intake and cognitive status
(Salerno-Kennedy & Cashman, 2005; Scarmeas, Stern, Tang, Mayeux, & Luchsinger,
2006; Solfrizzi et al., 2006; Solfrizzi et al., 2005).
Attention has also been given to the role of dietary fat in the reduction of
dementia risk. In a review of animal studies, Yehuda and colleagues concluded that not
only did EFAs mediate the relationship between brain biochemistry and cognitive
functioning, but that the levels and ratios of EFAs (e.g., unsaturated to saturated fats)
acted as a potential moderator of adverse cognitive effects (Yehuda, Rabinovitz, &
Mostofsky, 1999).
Subsequent findings from several population-based longitudinal studies (study
duration provided in parentheses) – namely a U.S. community-based prospective
study, i.e., the Chicago Health and Aging Project (6+ years) (Morris, Evans, Bienias,
Tangney, & Wilson, 2004); a Dutch study, i.e., the Rotterdam Study, a prospective
9

population-based cohort study (2.1 years in current study; 20+ years in total); a Finnish
study comprised of the Finnish cohorts in the Seven Countries Study (40+ years in
total) (Notkola et al., 1998); and a cohort sampling from the Italian Longitudinal Study
on Aging (8.5 years) (Solfrizzi et al., 2006) – suggested that sustained intakes of
saturated (Kalmijn et al., 1997; Morris, Evans, Bienias, Tangney, & Wilson, 2004;
Solfrizzi et al., 2006) and trans fats (Morris, Evans, Bienias, Tangney, & Wilson, 2004)
adversely affected cognitive functioning (Kalmijn, 2000), and were shown to increase
risk for incident dementia (Kalmijn, 2000; Kalmijn et al., 1997; Notkola et al., 1998).
In contrast, diets high in monounsaturated and polyunsaturated fats (including
the n – 3 series, also known as the omega-3 fatty acids) were shown to longitudinally
reduce the risk of cognitive decline (Kalmijn, 2000; Kalmijn et al., 1997; Solfrizzi et al.,
2006). To conclude, these were similar suggestions presented by the International
Academy on Nutrition and Aging in its most conclusive position statement yet regarding
the association between nutrition and cognitive decline, in which the organization
concurred and suggested that a high intake of saturated and trans fats was positively
associated with increased risk for dementia, whereas a diet high in unsaturated fats
(both poly- and mono-) was protective against cognitive decline (Gillette-Guyonnet et
al., 2007).

A.5. Latinas as target audience

The basic scientific research findings discussed above, coupled with reduced
awareness of foods containing saturated fats among Latinos of Mexican descent

10

compared to non-Latino Whites (Knapp, Hazuda, Haffner, Young, & Stern, 1988;
Neuhouser, Thompson, Coronado, & Solomon, 2004), prompted the decision to target
Latinos of Mexican descent for this study. The reasons include: First, a public health
challenge is anticipated when cardiovascular risk is coupled with ongoing population
increases. Individuals of Mexican ancestry comprise 59% of the U.S.’s total Latino
population, and this segment is estimated to grow over time due to immigration and
reproduction patterns (U.S. Census Bureau, 2004b). Second, Latinos of Mexican
descent have elevated lifestyle risks for cardio- and cerebro- vascular disease (Aronow
& Ahn, 2001; Fitten, Ortiz, & Pontón, 2001; Ford & Giles, 2003; Goff, Nichaman, &
Chan, 1997; Muntner, Garrett, Klag, & Coresh, 2002). Third, the southwestern border
states (California, Arizona, New Mexico, and Texas) all have large populations of
Latinos of Mexican descent (Bureau of the Census, 2000). Fourth, approximately one-
third of all Latinos in the U.S. lived in California, with older Latinos in California being
predominantly of Mexican descent (80%) (The California Policy Research Center,
1998). And fifth, Latinas in particular were chosen due to their prominent role as
nutritional gatekeepers in their families, their leadership role in food procurement and
preparation, their potential ability to influence both personal and familial dietary
behavior change (Ayala et al., 2001; McIntosh & Zey, 1989; Wansink, 2002, 2005,
2006), and their relative accessibility as a study population compared to men (Varma et
al., 2004). It was important to examine whether an individual’s beliefs about the
centrality of family (termed familism) influenced dietary fat outcomes as an intervening
construct in light of: (1) the selection of family nutrition gatekeepers as the target
audience in the present study; (2) findings from an empirical review on nutrition
interventions that supported the inclusion of culturally-based constructs to help improve
11

understanding of dietary fat interventions (Baranowski, Weber Cullen, & Baranowski,
1999); and (3) support from the cardiovascular disease prevention literature that
suggested health promotion interventions are influenced by cultural mores (Chyun,
Amend, Newlin, Langerman, & Melkus, 2003). Described in detail in A.9.b., familism is
defined as a collective loyalty to the family that supersedes the needs of the individual
(Flores, 2000), and is considered a cultural construct that is common in several ethnic
groups including Latinos.

A.6. Theoretical foundation

The rationale for the present study was built on epidemiological and
demographical evidence, and structured by a theoretical framework. The theoretical
backbone was constructed from the disciplines of risk communication and health
behavior as described next.

A.6.a. Risk communication

Mental models, whose origins are ascribed to Craik (1943), have become a
framework in recent years against which risk analysis and risk communication studies
hang (Atman, Bostrom, Fischhoff, & Morgan, 1994). In simple terms, mental models
within the context of disease prevention and health promotion research are cognitive
representations of: one’s belief about one’s own vulnerability to the disease; the
cognitive interaction between this vulnerability and the defenses one can enact to
combat this risk; and one’s belief about one’s own ability to enact these defenses
12

(Griffith, Dunwoody, & Neuwirth, 1999; Lundgren & McMakin, 2009; Morgan, Fischhoff,
Bostrom, & Atman, 2002). In the field of health risk communication, mental models
provide a heuristic by which to craft: a road map to identify an individual’s perceived
vulnerability to a disease; cognitive behavioral strategies to help the individual reduce
his or her perceived vulnerability to a disease; and, approaches to effectively build self-
efficacy and skills so as to enable the individual towards successful health behavioral
change (Griffith, Dunwoody, & Neuwirth, 1999; Kahlor, Dunwoody, Griffin, Neuwirth, &
Giese, 2003; Lundgren & McMakin, 2009; Morgan, Fischhoff, Bostrom, & Atman,
2002).
Diet is an extremely complex behavior expressed as a systematic pattern that is
the end result of an intricate series of several decisions (Baranowski, 1997; Campbell
& Desjardins, 1989). Hence, the mental model for risk communication was invoked as
an additional cognitive decisionmaking overlay onto principles of health behavior theory
described next. Together, the two perspectives guided the theoretical framework for
the present study.

A.6.b. Health behavior theory

Health behavior theory drove the development of the present study in two key
ways. The study, including design, measures, and intervention were based on the
Health Belief Model (HBM) (Rosenstock, 1974), with additional theoretical
considerations specific to the intervention, e.g., Social Cognitive Theory (SCT)
(Bandura, 1986). The additional theoretical considerations specific to the intervention
are explained in B.2.
13

Brain-focused public health interventions as conducted by the Alzheimer’s
Association, AARP, and the Centers for Disease Control and Prevention have been
implicitly premised on the assumption that individuals feel vulnerable to dementia
(Alzheimer's Association, 2007; American Association of Retired Persons, 2007;
Centers for Disease Control and Prevention and the Alzheimer's Association, 2007).
Termed perceived vulnerability or perceived susceptibility, this construct zeroes in on
an individual’s subjective risk of his or her developing dementia (Glanz, Rimer, &
Viswanath, 2008). The HBM is characterized as a value expectancy theory where the
yearning to prevent disease or ameliorate existing symptoms (value) is coupled with
the belief that a certain health behavior (e.g. dietary fat modification) would prevent or
ameliorate cardiovascular and cerebrovascular disease (expectancy) (Glanz, Rimer, &
Viswanath, 2008). The HBM was selected as the theoretical backbone for the present
study because the model prominently features perceived susceptibility (Glanz, Rimer,
& Viswanath, 2008). The HBM theorizes that perceived susceptibility – when parceled
with knowledge and informed by skills to reduce perceived barriers and increase self-
efficacy as discussed below – would lead to behavior change (Rosenstock, 1974). Vast
amounts of research focused on the expansion and application of the HBM have
suggested that the presence of perceived susceptibility to a disease is one of the
strongest contributors to individuals’ decision to enact preventive health behaviors
(Glanz, Rimer, & Viswanath, 2008; Janz & Becker, 1984), laying claim to the
importance of including perceived susceptibility as a moderator in the present study
which focused on disease prevention/risk reduction strategies, i.e., dietary fat
behaviors. For example, health behavior research indicates that perceived
susceptibility to cardiovascular risk factors such as hypertension,
14

hypercholesterolemia, angina pectoris, myocardial infarctions, and strokes are
positively correlated with behavioral change (Janz, 1988; Silagy, Muir, Coulter,
Thorogood, & Roe, 1993; Winkleby, Flora, & Kraemer, 1994) such as lower fat
consumption (Brunner, Rees, Ward, Burke, & Thorogood, 2007; Fuchs, Heath, &
Wheeler, 1992). This body of evidence led to this model’s becoming the basis for the
theoretical construction of the present study, whereby the investigator hypothesized
that individuals in the intervention conditions with a higher level of perceived
susceptibility to vascular disease at pretest would demonstrate greater improvements
in dietary fat knowledge, intentions, and self-reported behaviors from pretest to posttest
and pretest to follow-up compared to those with a lower level of perceived susceptibility
to vascular disease.

A.7. Dietary fat outcomes

The theoretical foundation for this study led to the prediction of several aspects
of dietary fat outcomes in this study (for a full review, see Ammerman, Lindquist, Lohr,
& Hersey, 2002), namely dietary fat knowledge, dietary fat intentions, and self-reported
dietary fat behaviors (i.e., avoiding, substituting, modifying, and replacing fats).

A.7.a. Dietary fat knowledge

Assessment of dietary knowledge is routinely used in nutrition interventions to
evaluate the impact of the intervention (Howard-Pitney, Winkleby, Albright, Bruce, &
Fortmann, 1997; Sapp & Jensen, 1997) and to assess whether the content of the
15

intervention was effectively conveyed (Wardle, Parmenter, & Waller, 2000). Based on
the previously described empirical research and health behavior theories, dietary fat
knowledge was expected to significantly differentially increase from pretest to posttest
and from pretest to follow-up in the intervention conditions compared to the pretest-
posttest condition and in the heart plus brain condition compared to the heart only
condition.

A.7.b. Dietary fat intentions

Behavioral intentions are defined as a motivational cognition associated with an
individual’s engagement in a behavioral performance (Armitage, 2004; for a review,
see Sheeran, 2002). The consideration of dietary fat intentions is customary in nutrition
interventions where the follow-up period is brief and dietary change may not have
transpired yet (Armitage, 2004; Sheeran, 2002), such as the one-month follow-up
period used in the present study.
Health behavior theory supports the conceptualization of dietary fat intentions
on a continuum (Abraham, 2008; Schwarzer, 2008). Consequently, this was the
primary conceptualization of this construct so as to capture an individual’s intention to
alter her dietary fat behaviors. A secondary conceptualization was to consider change
in their stage of change, defined as a discrete point of an individual’s readiness to
change and conventionally labeled as one of five stages, namely: precontemplation,
contemplation, preparation, action, and maintenance (Prochaska & DiClemente, 1986).
In this study, dietary fat intentions and stage were expected to significantly differentially
increase from pretest to posttest and from pretest to follow-up in the intervention
16

conditions compared to the pretest-posttest condition and in the heart plus brain
condition compared to the heart only condition.

A.7.c. Self-reported dietary fat behaviors

The ultimate goal of the intervention was to change dietary fat behaviors. To
note, all dietary fat behaviors examined in this study were self-reported. Evidence from
the field of nutrition research suggests that, measuring self-reported fat-reducing
behaviors is more sensitive to dietary change than measuring self-reported dietary
intake, e.g., food frequency (Kristal, Beresford, & Lazovich, 1994). In this study, self-
reported dietary fat behaviors were expected to significantly differentially increase from
pretest to posttest and from pretest to follow-up in the intervention conditions compared
to the pretest-posttest condition and in the heart plus brain condition compared to the
heart only condition.

A.8. Previous dietary fat outcome research among Latinas

Published literature of randomized-controlled trials for health promotion
intervention efforts to reduce dietary fats among community-dwelling younger to
middle-aged Latinas of Mexican descent is scant. Two interventions meeting these
criteria were found: (1) Secretos de la Buena Vida, a promotora de salud (lay peer
health educator)-led, randomized community-based study that sought to promote
health through a multi-faceted dietary behavioral intervention with Latinas of
predominantly Mexican descent (Elder et al., 2006); and (2) Mujeres Felices por ser
17

Saludables, a randomized intervention study to evaluate the efficacy of a combined
dietary and breast health intervention (Fitzgibbon, Gapstur, & Knight, 2004). Two
additional programs also met some, but not all of the criteria: (3) the Stanford Nutrition
Action Program (SNAP), a randomized dietary fat reduction intervention in which most
of the participants were young rather than middle-aged, primarily U.S. born Latinas of
Mexican descent (Howard-Pitney, Winkleby, Albright, Bruce, & Fortmann, 1997); and
(4) Salud para su Corazón, a cardiovascular disease prevention and outreach
intervention model aimed at a Latino sample of heterogeneous descent, but not
including any control condition (Alcalay, Alvarado, Balcazar, Newman, & Huerta, 1999).
All four studies collected data at least twice (i.e., before and after the intervention).
Overall, findings from these studies revealed the interventions’ ability to decrease
dietary fat consumption among Latinas. SNAP participants also made significant gains
in dietary fat knowledge and attitudes about fat consumption (Howard-Pitney,
Winkleby, Albright, Bruce, & Fortmann, 1997). Dietary fat-related behavior changes
(e.g., decreased intake of saturated fat and/or total fat) were found at posttest in all of
the studies. Follow-up assessment indicated that these changes were maintained up
to: three months (Howard-Pitney, Winkleby, Albright, Bruce, & Fortmann, 1997); eight
months (Fitzgibbon, Gapstur, & Knight, 2004); and one year, although the significant
group contrast between a promotora-led condition and a tailored communication
condition in the study dissipated during this long-term follow-up (Elder et al., 2006). A
longer-term assessment was absent from the Salud para su Corazón study (Alcalay,
Alvarado, Balcazar, Newman, & Huerta, 1999).

18

A.9. Intervening variables

To better understand the intervention used in the present study, two variables
were examined for their possible contributions as moderators: (1) susceptibility and (2)
familism. Specifically, three of the four susceptibility constructs (i.e., perceived
susceptibility to vascular disease, self vascular risk, and family vascular risk) and
familism were identified as factors that would influence the magnitude of behavioral
change for those in the two intervention conditions, but not for those in the pretest-
posttest condition; therefore an interaction effect was hypothesized for these constructs
(Hayes, Laurenceau, & Cardaciotto, 2008). A fourth susceptibility construct (i.e.,
perceived susceptibility to AD) was expected to behave differentially between
intervention conditions.

A.9.a. Susceptibility as a moderator

Four measures of susceptibility were examined in the present study to
determine whether they influenced study outcomes: (1) perceived susceptibility to
vascular disease; (2) perceived susceptibility to AD; (3) one’s own vascular risk profile
(or self vascular risk); and (4) family history of vascular risk (or family vascular risk).

Perceived susceptibility to vascular disease. Perceived susceptibility to
disease (Rosenstock, 1982), and in particular to vascular disease – including
cardiovascular heart disease (Rimal, 2001; Rimal & Real, 2003; Toft et al., 2007),

19

stroke (Samsa et al., 1997), and diabetes (Boulware, Carson, Troll, Powe, & Cooper,
2009; Gallivan, Brown, Greenberg, & Clark, 2009) – has been shown to moderate
healthy behaviors, including those which are dietarily-based. Thus, this study
hypothesized that a higher level of perceived susceptibility to vascular disease at
pretest would moderate dietary fat outcomes; that is, Latinas in the intervention
conditions who reported a higher level of perceived susceptibility to vascular diseases
at pretest would have greater improvements in dietary fat outcomes from pretest to
posttest and from pretest to follow-up compared to those who showed a lower level of
perceived susceptibility to vascular diseases.

Perceived susceptibility to AD. As previously detailed, brain-focused public
health interventions have been implicitly premised on perceived susceptibility to AD
(Glanz, Rimer, & Viswanath, 2008). Although the present study is the first known study
to evaluate whether perceived susceptibility to AD moderates dietary fat behavior,
recent findings indicate that perceived susceptibility to AD moderates other AD-related
health behaviors, such as cognitive screenings to assess AD. For example, a
population-based study utilizing a random-digit dialing strategy found that perceived
susceptibility to AD was a predictor of behavioral intention to screen for AD in a sample
of individuals without dementia (Galvin, Fu, Nguyen, Glasheen, & Scharff, 2008),
suggestive of the notion that perceived susceptibility to AD may influence other health
behaviors. Thus, this study hypothesized that a higher level of perceived susceptibility
to AD at pretest would moderate dietary fat outcomes; that is, Latinas in the heart plus
brain condition who reported a higher level of perceived susceptibility to AD at pretest
would have greater improvements in dietary fat outcomes from pretest to posttest and
20

from pretest to follow-up compared to Latinas with a lower level of perceived
susceptibility to AD in the heart only condition.

Self vascular risk. Epidemiological research shows that a personal history of
vascular disease or associated risk factors such as diabetes or obesity increases one’s
own vascular risk (Hendrie et al., 2006; Launer, 2006; Stroup-Benham, Markides,
Espino, & Goodwin, 1999). Research examining self-reported data suggested that
individuals with more self-reported cardiovascular risk were more likely to consider
their diet as a health risk and were in turn, more likely to improve their diet compared to
individuals at less risk for cardiovascular disease (Chow et al., 2010; Silagy, Muir,
Coulter, Thorogood, & Roe, 1993). Thus, it has been shown that the presence of self
vascular risk influences lifestyle modifications including dietary fat behavior change so
as to reduce self vascular risk (Hanlon et al., 1995). This study hypothesized that high
vascular risk in oneself moderated dietary fat outcomes; that is, Latinas in the
intervention conditions who reported greater self risk for vascular diseases at pretest
would have greater improvements in dietary fat outcomes from pretest to posttest and
from pretest to follow-up compared to those who had less self risk for vascular
diseases.

Family vascular risk. Similarly, epidemiological research indicates that a
family history of vascular disease or associated risk factors such as diabetes or obesity
increases one’s own vascular risk (Grundy et al., 1999; Leander, Hallqvist, Reuterwall,
Ahlbom, & de Faire, 2001; Yoon et al., 2002). In turn, it was expected that the
presence of family vascular risk would influence lifestyle modifications including dietary
21

fat behavior change so as to reduce vascular risk (McCusker et al., 2004; Yoon et al.,
2002). This study hypothesized that high vascular risk in one’s family moderated
dietary fat outcomes; that is, Latinas in the intervention conditions who reported greater
family risk for vascular diseases at pretest would have greater improvements in dietary
fat outcomes from pretest to posttest and from pretest to follow-up compared to those
who had less family risk for vascular diseases.

A.9.b. Familism as a moderator

As stated earlier, findings from research literature – coupled with the selection
of family nutrition gatekeepers for the target audience in this study – established
support for the inclusion of familism as a moderator. In a critical empirical review of
culturally-tailored interventions promoting nutrition behaviors among Latinos, Mier, Ory,
and Medina (2010) indicated that familism was a core value among Latinos and
subsequently, one that may influence dietary behaviors. In speaking with local
promotoras de salud from Esperanza Housing Coalition during pilot testing (described
in detail later), preservation of their own health was described as important in the
context of Latinas’ desire to maintain their role as caregivers, a sentiment echoed in
the research literature whereby Latino caregivers who adhered strongly to familism not
only placed a higher value on their family but also were more willing to provide for the
family (Knight et al., 2002). This study hypothesized that high familism moderated
dietary fat outcomes; that is, Latinas in the intervention conditions who showed greater
familism at pretest would have greater improvements in dietary fat outcomes from

22

pretest to posttest and from pretest to follow-up compared to those who had less
familism.

A.9.c. Covariates

Based on prior empirical work, four constructs were considered as covariates:
(1) general acculturation, (2) dietary acculturation, (3) presence of young children in
home, and (4) health literacy. Specifically, these constructs were identified as factors
that would make behavioral change difficult across study conditions (i.e., both the
intervention conditions and the pretest-posttest condition) from pretest to posttest and
from pretest to follow-up.
The first and second constructs concerned the role of acculturation. Two types
of acculturation were assessed: general and dietary. General acculturation was defined
as the process that occurs when a minority group adopts the day-to-day practices of
the host country, whereas dietary acculturation specifically refers to the process when
a minority group adopts eating patterns/food choices of the host country (Cabassa,
2003; Negy & Woods, 1992; Satia-Abouta, Patterson, Neuhouser, & Elder, 2002; Satia
et al., 2001). Latinos of Mexican descent residing in the U.S. had higher rates of
cardiovascular disease compared to those living in their home countries (Sundquist &
Winkleby, 1999), arguing for the effect of acculturation on risk for cardiovascular
disease (Satia-Abouta, Patterson, Neuhouser, & Elder, 2002). For instance, diet may
be adversely affected by the acculturation process, as Latinos tend to move away from
their native fiber-rich foods such as beans and legumes, and instead adopt less

23

healthful eating patterns such as increased fat intake (Ayala et al., 2001; Chavez,
Persky, Langenberg, & Pestano-Binghay, 1994). In this study, the investigator
predicted that being more versus less acculturated at pretest would make behavioral
change more difficult across study conditions (i.e., both the intervention conditions and
the pretest-posttest condition) from pretest to posttest and from pretest to follow-up.
Third, the number of children younger than 17 years of age living at home was
considered as a covariate. Research examining cross-sectional data from the
NHANES III suggested that the presence of children in this age group living at home
was associated with significantly higher total fat and saturated fat consumption
(Laroche, Hofer, & Davis, 2007). In this study, the investigator predicted that having
more children in the home versus fewer or no children at pretest would make
behavioral change more difficult across study conditions (i.e., both the intervention
conditions and the pretest-posttest condition) from pretest to posttest and from pretest
to follow-up.
Fourth, health illiteracy, or the inability to obtain, process, and understand basic
health information and services needed to make approximate health decisions (Weiss
et al., 2005), was identified as a risk factor for poorer health status and less knowledge
of one’s health problems (Howard-Pitney, Winkleby, Albright, Bruce, & Fortmann,
1997). In this study, the investigator predicted that possessing lower health literacy
would limit the extent of behavioral change, i.e., those with lower versus higher health
literacy at pretest would change less across study conditions (i.e., both the intervention
conditions and the pretest-posttest condition) from pretest to posttest and from pretest
to follow-up.

24

A.10. Feedback and motivation

Position statements by the American Heart Association, coupled with empirical
reviews and original findings from health behavioral intervention research – including
those found in randomized clinical trials of dietary interventions – recommend
feedback, problem solving, and motivational messages (Bond & Wing, 2009; DiLillo,
Siegfried, & West, 2003; Howard-Pitney, Winkleby, Albright, Bruce, & Fortmann, 1997;
Klein et al., 2004; Kumanyika et al., 2008; Turk et al., 2009). Findings culled from these
articles prompted the investigator to develop, implement, and test the efficacy of a
motivational component, as previous research suggested that the incorporation of a
motivational piece was important to improve and sustain complex dietary change such
as fat reduction (Bond & Wing, 2009; DiLillo, Siegfried, & West, 2003; Howard-Pitney,
Winkleby, Albright, Bruce, & Fortmann, 1997; Klein et al., 2004; Kumanyika et al.,
2008; Turk et al., 2009). A randomized controlled trial of a health communication
intervention for Latinas predominantly of Mexican descent found that reinforcement in
several formats, including telephone calls, may have led to greater decreases in self-
reported dietary fat intake compared to individuals not receiving the reinforcement
(Elder et al., 2006). As such, the present study assessed whether a motivational
component altered the dietary fat outcomes. In this study, the investigator predicted
that participants who received a motivational boost would demonstrate greater
improvements in their dietary fat outcomes compared to those that did not receive the
boost.

25

A.11. Treatment integrity variables

Treatment integrity, also called treatment fidelity, was integrated into the study
to evaluate whether the intervention provided to the participants was in accord with the
treatment protocols (A. Nezu & C. M. Nezu, 2008). Several aspects of treatment
integrity were identified and assessed in the present study, i.e., manipulation checks
(discussed next); horizontal diffusion (whether participants across conditions shared
information); self-reported facilitator adherence (whether facilitators adhered to the
particular intervention condition at any given time); self-reported facilitator competence
(whether facilitators felt competent in delivering the intervention condition); and
participant satisfaction. In standard fashion, the manipulation checks were
administered at posttest (Levine & Parkinson, 1994). The manipulation checks served
a tri-fold purpose: (1) to evaluate whether participants in the heart plus brain condition,
who heard about the dementia-diet connection in their workshops, were subsequently
more aware of the brain-diet association (i.e., to dementia) than those in the heart only
condition, whose workshops did not mention the dementia-diet connection; (2) to
assess whether participants in both of the intervention conditions, whose workshops
presented information that nutrition is relevant to cardiovascular health, subsequently
shown higher awareness of this connection compared to the pretest-posttest condition;
and (3) to evaluate whether perceived susceptibility to AD at posttest for the heart plus
brain condition was greater than those in the heart only condition, given that the
workshops for the heart plus brain condition discussed susceptibility to AD and that it
was affected by diet.

26

A.12. Study hypotheses

In this study, the investigator examined 14 hypotheses:

Hypothesis 1/Heart Plus Brain Health & Dietary Fat Knowledge: The investigator
hypothesized that Latinas would demonstrate greater improvements in dietary fat
knowledge from pretest to posttest and from pretest to follow-up in the heart plus brain
and heart only conditions compared to the pretest-posttest condition, and in the heart
plus brain condition compared to the heart only condition.

Hypothesis 2/Heart Plus Brain Health & Dietary Fat Intentions: The investigator
hypothesized that Latinas would demonstrate greater improvements in dietary fat
intentions from pretest to posttest and from pretest to follow-up in the heart plus brain
and heart only conditions compared to the pretest-posttest condition, and in the heart
plus brain condition compared to the heart only condition.

Hypothesis 3/Heart Plus Brain Health & Dietary Fat Behaviors: The investigator
hypothesized that Latinas would demonstrate greater improvements in self-reported
dietary fat behaviors from pretest to posttest and from pretest to follow-up in the heart
plus brain and heart only conditions compared to the pretest-posttest condition, and in
the heart plus brain condition compared to the heart only condition.

Hypothesis 4/Motivational Boost: The investigator hypothesized that a randomly
selected group of Latinas who received a motivational boost between posttest and
27

follow-up would demonstrate greater improvements in dietary fat knowledge,
intentions, and self-reported behaviors at follow-up, compared to those that did not
receive the motivational boost.

Hypothesis 5/Posttest Condition: The posttest-only condition was used to assess for
two key threats to internal validity, i.e., historical and testing effects (Kazdin, 2003),
specifically pretest sensitization (A. M. Nezu & C. M. Nezu, 2008). The investigator
hypothesized that if the scores for the posttest-only condition were higher than pretest
scores for the pretest-posttest condition, but similar to posttest scores for the pretest-
posttest condition, it would suggest a historical effect. Whereas, if the posttest-only
condition looked similar to pretest scores for the pretest-posttest condition, and the
pretest-posttest condition improved, the investigator hypothesized that there may have
been a pretest sensitization effect.

Hypothesis 6/Perceived Susceptibility to Vascular Disease as Moderator: The
investigator hypothesized that Latinas in the intervention conditions who had a higher
level of perceived susceptibility to vascular disease at pretest would demonstrate
greater improvements in dietary fat knowledge, intentions, and self-reported behaviors
from pretest to posttest and from pretest to follow-up compared to those that had a
lower level of perceived susceptibility to vascular disease.

Hypothesis 7/Perceived Susceptibility to AD as Moderator: The investigator
hypothesized that Latinas in the heart plus brain condition who had a higher level of
perceived susceptibility to AD at pretest would demonstrate greater improvements in
28

dietary fat knowledge, intentions, and self-reported behaviors from pretest to posttest
and from pretest to follow-up compared to those that had a lower level of perceived
susceptibility to AD in the heart only condition.

Hypothesis 8/Self Vascular Risk as Moderator: The investigator hypothesized that
Latinas in the intervention conditions who possessed more self vascular risk at pretest
would demonstrate greater improvements in dietary fat knowledge, intentions, and self-
reported behaviors from pretest to posttest and from pretest to follow-up compared to
those that had less self vascular risk.

Hypothesis 9/Family Vascular Risk as Moderator: The investigator hypothesized
that Latinas in the intervention conditions who possessed more family vascular risk at
pretest would demonstrate greater improvements in dietary fat knowledge, intentions,
and self-reported behaviors from pretest to posttest and from pretest to follow-up
compared to those that had less family vascular risk.

Hypothesis 10/Familism as Moderator: The investigator hypothesized that Latinas in
the intervention conditions who had a higher level of familism at pretest would
demonstrate greater improvements in dietary fat knowledge, intentions, and self-
reported behaviors from pretest to posttest and from pretest to follow-up compared to
those that had a lower level of familism.

Hypothesis 11/General Acculturation as Covariate: The investigator hypothesized
that Latinas who were less acculturated at pretest would demonstrate greater
29

improvements in behavioral change (i.e., dietary fat knowledge, intentions, and self-
reported behaviors) across all study conditions (i.e., both the intervention conditions
and the pretest-posttest condition) from pretest to posttest and from pretest to follow-up
than Latinas who were more acculturated.

Hypothesis 12/Dietary Acculturation as Covariate: The investigator hypothesized
that Latinas who were less dietarily acculturated at pretest would demonstrate greater
improvements in behavioral change (i.e., dietary fat knowledge, intentions, and self-
reported behaviors) across all study conditions (i.e., both the intervention conditions
and the pretest-posttest condition) from pretest to posttest and from pretest to follow-up
than Latinas who were more dietarily acculturated.

Hypothesis 13/Number of Children under the Age of 17 at Home as Covariate:
The investigator hypothesized that Latinas who resided with none or fewer children
under the age of 17 at pretest would demonstrate greater improvements in behavioral
change (i.e., dietary fat knowledge, intentions, and self-reported behaviors) across all
study conditions (i.e., both the intervention conditions and the pretest-posttest
condition) from pretest to posttest and from pretest to follow-up than Latinas who
resided with more children under the age of 17.

Hypothesis 14/Health Literacy as Covariate: The investigator hypothesized that
Latinas who were more health literate at pretest would demonstrate greater
improvements in behavioral change (i.e., dietary fat knowledge, intentions, and self-
reported behaviors) across all study conditions (i.e., both the intervention conditions
30

and the pretest-posttest condition) from pretest to posttest and from pretest to follow-up
than Latinas who were less health literate.
31

B. INTERVENTION

B.1. Overview of Buenos Habitos Alimenticios para una Buena Salud

A systematic review (for a detailed review, see Brunner, Rees, Ward, Burke, &
Thorogood, 2007) indicated that dietary advice, including of the fat-based type, was
associated with reductions in fat intake (particularly among women) and modifications
of cardiovascular risk factors. Moreover, a focus-group-based qualitative study that
examined message framing to reduce dietary risk for hypertension among Latinos and
African Americans suggested the use of culturally-tailored nutrition interventions that
are aimed at reducing cardiovascular risk factors (Horowitz, Tuzzio, Rojas, Monteith, &
Sisk, 2004; McMahon, Cathorall, & Romero, 2007). However, prior research indicated
that there were several limitations to nutrition interventions targeting Latinos, including:
difficulty following the recommendations in the context of their family lives, social
situations, and cultures; recommended foods being expensive; and recommendations
being an unwelcome departure from traditional and preferred diets, as well as being
socially isolating (Fitzgibbon, Gapstur, & Knight, 2004; Horowitz, Tuzzio, Rojas,
Monteith, & Sisk, 2004). Consequently, the intervention – based on health behavior
theory and existing best practices for a heart healthy dietary modification – was
developed to facilitate adoption of healthy dietary fat behaviors for Latinas of Mexican
descent.
Translated into “Good Eating Habits for Good Health”, Buenos Habitos
Alimenticios para una Buena Salud (BHA) was created to address many of these
barriers. The program name was created in collaboration with bicultural and bilingual
32

promotoras de salud so as to be linguistically and culturally appropriate. The
intervention that all conditions received was dubbed BHA. The brain health information
received by the heart plus brain health condition heavily angled on its emphasis of
hypertension, hypercholesterolemia, diabetes, and obesity relative to brain health, as
the evidence was strongest in this area (Alzheimer's Association, 2007; Centers for
Disease Control and Prevention and the Alzheimer's Association, 2007).
BHA was developed specifically for Latinas, particularly those of Mexican
descent. BHA was presented as two, two-hour workshops with two to six participants
per workshop. A published review of the research indicated a range of contact time
associated with the effectiveness of fat-based nutrition interventions that spanned from
one contact per participant to 50 hours of counseling over four years (Brunner, Rees,
Ward, Burke, & Thorogood, 2007). The length of time for each workshop was
established during pilot testing, and based on discussions with local promotoras de
salud who had extensive experience working with Latinos in East Los Angeles and had
witnessed their competing demands.

B.2. Application of theory

The goal of the intervention was to develop a theoretically- and empirically-
driven, culturally-tailored nutrition education intervention. Thus, the development of the
intervention was informed by health behavior theory, our own pilot testing, and
published formative nutrition education research conducted among Latinas of Mexican
descent, to support emphasis on: reducing barriers, increasing self-efficacy, and
providing cues to action by way of developing skills (Ayala et al., 2001). These
33

recommendations were fully consistent with behavior change principles supported by
the HBM and the SCT, i.e., perceived barriers, self-efficacy, and cues to action
(Bandura, 1977a, 1986; Rosenstock, 1974; Rosenstock, Stretcher, & Becker, 1988).
These theories would suggest that dietary behavior is: influenced by prior beliefs and
attitudes; incremental; supported by reinforcement; built into a habitual nature over a
lifetime; and inclusive of cognitions and skills (Rosenstock, 1982).

Barrier reduction. Micro- and macro-level barriers to fat-reduction were targeted
in the intervention. For example, personal and familial barriers were shared and
discussed by way of motivational interviewing strategies used during workshops.
Addressing barriers in the context of family was a key health communication strategy
identified for Latinos (Elder, Ayala, Parra-Medina, & Talavera, 2009). Additionally,
helpful tools to promote sustained change were introduced in the workshops, such as
low-cost, easy to adopt fat modification strategies that simultaneously preserved and/or
enhanced taste (which was a key barrier identified in pilot testing). On a macro level, a
comprehensive list of food stores in a 10 to 20 mile radius was distributed to
participants in the intervention conditions, as a condensed alternative guide to the
readily available mom-and-pop markets, and liquor and convenience stores mainly
known to supply processed, high fat products.

Self-efficacy development. Although self-efficacy was excluded in the original
HBM (Rosenstock, 1974), it was later incorporated into a revised model (Rosenstock,
Stretcher, & Becker, 1988) following the introduction of self-efficacy as a concept within
the SCT (Bandura, 1977a, 1977b, 1986). SCT is characterized as a self-regulatory
34

model that provides a theoretical infrastructure by which individuals when provided with
the necessary cognitions, beliefs, and skills can take effective action to reduce
personal risk (Finnegan & Viswanath, 2008).
The motivation to draw from this theoretical infrastructure stems from research
findings that bolstering self-efficacy increases healthier cooking among Latinas of
Mexican descent (Ayala et al., 2001). The self-efficacy promotion strategies used in the
present study were modeled after similar strategies developed for a randomized
controlled trial that effectively increased healthy dietary fat behaviors among Latinas of
Mexican descent by its use of culturally relevant food props, techniques, and materials
to promote ease to determine dietary fat content (Ayala et al., 2001; Elder et al., 2006).
These strategies included (1) nutrition label reading; (2) fast, easy, delicious recipes
conveniently made at home (cookbooks were provided); (3) a virtual presentation
depicting a grocery store tour and strategies to promote healthier, low-fat selections;
and (4) an in-class hands-on cooking demonstration and tasting to provide an
opportunity for skills modeling and positive reinforcement (Bandura, 1986, 1997). Other
areas targeted by the intervention (e.g., reducing barriers and providing cues to action)
were also considered to bolster self-efficacy. Additionally, the intervention was
presented in a group-based format, in adherence with cultural preferences mentioned
by promotoras de salud and social learning strategies, which suggest that groups can
provide important social support towards behavioral modification (Bandura, 1986).
Lastly, all intervention participants received a certificate of completion at the conclusion
of the workshops in accord with pilot testing feedback recommending this as a strategy
to increase participants’ self-efficacy towards dietary fat modification (see Appendix A).

35

Cues to action. Several skills were taught in the workshops. Using culturally
relevant food props and techniques, several easy to adopt dietary fat behaviors were
demonstrated, including how to: (1) obtain a healthy portion size; (2) substitute with
easily available low fat items; (3) set incremental goals to sustained health behavior
change; and (4) prepare foods using lower-fat methods. For example, in a study
among Latinas of Mexican descent, frying was the most common food preparation
technique directly observed in participants’ kitchens (Ayala et al., 2001). None of the
participants were observed preparing foods with lower-fat cooking techniques such as
steaming or broiling. In addition, a simple, easy-to-follow hands-on cooking
demonstration (Ayala et al., 2001) was conducted as mentioned previously.
Participants consumed the resulting nutritious, low-fat dish.

B.3. Pilot testing

In the summer of 2007, the investigator and two undergraduate bilingual and
bicultural research assistants pilot tested preliminary study materials in English and
Spanish. Pilot testers were told the purpose of the exercise was to help design a
nutrition class promoting fat reduction for Latinas in Southern California. The goal of
pilot testing for this study was to ensure: clarity of items and response options, item
modification, ecological validity, and in general, to ensure culturally appropriate
methods of assessment and intervention (Globe et al., 2002). A summary of the pilot
test findings is provided in Appendix B.
Because involving members of the target population in the design of the
nutritional intervention was seen as an important way to ensure that the target
36

audience’s preferences, needs, and characteristics were reflected (Hornik & Kelly,
2007), the researcher made arrangements with local organizations, i.e., Esperanza
Housing Coalition’s Promotoras Group (a grassroots health and social services agency
targeting Latinos), Compromiso y Visión (a statewide network of promotoras),
Sycamore-Hathaway Child and Family Services Center’s Promotoras de Salud para
Nutrición group (a nutrition-specific promotoras de salud group), and California
Hospital Medical Center – Catholic Healthcare West’s Promotoras De Salud Group
(community medical center-based promotoras’ group) to recruit a total of 15
promotoras de salud as pilot testers. These organizations shared demographic
characteristics with the USC Los Angeles Latino Eye Study (LALES) population and
had knowledge of best practices in health promotion for the Latino population residing
in East Los Angeles. All of the promotoras had extensive experience promoting healthy
nutrition in area Latino communities (range = 3 to 12 years). Pilot testing was held at
their respective offices and lasted two hours. Both monolingual (Spanish) and bilingual
pilot testers were recruited. Pilot testing was conducted in Spanish and English. Pilot
testers were compensated for their time with a $10 Target gift card. Residents from the
LALES sampling area were excluded to minimize study contamination.
These pilot testing findings augmented other pilot testing sessions conducted
previously in the LALES sample prior to the LALES baseline study (Globe et al., 2002).
Recommendations made in that study matched suggestions reiterated during pilot
testing. The identified challenges were addressed in similar ways (Globe et al., 2002;
Varma et al., 2004). First, to accommodate a large proportion of individuals who
worked outside the home, provided childcare, or both, research staff was available to

37

conduct interviews and to facilitate interventions on weekdays, weekday evenings, and
weekends. Second, to accommodate language barriers and trust issues, given the
presence of many first-generation immigrants who do not speak English and the
relatively high vigilance for immigration officials, bilingual and/or bicultural
undergraduate research assistants were a part of the research team. Finally, because
many Latinas in Los Angeles did not have access to child care facilities and, therefore,
had nowhere to leave their children or grandchildren when asked to participant in
research studies, research assistants and the investigator provided free child care in-
clinic. Findings from the baseline LALES study indicated that provision of child care
enhances study participation (Varma et al., 2004).

B.4. Intervention development

Considerable effort was put forth into the development and refinement of an
evidence-based manualized intervention, conducted in collaboration with a local
promotora de salud, a local diabetes and AD health educator, and the aforementioned
pilot testers. All of the collaborators had extensive experience educating Latinos in
East Los Angeles about diet and health. The intervention was further refined with five
bilingual employees of USC’s Housekeeping Services in spring 2008 to ensure
linguistic and cultural appropriateness of the intervention.
The intervention focused on reducing dietary fat barriers, and building dietary
fat self-efficacy and healthy dietary fat behaviors. Previous research indicated that
these areas were the most requested areas of change by Latina participants involved

38

in dietary interventions targeting Latinas of Mexican descent (Ayala et al., 2001).
Additionally, this researcher’s discussions with a highly experienced promotora de
salud and a local diabetes and AD health educator reaffirmed these findings for Latinas
of Mexican descent residing in East Los Angeles.
The “Brain Connection” curriculum drew on the latest science that indicated an
association between fat consumption and cognitive health. Specifically, the curriculum
included content about: (1) the relationship between metabolic syndrome and
increased risk for dementia; (2) a visual representation in which a non-pathological
brain is compared to a Alzheimer’s diseased-brain; (3) the relationship between
unsaturated fat consumption and reduced risk of cardiovascular as well as
cerebrovascular diseases; (4) the relationship between saturated fat consumption and
increased risk of these diseases; (5) knowledge about dementia, and (6) the distinction
between dementia and normal changes in memory with age.
The interventions were designed to be culturally-tailored to Mexican heritage as
knowledge of nutrition and different dietary behaviors such as food procurement,
planning, and preparation are heavily influenced by culture and ethnic traditions
(Chyun, Amend, Newlin, Langerman, & Melkus, 2003). To address knowledge and skill
sets that Latinas of Mexican descent indicated as highly sought elements (Ayala et al.,
2001), the culturally-tailored intervention targeted three areas of dietary fat behaviors
that have been shown to promote the increase of healthy fat consumption (e.g.,
unsaturated fats) and the decrease of less healthy fat consumption (e.g., saturated and
trans fats), in combination with an educational component on fats. These three areas
were: (1) food procurement (e.g., shopping on a budget, taking small steps, using the
Latin American Food Pyramid, smart shopping, reading food labels); (2) food
39

preparation (e.g., cooking with healthy fats, fat avoidance, fat substitution, cooking
healthy and tasty meals); and (3) food consumption (e.g., eating a balanced diet,
continuing to eat healthy when eating out or during holidays). The substitution of
unsaturated fats for saturated and trans fats is a key dietary modification strategy with
strong evidence for its effectiveness in coronary heart disease prevention (for a full
review, see Hu & Willett, 2002). Moreover, Hu and Willett (2002) also indicated that the
adoption of a combination of dietary behaviors bestowed greater benefits than a single
one. To convey this material, the intervention incorporated previously established
nutrition education techniques found to be culturally-relevant such as cooking
demonstrations, fotonovelas, experience sharing (Ayala & Elder, 2001; Ayala et al.,
2001; Elder et al., 2006) and game show formats. The intervention adhered to simple,
easy to understand lessons, in keeping with established strategies to communicate
accurate and culturally-tailored nutrition information (American Dietetic Association,
2006). Based on pilot testing feedback, vivid photographs and other visual aids were
prominently featured to circumvent potential concerns of low reading literacy in the
targeted study population.
In addition to developing a culturally-tailored intervention, substantial attention
was given in ensuring that the manualized intervention was appropriately translated. To
that extent, several individuals were hired for translation services. Qualified bilingual
and bicultural translators translated the recruitment, informed consent, and interview
materials, and the study intervention. Two bilingual and bicultural advanced research
assistants of Mexican descent additionally translated and back translated and edited all
study materials and the entire intervention to ensure not only linguistic, but also cultural
validity for a primarily Mexican American audience.
40

41

B.5. Intervention procedure

The BHA curriculum consisted of two workshops. A total of 69 workshops were
conducted between the two intervention conditions; i.e., 37 workshops (‘workshop I’)
covered the first part and 32 workshops (‘workshop II’) covered the second part (see
Table 1). There was not an identical number of workshop I and workshop II classes
because changes to participants’ schedules necessitated combining classes. Of the 69
workshops taught, 51 workshops were held in Spanish and 18 in English. Each
workshop took 2 hours to complete, for a total of 4 hours. Minus time allotted for breaks
and coverage of logistical details, the actual total instructional time was 220 minutes.
Participants receiving the intervention were greeted by study staff in the waiting
room of the LALES clinic, and led to the kitchen in the back of the clinic where the
workshops were held. Participants took seats around the kitchen table and were asked
to sign in. For the first workshop, the table was set up so that each participant had in
front of them: a folder containing workshop materials for that day, a nametag, a bottle
of water, and a healthy fruit snack; participants in the second workshop received the
same materials minus the snack. Following the manualized script, study staff
introduced themselves, and then facilitated introductions with the participants.
Logistical considerations such as child care services, location of restrooms, etc. were
covered prior to the start of each workshop.
All materials and handouts for each condition displayed a study logo that either
conveyed the heart only or heart plus brain connection (see Appendix C). All materials,
activities, and instructions in the scripted manual communicated identical information
except for the heart plus brain condition which received an additional component about
42

the association between cardiovascular risk and brain health (see Table 1). The
didactics shared between the two intervention conditions covered objectives such as
learning about fats; the fat-disease connection; goal setting for dietary fat changes;
becoming an informed consumer; and preparing and consuming foods in healthier
ways. The research team spent approximately 20 to 30 minutes teaching the “Brain
Connection” curriculum during the first workshop. Workshops without brain health
content devoted extended time to the other topics to maintain an equivalent duration of
time in comparison to the heart plus brain condition. At the end of the first workshop,
participants were led in an in-class exercise in which they developed three clear and
tangible goals for themselves to target prior to the second workshop. At the start of the
second workshop, participants were led in a discussion to share progression of and
barriers to these goals. At the conclusion of each workshop, participants were
requested to anonymously indicate their participant satisfaction, and study staff was
required to complete self-reports of their adherence to and competence in the
workshop they had conducted. At the conclusion of the second workshop, participants
were administered the posttest interview.
Several types of quality assurances were taken to maximize the likelihood that
the entire research team was following intervention procedures as intended. First, of
the 14 research assistants, 10 served as workshop facilitators. That is, there was no
“firewall” between staff that served as research assistants and facilitators; moreover,
the same facilitators taught workshops across study conditions. Availability was the key
factor in determining which facilitator was scheduled for a certain time slot. However,
experimental bias was reduced because workshop facilitators, though not blind to the
study, were invested in helping women improve their diets, regardless of the study
43

condition. Notably, workshop facilitators delivered a largely scripted intervention, and
fidelity checks were built into the study to assure that facilitators followed the script.
Experimental bias was further reduced because: the investigator ran only one
workshop (out of 69 workshops); the workshop facilitators were tasked with only
executing the intervention, not in altering the intervention in any form; and the
workshops were regularly conducted with co-facilitators. Exception to co-facilitation
occurred only due to scheduling conflicts, at which time only the most experienced
facilitator was sent solo into the field. Also, materials (e.g., facilitator manuals, class
handout packets, posters, etc) were clearly labeled and individually assembled prior to
each workshop so as to prevent cross-contamination between the intervention
conditions.
Second, the investigator conducted extensive ongoing research staff trainings
and held regular refresher trainings to ensure that all staff members were well-
prepared to implement the study, as well as oversaw weekly supervisory meetings. In
addition to addressing questions and concerns, the weekly meetings served to
approximate an actual workshop, with research assistants taking turns to be the weekly
meeting facilitator. In this role, they led the rest of the team in study-related didactics
and assisted each team member craft manageable weekly healthy nutrition goals.
Third, workshop facilitators were required to become highly familiarized with the
intervention manual before being gradually folded into the study team as a regular
workshop facilitator in the community. This was accomplished in a stepwise fashion, in
which facilitators were required to: learn the material; practice delivering the
intervention via a mock ”workshop" with the research study team; observe fellow
facilitators conduct workshops in the field; and pair themselves with a seasoned
44

facilitator when they conducted their first few workshops. The investigator conducted all
the trainings, participated in several workshop observations, and was in regular contact
with each study team member promptly to answer any training questions and/or
concerns. Additionally, the investigator regularly met with the workshop facilitators
(either as individuals or pairs) following each workshop to address any workshop
questions or concerns. Therefore, with all of these quality assurances in place, all
evidence suggests that the intervention was delivered as intended.

45

C. METHODOLOGY

The BHA study was a single-site randomized controlled trial conducted
between February 2008 and March 2009 (see Appendix D). The intervention and
virtually all of the assessments were conducted at the LALES Clinic in La Puente,
California, a city located 20 miles east of downtown Los Angeles.
Permission was given to recruit women who were already participating in
LALES (see Appendix E). Therefore, the base population consisted of those 3,722
females already participating in LALES I and II, a seven-year, NIH-funded ongoing two-
phase epidemiological study examining 6,357 Latinos’ eye care history and eye
disease prevalence (Varma et al., 2004). Individuals were eligible for participation in
LALES if they self-identified as Latino or of Latino heritage; were age 40 or older; and
resided in one of the selected La Puente census tracts. Additionally, the LALES
population was already a part of a dementia investigation called the Study of Vascular
Risk and Cognitive Status in a Latino Population, which examined the coincident
impact of atherosclerosis and vascular risk factors such as diabetes mellitus and
hypertension on cognitive health. Between the eye and cognition studies, there was an
abundance of data in support of cardiovascular risk in this population.
The Institutional Review Board (IRB) of the University of Southern California,
University Park Campus approved this study prior to participant recruitment.
Participants were administered IRB-approved materials in Spanish or English.

46

C.1. Design

A two factor mixed effects, prospective, single-blind, randomized controlled trial
design with one between-subjects variable (condition) and one within-subjects variable
(time) was used for this study (see Table 2). The between-factor variable included four
conditions: heart plus brain; heart only; a pretest-posttest waitlist; and a posttest only
waitlist. The intervention consisted of two workshops, each lasting two hours. Data
collection occurred at three time points: pretest, posttest, and follow-up. The first wait
list condition was assessed at pretest, posttest and follow-up, and the second was
assessed only at posttest to determine whether participating in the baseline
assessment itself had an effect. For the intervention conditions, the posttest was
conducted at the end of the second workshop, and follow-up took place one month
after posttest. This modified Solomon four-group design was selected to evaluate the
effect of pretesting on study outcomes (Kazdin, 2003; Solomon, 1949).
The study was designed to have a time lapse of one week between pretest and
the first workshop. The two workshops were designed to be spaced apart by a week.
Approximately half of the intervention participants were randomly assigned to receive a
motivational boost consisting of a reminder postcard and a phone call midway between
their posttest completion and their one-month follow-up assessment (see Appendix F).
The intervention’s short-term effectiveness was assessed when the intervention was
completed (posttest). The intervention’s moderate-term effectiveness was evaluated
one month after posttest (follow-up). A follow-up period of one month was chosen
based on the mobility of the population that might result in high attrition, and a potential
lack of resources to collect data past this timeline. At the end of the study, participants
47

in the wait list conditions were offered an invitation to participate in two two-hour
workshops based on materials given to participants in the heart plus brain health
condition (see Appendix G).

48
49

C.2. Participants

C.2.a. Estimated power and sample size

A power estimation (Cohen, 1987) conducted using G*Power (Faul, Erdfelder,
Lang, & Buchner, 2007) produced the sample size projections shown in Table 3. To
achieve a power of 0.95, the investigator determined that it was necessary to obtain a
sample size estimate of 81. This sample size estimate was based on a 3 X 3 repeated
measures analyses of variance (ANOVA) with 3 conditions and 3 times of
measurement that was based on equal cell sizes, an expected alpha level of 0.05, and
a medium effect size of 0.20. An additional 27 individuals were added to the estimate
to account for the posttest only condition, to total 108 individuals.
The size of the recruitment pool was derived from an estimated response rate
of 20%-25%, and estimated attrition rate of 15%-20%. These estimates take into
account the LALES I and II recruitment experiences (M. Torres, personal
communication, March 15, 2007). The resulting projections, shown in Table 3, led to
setting the recruitment pool at 540 individuals.

C.2.b. Recruitment from LALES base study

The vast majority of participants were recruited from the LALES base
population between February 2008 and August 2008. The base population consisted of
3,722 females from which the investigator excluded 715 females if the LALES team
reported that the LALES participant was: deceased; medically or mentally incapable
50

(this was defined as a LALES participant living in a convalescent home or were
determined too ill to leave home; M. Torres, personal communication, February 8,
2008); not Latino; missing contact information, being actively traced in the field, or did
not want to be contacted again for research participation; never available; or if the
LALES team could not locate a LALES participant and/or if a participant had moved.
Three thousand and seven females remained after the exclusions. Because the ethnic
characteristics of the community facilitated recruitment of a substantial number of
Latinas of Mexican descent, Mexican origin was not an enrollment requirement.
The investigator randomly selected 540 females (see Table 3) from the 3,007.
Each of these females was mailed a recruitment letter (see Appendix H). During the
mailing, the investigator learned that two of these females were in fact the same
individual; thus, this decreased the recruitment pool to 539 individuals. Forty one letters
were returned as undeliverable mail; however, research staff attempted to reach all
individuals by telephone, regardless of whether the letter had been returned. As in the
LALES, research staff attempted a maximum of five times to contact each individual
before labeling the person as “unreachable” (Varma et al., 2004). A study team
member telephone screened all interested individuals (see Appendix I).

C.2.c. Recruitment from La Puente community

During the course of recruitment, seven individuals from La Puente approached
research staff and expressed interest to participate in the study. Although none of
these seven women were in the LALES study, they met LALES study criteria (i.e., self-
identified as Latino or of Latino heritage; were age 40 or older; and resided in one of
51

the selected La Puente census tracts) and met screening criteria for the present study.
Therefore, it was presumed that these women were similar to the women in the LALES
base study as they lived in the same community and matched the demographic profile
of the larger sample.

52

53

C.2.d. Yoked participants

Pilot testing revealed that some Latinas may be intimidated to participate in a
research study on their own, and a recommendation was made that individuals be
permitted to invite a female friend or relative if they wanted. To be culturally-tailored
and contextually sensitive to the study population, the investigator permitted the pairing
in the following manner, and subsequently tested whether it mattered. Participants who
knew another participant were not individually randomized. Instead, a yoked
randomization method ensured that these pairs or clusters were assigned to the same
condition. A total of 14 participants were yoked (see Figure 1), four of whom were in
the posttest only condition, and 10 of whom yielded data across three time points. Of
the 93 individuals from the LALES who consented to participate in the present study,
three of them – one of whom was in the posttest only condition – were yoked with three
other females from the community (participating in BHA but not in LALES) and six of
them – two of whom were in the posttest only condition – were yoked with one another.
Additionally, of the seven individuals recruited from the community, two of them were
yoked with one another.

54

C.3. Procedure

The frontline staff consisted of 14 bilingual and/or bicultural undergraduate
research assistants. The study team pilot tested all study materials; helped to recruit,
screen, schedule, and consent participants; and conducted study interviews. In this
study, the term “research assistant” refers to a study team member involved in all
aspects of the study procedure except teaching the intervention curriculum. The term
“facilitator” is exclusively reserved for research assistants – who in addition to the

55

responsibilities of a research assistant –also taught the intervention curriculum. They
are called facilitators when describing their teaching of the curriculum, and they are
called research assistants when describing their other activities. Additionally, two
LALES staff members liaisoned with participants during the recruitment, screening, and
informed consent process to help us build trust and credibility with the study
population.
All potentially eligible participants, including the seven volunteers from the
community-at-large, were invited by telephone to meet with a research assistant at the
clinic to review the consent process, to enroll in the study, and to be randomly
assigned to a study condition (see Appendices J and K). All workshops, informed
consent meetings, and interviews were conducted at the LALES clinic in East Los
Angeles. The clinic was adjacent to the La Puente Community Center, a location
chosen because of its identity as a central community meeting place (M. Torres,
personal communication, February 20, 2007). A research assistant scheduled a
mutually convenient time for each informed consent and/or interview meeting,
reminded the individual of the clinic location, provided the study telephone number and
requested that the individual call the study office should a scheduling conflict arise. All
individuals were provided a telephone reminder one day prior to each meeting; in the
event that the scheduled meeting fell on a Monday, the reminder call was made on the
Friday preceding the interview.
Individuals were met by a research assistant in the waiting room of the clinic,
and led back to either the kitchen or to one of four ocular examination rooms. Research
assistants met with individuals one-on-one. After introductions, standard informed
consent procedures were followed. Subsequent to the obtainment of informed consent,
56

participants who did not know another participant in the present study were randomly
assigned to one of the four study conditions after they consented to participate. At this
time, the participant was presented with a numbered, sealed envelope. Each envelope
contained a study condition that had been previously prepared. The participant opened
the envelope, and was allocated to the study condition concealed in the envelope. The
order of study conditions concealed inside the envelopes was generated in an unequal
block randomization manner to oversample by 2:1 (see section D.1.). Standard double
blind procedures for random assignment (not intervention) were enacted in which other
than the investigator, all research staff and all study participants were blinded to the
order of the study conditions enclosed in the sealed envelopes; due to the nature of the
intervention, participants and research staff were not blinded to group membership. All
envelopes were prepared and sequentially numbered (by the investigator) so that the
research staff was required to disseminate the envelopes in sequence so as not to
tamper with the allocation process. Several concealment strategies were enacted to
guard against any potential for selection bias (Kunz, Vist, & Oxman; Torgerson, 1999).
First, the study condition contained in each envelope was printed in a small font on a
standard sheet of paper and folded several times to guard against premature detection
of condition allocation. Second, research staff was required to return the opened
envelope and study condition slip to the investigator after consenting every participant.
Third, multiple research assistants randomly assigned participants, thus minimizing the
possibility of any single team member correctly guessing the block pattern. And fourth,
the investigator neither revealed the blocking mechanism to research staff nor was
directly involved in recruitment efforts, thus further minimizing compromises to the
random assignment process (Kunz, Vist, & Oxman; Torgerson, 1999).
57

Participants in the posttest only condition were thanked for their time and
notified that they would be contacted within one week to schedule their interview.
Participants in either of the intervention conditions or in the pretest-posttest condition
were administered the pretest interview (or posttest or follow-up interviews at
subsequent time points). It was at this time that participants were read the instructions
carefully prior to oral administration of the interview. These instructions told the
participants the general content of the questions that they would be asked, the
expected length of time for the interview, and that participation was voluntary and
confidential. Participants were notified that the pretest interview lasted approximately
45 minutes, and that the posttest and follow-up interviews lasted approximately 30
minutes. Participants were asked if they had any questions, and were told that they
could refuse participation in the interview at any time without it affecting their
participation in the study. Participants were also told that an appreciation gift would be
given to them at the conclusion of the interview. Prior to the administration of each
questionnaire in the interview, participants were read scripted instructions specific to
each questionnaire. A packet of response option cards was provided to participants to
aid in responding to items with a lengthy set of response choices.
At the conclusion of each interview, participants were asked to confirm and/or
update their contact information; thanked for their time and participation; informed that
the research team would be in contact with them to schedule any remaining workshops
and/or interviews; and offered remuneration in the following manner. At the conclusion
of the pretest interview, participants were given a ten dollar gift card to a popular chain
store, as recommended by focus group participants and LALES study staff. In addition,
participants were provided stepped payments at two time points for retention purposes:
58

(1) two movie tickets following completion of the posttest interview and (2) a $25 gift
card to the aforementioned chain store after completion of the one-month follow-up
interview. LALES researchers had identified these payments as those that worked best
with the LALES study population (M. Torres, personal communication, March 15,
2007). Research assistants completed a participant observation form at the conclusion
of the interview. Posttests were routinely conducted at the conclusion of the second
workshop for participants in both of the intervention conditions. Participants in the
pretest-posttest condition were contacted by telephone to schedule their posttest
interview two weeks following their pretest. Participants receiving a follow-up interview
were contacted by telephone to schedule their interview four weeks after the posttest.
Participants in either of the intervention conditions who were randomly assigned to
receive a motivational boost were contacted by telephone two weeks following the
posttest so that a scripted motivational call could be conducted.

C.4. Measures

The interview administered to participants is provided in Table 4. All questions were
read to participants because of potential literacy concerns. The interview contained:
three outcome measures; four moderator measures; three covariate measures; and
several descriptive and demographic items, with the measures sometimes containing
multiple separate variables. The interview consisted mainly of established and widely
accepted measures (see Appendix L).
Of the eight treatment integrity measures, six were administered to study
participants, while workshop facilitators completed the two remaining (i.e., facilitator
59

adherence and facilitator competence). In addition, research assistants documented
their observations after the conclusion of each interview, including their observations of
participant’s comprehension, cooperation, gross mental capacity, and whether the
participant was literate. None of these were found to be problematic.

60
61

C.4.a. Dietary fat outcomes

There were three dietary fat outcomes: (1) dietary fat knowledge; (2) dietary fat
intentions; and (3) self-reported dietary fat behaviors, including four subscales:
avoiding fats; substituting fats; modifying fats; and replacing fats.

Dietary fat knowledge. Dietary fat knowledge was measured with 15 fat-
related items from the U.S. Department of Agriculture’s Diet and Health Knowledge
Survey (U.S. Department of Agriculture, 1994-1996). Respondents were given at least
two alternatives to choose from, plus “I don’t know” and that food items contained “the
same” amount of fat. Each question was scored correct or incorrect. Do not know and
missing responses were also scored as incorrect. The summary score was calculated
as the total number of correct items. Based on pilot testing, two items were modified to
ensure cultural specificity to the Latino diet: (1) “regular hamburger” was described to
be identical to “ground round” (both were described as the term carne molida);
therefore, pilot testers suggested that “ground round” (the original, and less common of
the two terms) be replaced with “ground sirloin”; and (2) because most pilot testers
were not aware of either the food terms “round steak” or “porterhouse steak”, the latter
being described by one pilot tester as ‘too expensive…we only get it for parties”, both
food items were replaced with two foods (i.e., Cup of Noodles and spaghetti noodles)
with widely unequal fat contents which pilot testers reported as being frequently
consumed in the targeted demographic.
Ten items with an item-to-total correlation of 0.10 or greater were retained;
specifically, five items about fat knowledge (i.e., skim/whole milk; regular
62

hamburger/ground sirloin; loin pork chops/pork spare ribs; hot dogs/ham) and five
knowledge items about fat and cholesterol labeling. Total scores ranged from one to
10. Kuder-Richardson 20 (KR-20) estimates assessed the measure’s internal
consistency. The KR-20 estimate for the 10 items was 0.43. The reliability was
comparable to a published reliability coefficient for similar and/or equivalent items from
a previous version of the USDA’s Diet and Health Knowledge Survey (0.46; Obayashi,
Bianchi, & Song, 2003). Dietary fat knowledge was assessed at pretest, posttest, and
follow-up.

Dietary fat intentions. The seven-item Stage of Change for a Low-Fat Diet
measure (Kristal, Glanz, Curry, & Patterson, 1999) was used to create a dietary fat
intention score to assess intentions towards dietary fat behavior change. An interval
score was based on an algorithm (see Table 5) developed by A. Kristal, the lead author
of the measure (personal communication with A. Kristal, February 22, 2010). The
algorithm produced a continuous dietary fat intentions score that ranged from one
(precontemplation) to five (maintenance) (Greene et al., 1999). A secondary method
was to treat the responses as ordinal data, so as to assess whether participants
changed in their stage of change (i.e., from pretest to posttest and posttest to follow-
up). Dietary fat intentions were assessed at pretest, posttest, and follow-up.

Self-reported dietary fat behaviors. Self-reported dietary fat behaviors were
measured with the Fat-Related Diet Habits Questionnaire (Kristal, Shattuck, & Henry,
1990; Shannon, Kristal, Curry, & Beresford, 1997) as modified by Neuhouser,
Thompson, Coronado, and Solomon (2004) for use with individuals of Mexican
63

descent. One of the benefits of this measure was that it captured culturally-specific
dietary fat behaviors pertaining to Latinos of Mexican descent living in the Western
U.S. The questionnaire consisted of 12 items about self-reported dietary fat behaviors
over the previous two weeks related to food procurement, preparation, and
consumption.

64

65

The measure assessed four specific self-reported dietary behaviors: avoiding fried
foods and/or fat as a flavoring; substituting specially manufactured low-fat foods;
modifying meats to be lower in fat; and replacing high-fat foods with fruits and
vegetables. Modifications of the measure were made based on pilot testing findings
(e.g., specifying the use of corn oil and lard for the ‘other fats’ category). Responses to
all questions were on a four-point scale (rarely/never, sometimes, often, usually). Items
ranged from one to four such that a higher score corresponded to a greater frequency
of lower fat intake. The score for each behavior was calculated as the mean of the non-
missing items if a participant responded to at least 50 percent of the items. Scale
developers expected a large number of missing items due to the fact that not all
participants eat all specified foods (Kristal, Shattuck, & Henry, 1990), and as such, the
50 percent rubric followed this guidance. The internal consistency of the overall
modified measure in the present study (Cronbach’s alpha = 0.62) was identical to the
internal consistency found in the original Fat-Related Diet Habits Questionnaire, which
had a Cronbach’s alpha of 0.62 and a test-retest reliability of 0.87 (Kristal, Shattuck, &
Henry, 1990; Shannon, Kristal, Curry, & Beresford, 1997). Self-reported dietary fat
behaviors were assessed at pretest, posttest, and follow-up.

C.4.b. Moderators

Susceptibility. Three measures of susceptibility were used: (1) perceived
susceptibility to vascular disease and AD; (2) self vascular risk; and (3) family vascular
risk.

66

Two perceived susceptibility measures were pilot tested: (1) a visual analogue
form and (2) a Likert scale form. Feedback received indicated that the visual analogue
form was difficult to comprehend compared to the alternate form, which was ultimately
chosen. The selected measure consisted of five items originally developed for the
MetLife Foundation’s Alzheimer’s Survey to assess perceived susceptibility to five
diseases (i.e., cancer, AD, heart disease, stroke, and diabetes) (MetLife Foundation,
2006). Each item was on a four-point scale that ranged from 1 = “strongly disagree” to
4 = “strongly agree”. For this study, cancer served the purpose of a distractor item
because it was not a focus of the workshops. Perceived susceptibility to vascular
disease and AD was assessed at baseline.
Self vascular risk was measured at baseline by a series of ten items that
assessed whether participants had ever been told by a physician that they were at risk
for or had vascular risk factors or diseases (i.e., diabetes; hypertension;
hypercholesterolemia; stroke; angina; myocardial infarction; heart failure or enlarged
heart; obesity; eye disease; and AD). These items were identical to questions used to
assess health status in LALES I and II. Each item was scored dichotomously as either
the presence or absence of a disease or risk factor. Do not know responses were
included with “no” responses. The score was a sum of items checked as present. Items
marked absent, do not know, and missing were scored as zero.
Family vascular risk was measured at baseline by a series of nine items that
assessed whether participants’ immediate family members had ever been told by a
physician that they were at risk for or had vascular risk factors or diseases (i.e.,
diabetes; hypertension; stroke; angina; myocardial infarction; heart failure or enlarged
heart; obesity; eye disease; and AD); hypercholesterolemia was inadvertently omitted
67

from the questionnaire. These items were identical to questions used to assess family
health history in LALES I and II. Each item was scored dichotomously as either the
presence or absence of a disease or risk factor. Do not know responses were included
with “no” responses. The score was a sum of items checked as present. Items marked
absent, do not know, and missing were scored as zero.

Familism. Familism was measured at baseline by the Pan-Hispanic Familism
Scale, a five-item familism measure that demonstrated factorial invariance across
languages (Spanish and English) and across three countries of origin (Mexico, U.S.,
Latin America) (Villarreal, Blozis, & Widaman, 2005). Responses to the five items were
on a five-point scale, ranging from 1 = “strongly disagree” to 5 = “strongly agree” and
the score was calculated as the mean of the non-missing items if a participant
responded to at least 80 percent of the items. At baseline, the internal consistency of
this measure was good (Cronbach’s alpha = 0.88).

C.4.c. Covariates

General acculturation. The abbreviated eight-item version of the 20-item
Acculturation Rating Scale for Mexican Americans-II (Cuellar, Arnold, & Maldonado,
1995) – developed for the 1982-1984 Hispanic Health and Nutrition Examination
Survey – was used at baseline to assess general acculturation (Cuellar, Harris, &
Jasso, 1980; Delgado, Johnson, Roy, & Treviño, 1990; Espino & Maldonado, 1990).
These items were selected in part because of their previous use in LALES I and II. The
eight questions asked participants about their: spoken language; preferred language;
68

spoken and written language ability; self-identification; parents’ ethnic identification;
and self and parents’ birth place (scored as one item). The first seven items were
scored on a scale of one to five, where one represented the strongest Spanish
language/Mexican orientation and five represented the strongest English
language/Anglo orientation. The eighth item ranged in score from a value of three (self
and both parents were born in Mexico) to a value of six (self and both parents were
born in the U.S.). The score was calculated as the mean of the non-missing items if a
participant responded to at least 80 percent of the eight items; since one item was on a
slightly different scale, hypothetically, the lowest score possible was 1.2 and the
highest score possible was 5.1. Higher scores indicated more English/U.S. orientation.
At baseline, the internal consistency of this measure was good (Cronbach’s alpha =
0.93).

Dietary acculturation. Dietary acculturation was measured at baseline with a
modified version of two orthogonal multidimensional dietary acculturation scales
developed to assess dietary acculturation among Chinese Americans (Satia-Abouta,
Patterson, Neuhouser, & Elder, 2002). The investigator adapted the items for use in a
Mexican American sample following pilot testing feedback. Although two scales were
developed (i.e., Latino and Western acculturation scales) to parallel the two measures
developed by Satia-Abouta et al. (2001; i.e., Chinese and Western acculturation
scales), item analysis of the Latino items showed it to be unsatisfactory as a scale, and
it was dropped.
The summary score for the Western acculturation scale, which ranged from
zero to nine, was measured as the frequency of participants’ consumption of these
69

food items: doughnuts; ketchup; Gatorade/Koolaid type beverages; hamburgers;
soda/carbonated beverages; hotdogs; pizza; Hot Cheetos; and sweet breakfasts.
There were no missing items. An item related to skim milk consumption was dropped
because of its low item-to-total correlation score. The internal consistency of the nine
items was low to fair (Cronbach’s alpha = 0.61) compared to Satia-Abouta et al.’s
Western acculturation scale which had a Cronbach’s alpha of 0.72 (Satia et al., 2001).
Dropping other items did not significantly increase the measure’s internal consistency.

Number of children under the age of 17 living at home. The number of
children under the age of 17 living at home was assessed at baseline with a single
item. Respondents who reported no children at home were assigned a zero.

Health literacy. Health literacy was measured at baseline by the Newest Vital
Sign (NVS; Weiss et al., 2005). The NVS – in the form of a nutrition label – was used to
assess reading and numeracy skills, as these are often considered abilities needed to
successfully maneuver through the modern era of health care information. Participants
were provided a hard copy of the NVS nutrition label to hold and to which they could
refer, as needed, while the interviewer asked the items aloud. Based on pilot testing,
confusion arose about the ice cream container size to which the questions referenced;
consequently, the study team offered an empty pint-sized ice cream container as a
visual prop to reduce confusion about the metric (i.e., one pint, not one and a half
quarts). The measure consisted of six items scored dichotomously as correct or
incorrect. Do not know were also scored as incorrect. There were no missing items.
The summary score which ranged from zero to six was calculated as the total number
70

of correct items. The NVS was validated on medical patients including patients in a
publicly funded clinic located in a primarily Spanish-speaking area of the southwestern
U.S. The measure demonstrated good internal consistency in the present study (KR-20
= 0.75), and was comparable to a previously published reliability estimate (0.76; Weiss
et al., 2005).

C.4.d. Treatment integrity

Manipulation check. Four manipulation checks (three quantitative and one
qualitative) were utilized as described below. All of the manipulation checks were
administered at posttest.
Participants were asked to answer five items regarding their belief about the
effectiveness of adhering to a low fat diet to prevent a variety of diseases (i.e., heart
disease, diabetes, AD, cancer, and stroke). The five items had been developed for this
study. Items were scored on a three-point scale that ranged from not effectively to very
effectively. Items were scored one to three such that a lower score corresponded to
lesser belief about the efficacy of a low fat diet in the prevention of chronic disease. For
the first quantitative manipulation check, the item regarding AD was examined to
determine if there were differences between the heart plus brain condition and the
heart only condition. For the second, the items concerning heart disease, diabetes, and
stroke were examined as vascular risk factors to determine if there were differences on
these items across both of the intervention conditions compared to the pretest-posttest
condition. For this study, cancer served the purpose of a distractor item because it was
not a focus of the workshops.
71

For the third quantitative manipulation check, perceived susceptibility to AD
(previously described) at posttest was examined to determine if there was a difference
on this item between the two intervention conditions.
Finally, a qualitative item assessed participants’ strategies to maintain brain
health. The item read “How do you consider maintaining a healthy brain? What do you
do to make sure that your brain stays healthy?”. If respondents attributed brain health
to healthy nutrition, their response was scored a one; otherwise, they were scored a
zero.

Horizontal diffusion check. To assess potential horizontal diffusion of the
intervention’s objectives across conditions (Bronfenbrenner, 1967), participants at
posttest were asked whether: they knew other participants in the study, and if so,
whom; they had conversations with them about the study; and/or they had tried any
ideas suggested by other participants.

Self-reported facilitator adherence. Facilitators were required to complete a
brief treatment fidelity check immediately following the conclusion of every workshop
(see Appendix M). Developed for this study, the self assessment ascertained whether
the staff adhered to the various sections of the intervention (e.g., brain health material
for the heart plus brain health study condition). The items were scored dichotomously.

Self-reported facilitator competence. Facilitators were required to complete a
brief self assessment following the conclusion of every workshop to assess their belief
as to whether that workshop was implemented in a competent manner (see Appendix
72

N). Developed for this study, the single item asked: “How effective do you believe
today’s workshop was in helping participants obtain better eating habits?” The item
was scored one to five such that a lower score corresponded to lesser belief about the
workshop’s effectiveness.

Participant satisfaction. Each participant was requested to complete an
anonymous five-item satisfaction survey at the end of each workshop (see Appendix
O). This survey was developed for this study. Anonymity also included not marking the
surveys for which workshop the participant took, thus preventing the investigator from
comparing conditions or study facilitators. Participants placed their surveys into a
sealed container and research staff retrieved them from the container only at the
conclusion of the study. For ease of understanding, responses to each of the five items
ranged from one to three which corresponded to a three-point faces scale of sad,
neutral, and happy, respectively. The score was calculated as the mean of the non-
missing items if a participant responded to at least four of the items.

C.4.e. Descriptive information

Demographic information collected at baseline included age, completed
educational level, and socioeconomic status. These characteristics were assessed with
three single items comparable to demographic items found in LALES. In addition, other
key sample characteristics were ascertained, including: self-rated health, as measured
by a reliable, self-reported single item measure of physical health (Finch, Hummer,
Reindl, & Vega, 2002); physical activity, as measured by a single global indicator of
73

peer-comparison based physical activity measured on a five-point scale (Sternfeld,
Cauley, Harlow, Liu, & Lee, 2000); and household food shopping and preparation
information.

74

C.5. Data analysis

All analyses were performed using SAS software, version 9.2 (SAS Institute,
2009). A two-tailed test of statistical significance was used in the interpretation of all
analyses. Although a statistical significance in the tables in the Results section was
defined by p-values < 0.05, adjustment of significance consistent with a Bonferroni
correction was considered in making interpretations of the findings; specifically, the p-
value of 0.05 was divided by the number of outcomes (three) to produce a Bonferroni
correction < 0.01 (actual value = 0.0167). Therefore, the main findings were considered
significant if they survived this correction.

C.5.a. Preliminary analyses

Each questionnaire was entered twice in Survey Monkey by two study team
members. Data cleaning occurred in several steps. First, the double-entered data were
compared, and any observed discrepancies and out-of-range values were cleaned.
Second, the frequency and score distributions of individual items were examined.
Additional outliers identified at this stage were checked against the raw data and
appropriately cleaned.
The reliability of multi-item scales was examined with Cronbach’s alphas for
items measured continuously and Kuder-Richardson 20 coefficients for items
measured dichotomously to determine the internal consistency of the study measures
(Nunnally & Bernstein, 1994). Descriptive statistics illustrated baseline characteristics

75

of the sample. Continuous variables were characterized with arithmetic means and
standard deviations, and categorical variables with frequencies.

Test of group equivalence at pretest. Random assignment was utilized to
minimize the probability of various threats to internal validity (e.g., selection bias) so as
to increase the likelihood that differential change across study conditions was due to
the intervention alone (Kazdin, 2003). However, because random assignment may not
have generated equivalent conditions, the investigator tested group equivalence by
comparing conditions after random assignment. If differences were found, the variable
will be entered as a covariate in the test of change on the outcomes.
ANOVAs and chi-square analyses were performed to detect the presence of
any statistically significant differences across study conditions on all demographic and
study items. Fisher exact tests were used to compare categorical variables across
conditions when cell sizes existed with fewer than five participants.
Additional analyses (i.e., ANOVAs and t-tests) investigated whether certain
measurable aspects of the study design, i.e.: (1) being yoked, (2) participants of non-
Mexican descent and (3) participants involved in the study as community volunteers,
significantly differed from their counterparts.

Posttest-only condition. A posttest-only condition was included in the
modified Solomon four-group design so as to assess several threats to internal validity,
i.e., historical and testing effects (Kazdin, 2003), specifically pretest sensitization (A. M.
Nezu & C. M. Nezu, 2008). Student’s t-tests for independent and dependent groups as
well as chi-square analyses were used to compare participants in the posttest-only
76

condition with those in the pretest-posttest condition. Fisher exact tests were used in
cell sizes with fewer than five participants to compare categorical data.

Time lag. Despite best efforts, participants’ time interval between workshops,
as well as between assessment and intervention occasionally varied as a result of
personal and logistical barriers, e.g., illness, unexpected child care demands,
inclement weather, or a lack of transportation. Consequently, the investigator examined
whether the variability in time lags affected findings. Specifically, Pearson correlations
assessed whether the time lag, assessed by the amount of elapsed time between
workshops, and between assessment and intervention (i.e., screening to IRB consent;
IRB consent to pretest; pretest to workshop 1; workshop 1 to workshop 2; posttest to
one month follow-up; posttest to motivational boost, and motivational boost to follow-
up) was correlated with dietary fat outcomes across conditions. Where correlations
were found, lag was entered as a covariate in the tests of change on the outcomes.

Test of selective attrition. Student’s t-tests and chi-square analyses were
conducted to compare participants who were retained with those that attrited. Fisher
exact tests were used in cell sizes with fewer than five participants to compare
categorical data.

C.5.b. Treatment integrity

Manipulation check. Four manipulation checks were considered for analysis.
First, a two-group ANOVA was computed to examine whether participants’ belief of the
77

efficaciousness of a low fat diet to AD at posttest varied between the two intervention
conditions.
Second, a two-group ANOVA was performed to examine whether participants’
belief of the efficaciousness of a low fat diet to vascular disease (i.e., heart disease,
diabetes, and stroke) at posttest varied between the intervention conditions and the
pretest-posttest condition. The score for vascular disease was calculated as the mean
score for these three items.
Third, a two-group ANOVA was performed to assess whether participants’
perceived susceptibility to AD at posttest varied between the two intervention
conditions.
Fourth, findings from the qualitative manipulation check were analyzed via
content analysis. Specifically, thematic analysis was used, in which codes were
created from themes after a careful examination of the responses (Miles & Huberman,
1994). Codes were then cross-checked against the entire data set, and codes that
ascribed brain health to healthy nutrition were clustered together to form a brain health-
nutrition theme. A chi-square analysis of this theme by study condition was conducted.
If the chi-square showed statistical significance, a multiple comparison test for
proportions was conducted to localize the difference between intervention conditions
(Elliott & Reisch; personal communication with A. Elliott, June 3, 2010; Zar, 1998).

Self-reported facilitator adherence. Frequencies were conducted to compare
self-reported facilitator adherence to curriculum modules in both intervention
conditions.

78

Self-reported facilitator competence. A Student’s t-test was performed to
detect the presence of any statistically significant difference between intervention study
conditions in self-reported competence of facilitators.

Participant satisfaction. Only a total mean score for participant satisfaction
was available for calculation because all surveys were anonymous and did not include
study condition or workshop time point (i.e., workshop 1 or workshop 2). This
information is included descriptively.

Motivational boost. A series of ANOVAs were conducted to investigate
whether the motivational boost resulted in a significant difference for the dietary fat
outcomes across intervention conditions.

C.5.c. Primary outcomes

Completer analysis. A repeated measures ANOVA with between-subjects
factors was performed utilizing PROC GLM with repeated measures to test within-
subject (time), between-subject (condition), and within-subject by between-subject
interaction (time by condition) effects. The time factor had three levels: pretest,
posttest, and follow-up. The condition factor had three levels: heart plus brain
condition, heart only condition, and pretest-posttest condition. A significant interaction
would indicate that different conditions changed differently. Planned comparisons were
used to identify which conditions differed from one another. The planned comparisons
corresponded to the hypotheses. They (1) compared the two intervention conditions to
79

the pretest-posttest condition to test whether any intervention was superior to no
intervention; and (2) compared the heart plus brain condition to the heart only
condition, to determine whether the addition of the brain component improved
outcomes over the heart oriented program alone. Separate planned comparisons were
calculated based on a comparison between two measurement timepoints: from pretest
to posttest, and from pretest to follow-up. Computationally, a difference score between
the two timepoints was used as the dependent variable, condition was the independent
variable, and the two planned comparisons were specified as contrasts. In the
completer analyses, only participants with valid data at all time points on the outcome
being analyzed were included in the analyses.
Movement in dietary fat stage of change was examined in a secondary
analysis. Individuals were categorized as improving, not changing, or regressing in
stage of change, first from pretest to posttest, and then from posttest to follow-up. Chi
square analyses were conducted for each of the two time periods comparing the three
conditions on these three categories of change. If the findings were significant, the chi
square was followed by a multiple comparison of proportions test to isolate any
statistical difference between conditions.

Intent-to-treat analysis. A repeated measures ANOVA with between-subjects
factors was performed utilizing PROC MIXED to test the within-subject (time),
between-subject (condition), and within-subject by between-subject interaction (time by
condition). The analysis and planned comparisons paralleled the completers analysis.
The purpose of the intent-to-treat analysis was to impute missing values to allow for the
retention of participants who skipped a measure once or twice and/or attrited from the
80

study. Participants who did not provide a response to an item at any timepoint were
excluded from the analyses. The PROC MIXED models were analyzed with a
compound symmetry matrix so that it matched the covariance structure assumed in
PROC GLM (Howell, 2008). The only difference between the PROC GLM completer
analysis and the PROC MIXED intent-to-treat analysis was the inclusion of participants
with partial data.
.
C.5.d. Intervening variable analyses

Tests for moderators. Tests for moderators addressed possible constructs
that may have influenced any observed group differences in change between the two
intervention conditions and pretest-posttest conditions.
The distribution of each moderator was examined. If the distribution of the
moderator was not normal, a median split was used for the moderator, whereby values
of -1 and +1 indicated scores below and above the median, respectively. If the
distribution of the moderator was normal, the moderator was centered through
converting into a z-score prior to testing for interaction as recommended by Aiken and
West (1991). Z-scores were calculated by subtracting the mean score from each score
and dividing the difference by the standard deviation.
Centered contrast codes were created to identify the intervention conditions
(+1) versus the pretest-posttest condition (-1). To test for moderation, a linear
regression equation was run in PROC GLM for each moderator and each of the
outcomes, first pretest to posttest and then pretest to follow-up. Each regression
equation contained a dependent variable consisting of the outcome’s difference score
81

(from pretest to posttest, or from pretest to follow-up, respectively); and the
independent variables, consisting of: the centered moderator term, the centered term
for condition, and an interaction of those two terms. A moderator was thought to be
significant if the condition by moderator interaction term explained a statistically
significant amount of variance for change on the dietary fat outcome (Baron & Kenny,
1986; Chmura Kraemer, Wilson, Fairburn, & Agras, 2002).

Covariates. Four covariates, i.e., general acculturation, dietary acculturation,
number of children under the age of 17 living at home, and health literacy, were
examined to determine whether they would predict more or less dietary fat behavior
change across all conditions.
The distribution of each covariate was examined. If the distribution of scaled
covariates was not normal (i.e., general and dietary acculturation, and health literacy),
the variable underwent a median split whereby values of -1 and +1 indicated scores
below and above the median, respectively. If the distribution of the discrete covariate
(i.e., number of young children at home) was not normal, the potential covariate was
dichotomized with values of 0 and +1. If the distribution of the covariate was normal,
the potential covariate was entered as is into the model.
To test covariates, a linear regression equation was run in PROC GLM for each
of the covariates and each of the outcomes, first pretest to posttest and then pretest to
follow-up. Each regression equation contained a dependent variable consisting of the
outcome’s difference score (from pretest to posttest, or from pretest to follow-up,
respectively); and two independent variables, condition and covariate, with the latter
entered second to denote its covariate status. Only participants with valid data at all
82

time points were included in the analyses. A covariate was considered present if there
was a statistically significant effect. In those instances, the investigator looked at the
main effect for condition in comparison with the main effect for condition in the absence
of the covariate.

83

C.5.e. Effect size

Effect sizes for differential change over time by group were based on the
omega-squared term ( ω
2
) and calculated using the EFFECTSIZE function in SAS (SAS
Institute, 2009). The ω
2
assessed the strength of any effects found in the planned
comparisons in the completers analysis (Cohen, 1987).
84

D. RESULTS

D.1. Achieved sample size and actual power

Table 6 depicts the flow of participants from screening to follow-up in a
Consolidated Standards for Reporting of Trials (CONSORT) table (Begg et al., 1996),
showing who: was contacted for screening; met criteria for the study; and who was
randomized to one of the four arms of the study.
Of the 539 females, 420 were excluded for the following reasons: no working
telephone number (n = 170); unreachable (n = 98); not interested (n = 38); scheduling
conflicts (n = 31); medical illness (n = 16); out of town (n = 14); participating in other
nutrition classes (n = 13); no transportation (n = 13); moved (n = 9); refused to
talk/hung up (n = 7); passed away (n = 6); family illness (n = 3); death in family (n = 1);
and not of Latina origin (n = 1). An additional five individuals who were: on a special
diet (e.g., either by personal choice or for medical reasons); not over age 40; and not
expecting to be in the Los Angeles for the study duration were excluded. This
recruitment process resulted in 114 participants. None of the remaining exclusion
criteria (i.e., individuals who were pregnant, could be pregnant, or were actively trying
to become pregnant during the study duration, or self-reported a diagnosis of an eating
disorder) was reported in the recruitment sample. With the addition of the 7 volunteers,
a total of 121 participants were invited to an individual meeting where informed consent
was administered and participants were randomized to condition. Twenty one potential
participants were lost prior to randomization due to: no longer interested (n = 8);
unreachable (n = 6); scheduling conflicts (n=5); family illness (n = 1); and death in
85

family (n = 1). One hundred females met criteria, were consented, and took part in the
study. The yield from the LALES pool was 17%. When recruitment proved to be
difficult, the investigator became concerned about attrition from the intervention
conditions and consequently elected to double the size of the intervention conditions
relative to the control conditions. On this basis, 32 participants were in the heart plus
brain condition; 33 were in the heart only condition; 17 were in the pretest-posttest
condition; and 18 were in the posttest only condition.
Achieved power was estimated taking the minimum cell size and assuming
equal cell sizes. As shown in Table 7, with a minimum cell size of n = 17, this
assumption corresponded to a total n = 51, suggesting a conservative lower bound on
power = 0.79 for an effect size = .20. However, the actual power was almost certainly
greater than 0.80, as the sample size for the three conditions with repeated
measurements was actually larger (n = 82), and the extra 31 participants (82 – 51 =
31) helped to increase power, although not as much as they would have had the cell
sizes been equal.

D.2. Baseline characteristics

D.2.a. Sample characteristics

Demographic and study characteristics at baseline of the study population by
study condition are presented in Table 8.
86

Table 6
Participant Ascertainment from Screening to Completion of One-Month Follow-up
Assessment The number of completed cases may vary according to the end point
because of missing data.

87

88

89
Table 8, continued

90
Table 8, continued

91

D.2.b. Test of group equivalence at pretest

Participants were not statistically different across conditions at baseline
according to being: yoked or not; of Mexican descent or not; or a community volunteer
or not. Additionally, participants did not statistically significantly differ across conditions
at baseline on the vast majority of the measures of study outcomes or moderators, or
on demographic characteristics; the sole exception was the number of family members
living at home. That is, the posttest only condition and the pretest-posttest condition at
posttest statistically differed in their responses to the number of family members that
lived in their home over the past one year (p = 0.01, Fisher’s exact test); i.e., the
posttest only condition had more 4 or 5 person households whereas the pretest-
posttest condition at posttest had more very large households of 6 or more.
Completers in the two intervention conditions plus the pretest-posttest condition
were not significantly different at pretest on any of the three dependent variables (see
Table 9). Notably, some of the items or scales had substantial missing data, reducing
the power to detect differences across conditions (see Table 10). The interview format
– which required all items to be read to participants – motivated a low rate of
missingness on all items except for dietary fat behavior items. In that instance,
missingness for self-reported dietary fat behavior items was driven by the skip pattern
where not engaging in a specific dietary fat behavior resulted in the item being scored
as missing.

92

93

94

Table 10, continued
95

D.2.c. Posttest only condition

Three group comparisons (see Table 11) were assessed to ascertain two key threats
to internal validity, i.e., historical and pretest sensitization effects: (1) pretest-posttest
condition at pretest versus posttest-only condition and (2) pretest-posttest condition at
posttest versus posttest-only condition. A third analysis compared the pretest-posttest
condition at pretest versus the same condition at posttest. First, comparisons between
the pretest-posttest condition at pretest and the posttest only condition showed no
statistically significant differences on scaled scores with the exception of dietary fat
knowledge; i.e., the posttest only condition had a significantly higher mean score for
dietary fat knowledge (mean = 5.39, SD = 1.29) compared to the pretest-posttest
condition at pretest (mean = 4.00, SD = 1.73), t (33) = 2.70, p = 0.01. Second,
comparisons between the pretest-posttest condition at posttest and the posttest only
condition revealed no statistically significant differences on scaled scores. Thus, there
was no significant evidence of test sensitization. Third, comparisons between the
pretest-posttest condition at pretest and at posttest revealed the absence of statistically
significant differences on all scaled scores and all demographic characteristics.

D.2.d. Time lag

Analyses revealed that the duration between study time points (i.e., screening
to IRB consent; IRB consent to pretest; pretest to workshop 1; posttest to one month
follow-up; posttest to motivational boost, and motivational boost to follow-up) did not
statistically significantly correlate with study outcomes. However, the time lag from
96

workshop 1 to workshop 2 was significantly negatively correlated with dietary fat
knowledge at follow-up (r = -0.25, n = 72, p = 0.04), but not at posttest (r = -0.02, n =
74, p = 0.86). Consequently, the time lag from workshop 1 to workshop 2 was entered
as a covariate in the model testing the main study outcome for dietary fat knowledge.

D.2.e. Test of selective attrition

A total of eight participants attrited from the study (see Table 12), a 10% rate of
attrition for those not in the posttest only condition where there was only one time of
measurement. Three participants were in the heart plus brain condition; four were in
the heart only condition; and one participant was in the pretest-posttest condition. None
of the attrited participants were yoked with other study participants. Attrition occurred at
various points in the study timeline: after pretest but prior to the intervention (n = 3); to
midway through the intervention (n = 2); to immediately prior to one-month follow-up (n
= 3). There was very little attrition; tests comparing completers to attriters showed a
minimal number of significant differences. Chi-square analyses and t-tests were used
to compare completers and attrited individuals on the scaled scores and demographic
characteristics at baseline. No significant differences were observed on any

demographic variables. On the main outcomes, self-reported total fat behaviors and fat
modification behaviors were significantly higher among completers than attriters, t (75)
= 2.84, p = 0.0058 and t (75) = 2.23, p = 0.03, respectively. Those who attrited
reported less healthy fat behaviors at baseline (see Table 13).

97

98

99

100

D.3. Treatment integrity

D.3.a. Manipulation check

Four manipulation checks were examined. Table 14 displays the mean scores
for individual items in the quantitative manipulation check for completers by study
condition. One test asked whether participants in the heart plus brain condition had
greater awareness at posttest of the connection between diet and brain health
compared to participants in the heart only condition who were not exposed to this
information. A second test asked whether participants in the intervention conditions
had greater awareness at posttest of the connection between diet and vascular health
(i.e., heart disease, diabetes, and stroke) compared to the pretest-posttest condition. A
third test asked whether participants in the heart plus brain condition were more likely
to endorse greater perceived susceptibility to AD at posttest compared to participants
in the heart only condition. A fourth test determined whether participants in the heart
plus brain condition were more likely to mention diet as a factor in brain health
compared to participants in the heart only condition. Tests 1, 3, and 4 thus checked
whether the healthy brain message was received; test 2 checked whether the healthy
heart message was received. First, the two-group ANOVA did not reveal a statistically
significant difference between participants in the heart plus brain condition (mean =
2.63, SD = 0.63) versus the heart only condition (mean = 2.52, SD = 0.58) regarding
their understanding that a low-fat diet can prevent AD, F (1, 52) = 0.46, p = 0.50.
Second, a two-group ANOVA comparing participants in the intervention conditions
(mean = 2.84, SD = 0.28) to those in the pretest-posttest condition (mean = 2.63, SD =
101

0.42) revealed a statistically significant difference regarding their understanding that a
low-fat diet can prevent vascular risk factors, F (1, 72) = 5.80, p = 0.02. Follow up
analyses looking at individual items showed that the significant difference was primarily
driven by their understanding that a low-fat diet can prevent stroke, F (1, 70) = 5.89, p
= 0.02.

102

103

Third, the two-group ANOVA showed that there was a significant difference in
perceived susceptibility to AD at posttest between intervention conditions, F (1, 54) =
5.39, p = 0.02, with the heart plus brain condition (mean = 2.59, SD = 0.80) showing a
significantly higher level of perceived susceptibility to AD than the heart only condition
(mean = 2.07, SD = 0.88).
Fourth, results from the qualitative manipulation check indicated that 68.97% of
participants in the heart plus brain condition ascribed brain health to nutritious dietary
behaviors compared to 44.83% of participants in the heart only condition and 18.75%
of those in the pretest-posttest condition. An omnibus test revealed that this difference
was statistically significant, χ
2
(2, n = 74) = 10.69, p < 0.01. A multiple comparison test
for proportions (Elliott & Reisch; personal communication with A. Elliott, June 3, 2010;
Zar, 1998) revealed that there was a significant difference between the proportions of
the heart plus brain condition (0.69) and the pretest-posttest condition (0.19), but that –
in the comparison of interest – the heart plus brain condition was not significantly
different from the heart condition.
These results suggested that there was mixed evidence for differential
sensitization across study conditions for the importance of brain health reflecting the
differences in content in the two types of nutrition classes. Notably, participants in the
heart plus brain condition were significantly differentially sensitized for perceived
susceptibility to AD at posttest but not for more awareness of the connection between
brain health and nutrition. Also, participants in the intervention conditions were
significantly differentially sensitized for the importance of nutrition for reducing vascular

104

disease risk, in particular, stroke prevention, compared to the pretest-posttest
condition.

D.3.b. Horizontal diffusion

None of the participants reported interaction with participants in study
conditions besides their own.

D.3.c. Self-reported facilitator adherence

Facilitators reported their self-reported facilitator adherence a total of 63 times.
Although the same facilitators conducted classes for different conditions, 100% of the
heart plus brain workshops correctly covered the brain health curriculum, whereas
none of the heart only workshops covered the same curriculum (see Table 15).
Although the investigator had limited information from which to confirm treatment
adherence, she does not have any evidence that cross-contamination occurred in light
of the efforts previously taken to have facilitators strictly adhere to specific intervention
conditions.

D.3.d. Self-reported facilitator competence

Facilitators who conducted workshops for the heart plus brain health condition
reported a self-reported facilitator competence mean score of 4.59 (n = 34). Facilitators
who conducted workshops for the heart only condition indicated a self-reported
105

facilitator competence mean score of 4.48 (n=29). There was no statistically significant
difference in facilitators’s self-reported perceived competence between intervention
conditions, t (61) = -0.78, p = 0.44.
106

Table 15
Percentage of Workshops That Covered Each BHA Curriculum Module
107
108

D.3.e Participant satisfaction

Participants across all workshops and all conditions indicated a high
satisfaction score (mean = 2.86, SD = 0.28).

D.3.f. Motivational boost

Findings from a series of analyses of variance revealed that treatment
outcomes for participants that received a motivational boost were not statistically
significantly different at follow-up from those who did not (see Table 16). Therefore,
whether or not the participant received the motivational boost was not considered
further.

109

110

D.4. Primary outcomes

D.4.a. Completers analysis

Mean scores and standard deviations for completers in the intervention
conditions and the pretest-posttest condition are shown in Table 17. The completer
analysis – shown in Table 18 – is comprised of F-scores from repeated measures
ANOVAs for the effect of time, condition, and the interaction of time by condition for
completers in the intervention conditions and the pretest-posttest condition. These
analyses showed a significant effect for time on dietary fat knowledge, dietary fat
intentions, and self-reported dietary fat behaviors, i.e., total fat behaviors, avoiding fats,
and modifying fats: F (2, 70) = 4.43, p = 0.02; F (2, 70) = 4.05, p = 0.02; F (2, 65) =
11.71, p < 0.0001; F (2, 69) = 8.50, p < 0.01; and F (2, 65) = 5.35, p < 0.01,
respectively. These results suggested that, among participants who continued to
provide data, across those who received an intervention and those who did not,
significant global increases were evidenced in: (1) dietary fat knowledge; (2) dietary fat
intentions; (3) self-reported total fat behaviors; and two of its subscales, namely (4)
self-reported avoiding fats; and (5) self-reported modifying fats.
In support of the third hypothesis, total fat behaviors showed a significant time
by condition interaction, F (4, 132) = 2.91, p = 0.0241. However, this finding was not
significant after Bonferroni adjustment for three outcome measures. Neither of the
other two outcomes showed a significant time by condition interaction.
As indicated earlier, the time lag from workshop 1 to workshop 2 was
significantly negatively correlated with dietary fat knowledge at follow-up. When time
111

lag was entered as a covariate into a repeated measures model for dietary fat
knowledge among completers, there was no significant difference over time, F (2, 136)
= 0.72, p = 0.49, and dietary fat knowledge did not show a significant time by condition
interaction, F (4, 136) = 0.24, p = 0.91.

112

113

114

For dietary fat knowledge, consistent with the lack of a significant time by
condition interaction, examination of the planned comparisons failed to reveal
significant differences in gains in dietary fat knowledge across conditions, either from
pretest to posttest or from pretest to follow-up. For dietary fat intentions, planned
comparisons revealed no differences across intervention conditions from pretest to
posttest or pretest to follow-up.
For self-reported behaviors, planned comparisons contrasting the heart plus
brain condition and the heart only condition with the pretest-posttest condition indicated
that participants in the intervention conditions were significantly more likely to self-
report change in behavior between pretest and follow-up, F (1, 66) = 10.09, p = 0.002,
although not between pretest and posttest. Looking at the four component subscales,
there was a significant improvement on fat avoidance behaviors at follow-up compared
to pretest, F (1, 70) = 7.58, p < 0.01, but not at posttest compared to pretest. Based on
ω
2
estimates, the effect sizes for study outcomes with significant contrasts were: 0.12
(self-reported total fat behaviors) and 0.08 (self-reported avoiding fat behaviors). None
of the other three subscales showed significant change. Planned comparisons between
heart plus brain and heart only were non-significant. To summarize, participation in
either of the intervention conditions resulted in self-reporting more improvement in
dietary fat altering behaviors, especially more fat avoidance, between pretest and
follow-up, compared to the pretest-posttest condition; however, the two intervention
conditions did not differ in their effects.
In addition to capturing dietary fat intentions as a continuous measurement as
described above, dietary fat intentions were also assessed as a discrete stage score.
Findings indicated that there were no significant differences between the three study
115

conditions from pretest to posttest (see Table 19), χ
2
(4, n = 74) = 1.35, p = 0.85, or
from posttest to follow-up (see Table 20), χ
2
(4, n = 74) = 1.18, p = 0.88. Thus, the post
hoc multiple comparison test for proportions – conventionally used to localize any
significant difference – was not conducted in light of the null findings.

116

117

D.4.b. Post hoc analyses for completers

Post hoc analyses were motivated by the discovery that participants in the
workshops did not improve significantly more on the dietary fat knowledge measure
than the pretest-posttest condition, leading to a direct test of whether participants were
taking in the information taught during workshops. Consequently, the investigator
identified four of the ten dietary fat knowledge items where the items were explicitly
taught in the workshop curriculum for both intervention conditions. Results of post hoc
analyses indicated that there was not a statistically significant difference between the
combined intervention conditions and the pretest-posttest waitlist condition for any of
these items, i.e., whether skim or whole milk had more saturated fat t (75) = 0.2, p =
0.60; and whether more total fat was in regular hamburger or ground sirloin, t (75) =
0.33, p = 0.74; loin pork chops or pork spare ribs, t (74) = 1.39, p = 0.17; and hot dogs
or ham, t (75) = -0.20, p = 0.84.

118

D.4.c. Intent-to-treat analysis

Imputed mean scores and standard deviations for all participants in the
intervention conditions and the pretest-posttest condition, as obtained from the PROC
MIXED procedure, are shown in Table 21. The intent-to-treat analysis – shown in Table
22 – paralleled the completers analysis. The table shows results from repeated
measures ANOVAs using PROC MIXED for the effect of time, condition, and the
interaction of time by condition for all participants (regardless of study completion).
Similar to the completer analysis, the intent-to-treat analysis showed a significant effect
for time on dietary fat knowledge, dietary fat intentions, and self-reported dietary fat
behaviors, i.e., total fat behaviors, avoiding fats, and modifying fats: F (2, 149) = 6.43, p
< 0.01; F (2, 148) = 5.28, p < 0.01; F (2, 140) = 14.30, p < 0.0001; F (2, 146) = 6.49, p
< 0.01; and F (2, 139) = 6.72, p < 0.01, respectively. These results indicate significant
global increases in: (1) dietary knowledge; (2) dietary fat intentions; (3) self-reported
total fat behaviors; (4) self-reported avoiding fat behaviors; and (5) self-reported
modifying fat behaviors.
In addition, the main effect for condition on dietary fat intentions was statistically
significant, F (2, 77.7) = 4.26, p = 0.02, with the heart plus brain condition scoring
highest and the pretest-posttest condition scoring lowest, but with no differential
change over time.
In support of hypothesis 3, total fat behaviors showed a significant time by
condition interaction among all participants, F (4, 141) = 3.03, p = 0.0196. However,
this finding was not significant after Bonferroni adjustment for three outcome
measures.
119

120

121

Examination of the planned comparisons in the intent to treat analysis across
study conditions failed to reveal significant differences in change in dietary fat
knowledge across conditions; pretest to posttest or pretest to follow-up.
Planned comparisons for dietary fat intentions indicated no differences across
conditions in change from pretest to posttest or from pretest to follow-up. Examination
of the planned comparisons for total fat behaviors showed a significantly greater
change for participants in the intervention conditions compared to the pretest-posttest
condition when comparing pretest to follow-up, F (1, 140) = 8.78, p = 0.004, but not when
comparing pretest to posttest. None of the four component subscales showed a significant
interaction or planned comparison. The planned comparisons between intervention
conditions were non-significant. The mean scores are shown in Figures 2-6 for all
measures with significant main effects for time or significant interaction effects.
122

Figure 2. Dietary fat knowledge mean scores for all participants across three time
points.

4
4.5
5
5.5
6
Pretest Posttest Follow-up
Time
Score
Heart plus brain
Heart only
Pretest-posttest

Figure 3. Dietary fat intentions mean scores for all participants across three time
points.

3.7
3.8
3.9
4
4.1
4.2
4.3
4.4
Pretest Posttest Follow-up
Time
Score
Heart plus brain
Heart only
Pretest-posttest

123

Figure 4. Self-reported total fat behavior mean scores for all participants across three
time points.

2.9
3
3.1
3.2
3.3
3.4
3.5
Pretest Posttest Follow-up
Time
Score
Heart plus brain
Heart only
Pretest-posttest

Figure 5. Self-reported avoiding fat behavior mean scores for all participants across
three time points.

3
3.1
3.2
3.3
3.4
3.5
3.6
3.7
Pretest Posttest Follow-up
Time
Score
Heart plus brain
Heart only
Pretest-posttest

124

Figure 6. Self-reported modifying fat behavior mean scores for all participants across
three time points.

3.3
3.4
3.5
3.6
3.7
3.8
Pretest Posttest Follow-up
Time
Score
Heart plus brain
Heart only
Pretest-posttest

125

D.5. Intervening analyses

Next, two types of intervening variables were considered. One, moderators, or
variables that might alter the strength of the relationship between study condition (i.e.,
intervention conditions and the pretest-posttest condition) and dietary fat outcomes
were examined. Two, covariates, or variables that might affect change on the
outcomes regardless of study condition, were evaluated (Hayes, Laurenceau, &
Cardaciotto, 2008).

D.5.a. Tests for moderators

As shown in Table 23, two potential moderators were considered: (1)
susceptibility and (2) familism. Susceptibility was operationalized in four ways:
perceived susceptibility to vascular disease; perceived susceptibility to AD; self
vascular risk; and family vascular risk. Familism was represented by one variable.

126

As seen in Table 23, a test for interaction effects did not show that perceived
susceptibility to vascular disease or AD, family vascular risk, or familism significantly
moderated the association between conditions (intervention versus control) on any of
the significant dietary fat outcomes. Similarly, a test for interaction effects did not show
that the amount of self vascular risk significantly moderated the association between
conditions (intervention versus control) and any of the significant dietary fat outcomes
with one exception, namely dietary fat knowledge from pretest to follow-up, F (1, 52) =
4.32, p = 0.04. In partial support of the eighth hypothesis, Figure 7 shows that
individuals in the intervention conditions who possessed more self vascular risk at
pretest demonstrated greater improvements in dietary fat knowledge from pretest to
follow-up compared to those that had less self vascular risk.

127

Figure 7. Dietary fat knowledge mean scores for completers from pretest to follow-up
by self vascular risk. A positive difference score represents improved dietary fat
knowledge.

-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
Low High
Self vascular risk
Dietary fat knowledge difference score
from pretest to follow-up
Intervention conditions
Pretest-posttest condition

128

D.5.b. Covariates

Four covariates were considered, i.e., general acculturation; dietary
acculturation; presence of young children living at home; and health literacy. As shown
in Table 24, the presence of young children living at home significantly accounted for a
proportion of the variance for one dietary fat subscale across study conditions,
although in the opposite direction from what was predicted in hypothesis 13. That is,
the presence of young children living at home was associated with greater
improvements in self-reported fat avoidance behaviors from pretest to posttest, F (1,
69) = 5.35, p = 0.02 (see Figure 8) and from pretest to follow-up, F (1, 69) = 4.43, p =
0.04 (see Figure 9) compared to absence of young children living at home. The
presence of young children living at home did not significantly account for any of the
variance for self-reported total fat behaviors either from pretest to posttest, F (1, 65) =
2.43, p = 0.12 or from pretest to follow-up, F (1, 65) = 0.79, p = 0.38, or for any of the
other self-reported fat behavior subscales.

129

130

Figure 8. Self-reported dietary fat avoidance behavior mean scores for completers from
pretest to posttest by young children living at home. A positive difference score
represents improved self-reported dietary fat avoidance behaviors.

0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
None One or more
Young children living at home
Self-reported fat avoidance difference score
from pretest to posttest

Figure 9. Self-reported dietary fat avoidance behavior mean scores for completers from
pretest to follow-up by young children living at home. A positive difference score
represents improved self-reported dietary fat avoidance behaviors.

0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0.55
None One or more
Young children living at home
Self-reported fat avoidance difference score
from pretest to follow-up

131

E. DISCUSSION

E.1. Findings

E.1.a. Overview

This is the first known study to test the effectiveness of a heart plus brain
health-focused nutrition intervention compared to a heart health-focused nutrition
intervention. Despite the extensive research that suggests Latinos are at increased risk
for vascular disease and consequently heart and brain disease (Aronow & Ahn, 2001;
Fitten, Ortiz, & Pontón, 2001; Ford & Giles, 2003; Goff, Nichaman, & Chan, 1997;
Haan et al., 2003; Muntner, Garrett, Klag, & Coresh, 2002; Plassman et al., 2007;
Stroup-Benham, Markides, Espino, & Goodwin, 1999; Sundquist & Winkleby, 1999),
the diet and health randomized controlled trials for Latinas known to this investigator
did not mention the relationship between diet and brain health.
This study hypothesized that participants in the heart plus brain health condition
would have greater gains in dietary fat outcomes than the heart only condition, and that
those in either intervention condition would show greater gains than a pretest-posttest
waiting list control condition. Findings from this study revealed no significant
differences in outcomes between the heart plus brain and the heart only intervention
conditions, with neither condition demonstrating superiority to the other. Surprisingly,
there was only modest evidence of a differential effect for participating in an
intervention compared to being on a wait list. Those who participated in either
intervention condition showed greater improvement than the pretest-posttest control

132

condition on self-reported behaviors related to fats, especially to self-reported
avoidance of fat, when comparing pretest to one-month follow-up. The effect size for
self-reported total fat and fat avoidance behaviors was 0.12 and 0.08, respectively.
However, there were no significant differences on self-reported behaviors related to
fats between pretest and posttest, and there were no significant differences on either
dietary fat knowledge or dietary fat intentions between those who participated in either
condition and the pretest-posttest control group. Essentially the same results were
found in an intent-to-treat analysis, indicating that selective attrition did not explain the
findings.
Two intervening variables were significantly related to change on the key
outcomes. As hypothesized, one’s own vascular risk was a significant moderator.
Individuals in the intervention conditions who reported a higher level of vascular risk
had greater gains in dietary fat knowledge from pretest to follow-up than those with
lower self vascular risk. Unexpectedly, for all women in the study, the presence of
young children living at home was related to significantly greater improvement in self-
reported fat avoidance behaviors from pretest to posttest and pretest to follow-up
compared to individuals not residing with young children.
Participants in the heart plus brain condition had significantly greater perceived
susceptibility to AD at posttest compared to those in the heart only condition,
suggesting that the two interventions differentially affected perceived susceptibility,
albeit modestly. However, at posttest participants exposed to the heart plus brain
condition did not evidence greater awareness of the connection between diet and AD
than did participants in the heart only condition. Thus, the results should not yet be

133

taken as arguing conclusively against the augmentation of standard health education
materials currently in practice with a brain health focus. Rather the question remains
open. There may need to be a stronger presentation of the brain material, or a simpler
test of whether the brain health message was received.

E.1.b. Dietary fat knowledge

The present study hypothesized that the highest gains in dietary fat knowledge
would be among participants in the intervention conditions, especially those in the
heart plus brain health condition. Findings revealed that no significant differences in
increases of dietary knowledge across conditions on either the total dietary fat
knowledge scale or a subset of items that corresponded to material that had been
emphasized in the intervention. However, individuals in the intervention conditions who
had greater self vascular risk showed significantly larger gains in dietary fat knowledge
from pretest to follow-up, compared to those with less vascular risk, underlining the
connection between susceptibility and health behavior change (Chow et al., 2010;
Silagy, Muir, Coulter, Thorogood, & Roe, 1993).
The SNAP study, a randomized controlled trial of a culturally-tailored,
cardiovascular-based dietary fat intervention aimed at Latinos of Mexican descent, is
one of a few comparison studies available in the literature. In contrast to this study’s
findings, SNAP found a significant difference on dietary fat knowledge between the
intervention condition (culturally tailored and dietary fat reduction-focused) and the
control condition (general nutrition classes (non-culturally tailored and without a dietary

134

fat reduction emphasis) (Howard-Pitney, Winkleby, Albright, Bruce, & Fortmann, 1997).
SNAP also had several methodological characteristics absent in the present study that
may have contributed to its statistically significant finding, including: a large sample
size (n = 351) compared to a smaller sample size in the present study (n = 74
completers across three timepoints) and a protracted intervention consisting of weekly
90-minute sessions held over six weeks compared to two two-hour sessions. Also
advantaging the culturally tailored intervention was the fact that intervention and control
conditions were not matched with respect to emphasis on behavior change, unlike the
effort in the present study completely to match the two intervention conditions on all
aspects of the intervention with the single exception of the brain health content.

E.1.c. Dietary fat intentions

This study hypothesized that the heart plus brain condition would lead to a
greater intention to change compared to the heart only and pretest-posttest conditions.
On the contrary, dietary fat intentions as measured by continuous data did not show
significant time by condition interactions. Furthermore, data from this study showed a
ceiling effect for dietary fat intentions to suggest that the level at which an individual
was situated at baseline may have been near proximity to levels of dietary fat
intentions that reflect initiation or continuation of behavior change. These findings merit
explanation. It is possible that the mere decision to opt for enrollment in a nutrition
research study was premised on a sufficient level of intention that precipitated
participation (Greene et al., 1999). This notion would be consistent with the high scores

135

at pretest. In turn, ceiling effects limited the degree of change that could be attained,
which might have made it more difficult to detect differential change, i.e., a time by
condition effect. Also, it is not inconceivable to consider that some participants across
study conditions may have been interested and/or already engaging in dietary behavior
changes prior to study enrollment. The fact that dropouts were significantly lower than
completers on baseline self-report dietary fat behaviors is also consistent with the
suggestion that women chose to participate because they were already committed to
trying to change their behaviors.

E.1.d. Dietary fat behaviors

Both intervention conditions exhibited significant gains in self-reported total fat
behaviors and specifically in self-reported fat avoidance behaviors from baseline to
follow-up compared to the control condition. Contextualized, these findings replicated a
similar pattern found in the SNAP study. In SNAP, both conditions were administered a
dietary intervention and had greater self-reported dietary fat reductions two months
after baseline than at baseline; that is, as in the present study, neither intervention in
SNAP significantly proved better than the other towards greater self-reported total
dietary fat reduction (Howard-Pitney, Winkleby, Albright, Bruce, & Fortmann, 1997). It
is challenging to directly contrast findings with additional studies, as research designs
used in other studies vary. For example, Mujeres Felices por ser Saludables, a
randomized controlled trial of a combined dietary and breast health intervention for
Latinas found that self-reported dietary fat intake was significantly lower at eight month

136

follow-up compared to a control group that received mailings on general health
information (e.g., dental care, seat belt safety) (Fitzgibbon, Gapstur, & Knight, 2004).
Similarly, Secretos de la Buena Vida, a randomized controlled trial of a promotora-led
dietary communication intervention for Latinas found that participants in the promotora-
led condition versus those in two control conditions (i.e., tailored print materials, and
standard print materials) exhibited significantly lower levels of self-reported total fat and
saturated fat consumption two weeks prior to the conclusion of the intervention,
although these findings diminished at one-year follow-up (Elder et al., 2006). Despite
the range of study designs, it seems accurate to conclude that culturally-tailored dietary
fat interventions for Latinas can alter self-reported dietary fat behavior.
Examination of the present finding from a theoretical perspective informs the
discussion in a few additional ways. One, the BHA intervention was informed by risk
communication as well as health behavior theories, namely HBM and SCT, to: reduce
barriers; increase self-efficacy; and develop skills for behavioral change. However, the
assessment tools used to measure change were heavily weighted towards evaluating
outcomes (i.e., dietary fat knowledge, intentions, and self-reported dietary fat behavior
changes) rather than evaluating processes of change (e.g., barriers, self-efficacy, and
the process by which skills develop). Thus, the results reflect the measures used and
may have missed nutrition class content in both intervention conditions that stressed
the identification, promotion, and demonstration of recommended behavioral changes.
Two, there was a marked difference in perceived susceptibility for AD at
posttest between the intervention conditions (i.e., the heart plus brain condition had a
significantly higher perceived susceptibility to AD versus the heart only condition).

137

However, this did not translate into further gains for the heart plus brain condition. One
can draw upon this finding further within the context of risk communication literature.
This literature suggests that the assembly of a comprehensive cognitive schema is a
gradual, step-wise process, whereby behaviors represent the endpoint in an
individual’s mental model that is preceded by several constructs, including from left to
right: cognitions, and the interaction between cognitions and behaviors (Kahlor,
Dunwoody, Griffin, Neuwirth, & Giese, 2003). Lundgren and McMakin (2009) and
Morgan et al. (2002) espouse that mental models provide a heuristic for this step-wise
process in identifying a road map for an individual’s thoughts about disease prevention
and health promotion efforts, composed of: (1) an individual’s perceived vulnerability to
a disease; (2) cognitive behavioral strategies to help the individual reduce their
perceived vulnerability to a disease; and, (3) approaches to effectively build self-
efficacy and skills so as to enable the individual towards successful health behavioral
change. The third manipulation check showed that individuals in the heart plus brain
condition had significantly higher levels of perceived susceptibility to AD at posttest
compared to the heart only condition, thus providing support for the presence of the
first factor. From a risk communication perpective, perceived susceptibility to a disease
is one of the first steps in an individual’s decisionmaking heuristic, and consequently is
positioned the farthest downstream from behavior change (Kahlor, Dunwoody, Griffin,
Neuwirth, & Giese, 2003; Lundgren & McMakin, 2009; Morgan, Fischhoff, Bostrom, &
Atman, 2002). The heart plus brain condition appears to have succeeded with respect
to this first step.

138

As previously detailed, both intervention conditions offered several cognitive
behavioral approaches to help the individual modify vascular risk by way of altering
dietary fat behavior. These correspond to the second factor, i.e., strategies. Notably,
there was no difference in the strategies offered to the two conditions, and there was
no significant difference for self-reported total fat behaviors between the two
intervention conditions. Quite possibly the mental models of participants in the heart
plus brain condition may not (yet) be on dietary fat modifications related to brain health.
Moreover, the heart plus brain condition did not significantly differ from the heart only
condition on their belief that a low fat diet would prevent AD. Research on mental
models suggests that this finding closely aligns with the challenging cognitive
interactions that occur in the cognitive space between an individual’s perceived
susceptibility to AD and the behavior (i.e., dietary fat modification) that one can
potentially enact to reduce their perceived risk.
The third factor in this model of change – self-efficacy – was not directly
assessed in this study, although a risk communication angle lends insight. That is, a
risk communication perspective is consistent with the significant gains in self-reported
total fat behavior exhibited by both intervention conditions (Lundgren & McMakin, 2009;
Morgan, Fischhoff, Bostrom, & Atman, 2002). Furthermore, these findings suggest that
a mental model for brain health promotion may have begun to form, but may need to
be supported by specific strategies and a stronger connection of perceived vulnerability
to strategy.
The other three specific self-reported dietary fat outcomes did not show
significant group by time effects. Three considerations bear mention. As suggested

139

above, nomothetic measures such as self-reported dietary fat intake measures
designed to capture dietary outcomes and normed on large samples may not be as
optimal for detecting incremental gains – central to complex dietary behaviors – as can
be had with idiographic assessments that capture an individual’s progress (Fernández-
Ballesteros & Botella, 2008). Second, substituting, modifying, or replacing a saturated
fat may have been seen as “advanced” behaviors in comparison to a “beginner” step
like avoiding a saturated fat; that is, avoiding a food item may fall to the left of the
other, more advanced steps. Moreover, the spectrum also may portray a continuum of
required effort; that is, self-reporting substituting, replacing, and/or modifying fats may
involve more “active” behaviors whereas avoiding may primarily entail “passive”
exclusion of certain foods or techniques by which to prepare foods (Kristal, Shattuck, &
Henry, 1990). The curriculum put forth in both of the intervention conditions
emphasized a successive approximational approach in building self-efficacy that is in
line with well-established behavioral change principles (Bandura, 1986, 1997).
Participants may have considered avoiding fats a manageable first step on the
continuum. Third, the challenge inherent in capturing dietary change for a specific
nutrient (e.g., fat, fiber, antioxidants) has been well established in the literature
(Kumanyika et al., 2000). Fourth, avoiding fats actually encapsulates several lower-
order and potentially complementary behaviors such as recognition of foods in which
fats are present and interpretation of food labels (Kumanyika et al., 2000); in risk
communication terms, avoiding fats is a simplified behavioral strategy for an underlying
heuristic that is representative of a collective set of dietary fat behaviors involving fat
substitution, modification, and replacement. Interpreted through the lenses of Bandura

140

(1986; 1997), Fernández-Ballesteros and Botella (2008), and Kumanyika (2000), the
significant finding of self-reported fat avoidance is a promising first step.

E.1.e. Brain health focus

The idea that educating women about the importance of nutrition for reducing
dementia risk would boost behavior change received little support. However,
participants in the heart plus brain condition were significantly differentially sensitized
for perceived susceptibility to AD. Additionally, participants in both intervention
conditions were significantly differentially sensitized for the importance of nutrition
towards vascular disease, in particular, stroke prevention, compared to the pretest-
posttest condition. The other manipulation checks failed to provide evidence to suggest
that the heart plus brain condition and the heart only condition significantly differed on
recognition of the connection between nutrition and brain health. Although surprising,
these findings are nonetheless of interest as discussed next. In particular, in the
absence of a clearer and more consistent manipulation check, and lack of significant
differences between heart only and heart plus brain interventions, they could reflect
inadequacies in delivering the brain content.
Still, the fact that a heart plus brain health intervention did not show any
significant differences compared with a standard heart health curriculum is of potential
relevance for health education and health communication purposes. The recent rise in
brain health promotion campaigns – including the development of brain health
initiatives such as the Alzheimer’s Association’s “Maintain Your Brain” campaign and

141

the AARP’s “Staying Sharp” campaign (Alzheimer's Association, 2007; American
Association of Retired Persons, 2007) – are premised on the idea that emphasizing
brain health will motivate behavior change. Placed into the context of health behavior
theory, the HBM suggests that perceived susceptibility is paramount to the adoption of
preventive behaviors. This framework contributed to the development of the
hypotheses in the present study to predict that the heart plus brain condition would
have greater dietary fat outcomes compared to the remaining study conditions. What
was supported in this study is that women in the heart plus brain condition had greater
perceived susceptibility to AD at posttest compared to the heart only condition,
although this did not translate to significantly differential outcomes between
intervention conditions. This result could be applied to designing more effective brain
health campaign content. The fact that the manipulation checks showed modest
evidence that the women heard different messages does make it clear that a brain
health message was apparently harder to give to the audience than anticipated.
Placed into the context of applied public health campaigns, the national
dialogue on AD and lifestyle modification is still early. Although the diet and AD
message may be timely and relevant to certain groups, including Caucasians who are
widely aware of dementia (Connell, Scott Roberts, McLaughlin, & Akinleye, 2009;
Roberts et al., 2003), the message may be ahead of its time and appearing seemingly
irrelevant to some, as findings from health disparities research indicate that Latinos are
still at reduced awareness for dementia compared to their Caucasian counterparts
(Ayalon & Arean, 2004).

142

Moreover, the public health campaigns also face another challenge: they have
not yet attuned themselves to a variety of culturally-driven perspectives on health and
disease including the mental models currently in existence among those at high risk for
cardiovascular and cerebrovascular disease, including Latinos. Research on culture
and health communication unequivocally demonstrate the importance of cultural
sensitivity and cultural integration into message development, ranging from the
integration of surface characteristics such as linguistic validity to deeper characteristics
such as cultural values, health beliefs, and culturally-driven behaviors (for a full review,
see Kreuter & McClure, 2004). Two examples of the interplay between culture and
brain health are discussed next.
One issue is whether dementia is considered a sufficiently high level priority
among the Latinas sampled – especially when it is placed on a wider spectrum of other
potentially more salient health and non-health issues? Lack of support for hypothesized
differences between heart and heart plus brain conditions may reflect assignment of a
lower priority level for dietary modification to reduce risk for dementia – information that
may be helpful in the future but is not as pertinent as issues relegated to the here-and-
now. Anecdotally, participants in the present study led lives filled with several roles and
wrought with daily economic, domestic, and caregiving stressors. Thus, their not
translating heightened perceived susceptibility to dementia into behavior change may
be a function of the participants’ attending to matters most salient to them at the
present time.
Second, the brain health curriculum solely emphasized the biomedical model of
brain health and disease. Yet, the biomedical model may or may not accurately match

143

every participant’s perceptions of brain disease. A recently published qualitative study
among economically disadvantaged Latinos of Mexican descent living in South Texas
revealed a prevailing belief in curanderismo to help explain the underlying etiological
explanation for dementia (Sharkey, Sharf, & St. John, 2009). Curanderismo, a health
belief system based on a trifecta borne from harmony with nature, spirit, and self, is
considered to be a belief system that lies in parallel with biomedical health
perspectives (Trotter & Chavira, 1997). Etiologically, Latinos who believed in
curanderismo consider dementia to be the result of punishment for a sin for which
amelioration could be sought with biomedical treatments (Sharkey, Sharf, & St. John,
2009). Thus, the augmentation of the biomedical model with alternative etiological
beliefs about dementia may further increase cultural validity and prove effective in the
expansion of health communication and health behavior intervention strategies
(Anderson, Day, Beard, Reed, & Wu, 2009).

E.1.f. Latinas as nutrition gatekeepers in their families

This study identified Latinas to be “nutrition gatekeepers” and recognized them
as important agents to help foster nutritional changes in themselves and their families
(Wansink, 2006). The study team and the investigator anecdotally learned from the
participants that their forays to implement gradual dietary changes in their homes were
met with a wide range of reactions, including encouragement, support, gratitude, as
well as defiance to try new foods, nostalgia (‘we always made it that way’), to
preference for alternative foods. In light of this range of responses, it was not surprising

144

to witness some participants return to the second workshop and comment on their
experiences at experimentation in small steps – and their resulting discouragement to
effect substantial change. Wansink (2006) addresses this as a seemingly common
experience among nutrition gatekeepers. That is, Wansink (2006) suggests that it is
not uncommon for gatekeepers to underappreciate their own influence over their
family’s food decisions. Coined the “72% solution”, Wansink (2006), in turn, suggests
that nutrition educators promote a moderate level of influence (i.e., 72%) among
gatekeepers and in doing so, allow gatekeepers to unburden the responsibility of
shouldering all food decisions made for and by their family. However, the presence of
family members was not always a barrier to dietary fat change, despite anectodal
evidence to the contrary. Cross-sectional data from the NHANES III indicated evidence
for a specific social contextual factor in dietary fat change, i.e., that the inhabitation of
children younger than 17 years of age was associated with significantly higher total fat
and saturated fat consumption (Laroche, Hofer, & Davis, 2007). But, findings from the
present study interestingly showed a reverse pattern of results; that is, individuals
residing with young children under the age of 17 showed significantly greater
improvements in self-reported dietary fat avoidance behaviors from pretest to posttest
and pretest to follow-up versus individuals not living with young children.
Although our study primarily focused on Latinas as individuals, their role as
caregivers certainly needs to be acknowledged for many reasons – not the least of
which is rapport-building. Mosca et al. (2004) found that female caregivers were
associated with an increased risk of cardiovascular disease that may be possibly

145

explained by poor lifestyle habits and psychosocial risk factors (Aggarwal, Liao,
Christian, & Mosca, 2009; Lee, Colditz, Berkman, & Kawachi, 2003). Perhaps the initial
hypothesis that Latinas in the intervention conditions with a high level of familism would
embrace self-care by way of improved dietary fat outcomes was too ambitious;
alternatively, a plethora of research from the caregiving literature suggests that it is
altogether too common for caregivers to decrease or deny themselves self-care in light
of their caregiving responsibilities (for a review, see Aranda & Knight, 1997). In light of
the daily caregiving stressors endured by many of our participants and the challenges
they faced in implementing gradual dietary changes, the investigator surmises that the
absence of a significant finding related to the motivation boost is due to potential
insufficiency of the boost as designed. Future studies are encouraged to test the
efficacy of sustained reinforcement strategies as a form of motivation that has been
shown to be important for dietary fat change among Latinas predominantly of Mexican
descent (Elder et al., 2006).

E.2. Limitations

Some limitations of this study should be noted. First, the study had modest
statistical power in part because of small cell sizes in the waiting list conditions.
However, the overall study sample size across the three repeated time conditions (n =
82) likely helped to increase the final power value beyond the calculated minimum of
0.79. Second, the present study relied exclusively on self-report. Despite this

146

challenge, in the field, procedures were employed to reduce respondent bias, including
staff training to standardize participant interviews. Third, the study was constrained by
the complicated construction of the self-reported dietary fat behaviors measure.
Specifically, the skip patterns within the two-tiered assessment (e.g., “Did you eat
chicken…?” followed by “How often was the chicken fried or cooked with lard or oil?”)
resulted in missing data in instances in which the participant did not engage in the
behavior, did not know, or refused to answer the item. As mentioned earlier, this issue
resulted in a small sample size for some of the fat behavior outcomes (e.g., replacing
fats). Fourth, the length of the intervention may have been insufficient; individuals may
have benefited from a protracted intervention implemented over the span of several
weeks. The length and pacing of the intervention emerged from consultation and pilot
testing. Nonetheless, it remains unanswered whether individuals in this study were
fatigued or burdened with the amount of information presented to them within a
relatively brief timeframe. Fifth, the follow-up period was brief which may have been an
insufficient time period in which to capture change across all study outcomes. Sixth,
the investigator excluded individuals who either lacked transportation or participated in
other nutrition classes, groups that may have had higher vascular risk (e.g.,
homebound, sedentary lifestyle, medical condition) who may or may not have altered
the findings. Seventh, the intervention hypothetically could have addressed even more
social and physical environments that are more likely to influence health behaviors
(Glass & McAtee, 2006; Kumanyika et al., 2000). Still, the research study team’s field
experiences as well as the findings from the present study suggested that these
nutrition gatekeepers were enacting dietary fat change albeit the changes being

147

gradual and ongoing. And eighth, effect size estimates were small (Cohen, 1987) and
were under the 0.20 used to estimate sample size needed for recruitment. Although
there is debate as to how best to accurately capture the magnitude of the effect
(Bakeman, 2005) – especially in community-based research settings that may not be
as sterile as laboratory settings – the small effect sizes, coupled with the modest
number of statistically significant differences to detect intervening variables hints at the
possible involvement of additional constructs which are yet to be accounted.

E.3. Strengths

Methodological strengths of this study also merit attention. First, this
investigation is the first known randomized controlled trial to evaluate the effectiveness
of a heart plus brain-focused nutrition intervention among Latinos. The experimental
design afforded the opportunity to examine the causal relationship between study
condition and dietary fat outcomes. Additionally, random assignment is considered to
increase the likelihood of group equivalence and to make certain threats to internal
validity (i.e., selection by history, and selection by maturation) less plausible (Kazdin,
2003). Second, the study intervention was based on principles of risk communication
and health behavioral theories. Third, there were two wait list conditions. In the first
wait list condition designed to assess testing effects, participants received the same
interviews but not the intervention. In the second wait list condition designed to
evaluate testing sensitization to the baseline measure, participants received the
posttest interview only.

148

Fourth, there was a very low attrition rate (10%). Recruitment and retention
challenges in Latino populations and recommendations to adapt flexible and creative
study designs in light of these challenges have been well-documented in the research
literature (Paskett, 2008; Varma et al., 2004). Moreover, a literature review at the
beginning of this decade revealed a critical shortage of racial and ethnic populations in
dementia clinical research (Olin, Dagerman, Fox, Bowers, & Schneider, 2002), and
identified several strategies to increase participation. Efforts to adopt these
recommendations appear to have succeeded by: conducting the study at a familiar
location to the participants; offering a flexible schedule for participation; involving a
bilingual and/or bicultural research team; and providing free-of-charge child care in
clinic. The recruitment strategies contribute to increased understanding of conducting
dementia-related clinical research in community-based Latino populations.
Fifth, all study materials were developed to be culturally-tailored for Latinos of
Mexican descent. An empirical review of the cultural appropriateness for health
education programs acknowledges the value in accounting for the cultural
characteristics for not only the larger group (e.g., Latinas), but also for specific
individuals within that group (e.g., Latinas of Mexican descent) (Kreuter, Lukwago,
Bucholtz, Clark, & Sanders-Thompson, 2002). Moreover, the culturally nuanced
aspects of diet particularly warranted special cultural consideration (Horowitz, Tuzzio,
Rojas, Monteith, & Sisk, 2004; McMahon, Cathorall, & Romero, 2007). Consequently,
all materials were developed in close partnership with knowledgeable research staff,
community health educators, promotoras de salud and community volunteers to ensure
linguistic and cultural validity (Hornik & Kelly, 2007). Utilization of community-based

149

participatory research is a particularly appropriate methodology for addressing health
disparities of all types (U.S. Department of Health and Human Services, 1998).

E.4. Lessons learned

This study affords an opportunity to highlight strategies that may help improve
similar studies in the future.

Recruitment efforts. First, the investigator recommends augmenting the
recruitment strategies in place with a plan to address logistical concerns that prevented
the participation of some, e.g., lack of transportation. Although this was outside the
scope of our resources, resolution of logistical concerns may increase study
participation and participant completion.

Intervention. Second, the investigator recommends that future studies
evaluate whether variations in the “dose” of the brain health material in the curriculum
makes a difference. The primary reason for not increasing the amount spent on brain
health material was borne out of concerns that devoting more time to brain health
would lead to one of two other problems: Either (a) there would be less time in the
heart plus brain condition to teach the other content, or (b) the heart plus brain classes
would have to be lengthened relative to the heart classes. Further, the nature of the
experimental design led to making brain health a discrete segment rather than
integrated throughout, in order to have exact equivalence of the heart material.

150

However in light of our limited significant findings, further evaluation of the dose
appears warranted. Relatedly, the subtlety of the brain health curriculum merits
examination; that is, was the content too subtle to be absorbed? Manipulation check
results suggest that may have been a problem. Third, the investigator recommends
empirically testing whether study outcomes would improve if the curricula were
dispensed over a larger duration of time versus a concentrated amount (e.g., two two-
hour workshops). The concern arises from post hoc findings suggesting participants
were not attending to some material taught in the interventions. Perhaps the material
was too concentrated and thus if that led to premature information overload or
saturation. Fourth, the “dose” of the motivational boost deserves re-examination to
further understand the intrinsic and extrinsic factors that drive and sustain dietary fat
behavior change. Fifth, the investigator was severely constrained in her attempt to
evaluate adherence to the intervention conditions past a simple adherence check. In
the future, it is recommended that a comprehensive adherence check assess treatment
fidelity.

Measurement. Sixth, the investigator advocates for a more overt manipulation
check than what was used in the present study to help drive a successful induction of
the manipulation check (Kidd, 1976). The choice of the manipulation check chosen for
this study was driven by the concern that numerous mentions of terms like “Alzheimer’s
disease” would prematurely sensitize participants across intervention conditions to the
disease. Seventh, the investigator encourages the supplementation of nomothetic self-
report dietary fat outcome measures with more frequently self-reported idiographic

151

accounts of cognitive and behavioral processes so as to evaluate ongoing and
incremental change as well as to assess the extent of convergent validity between
these data.

E.5. Conclusion

This study shed insight into the development, implementation, and evaluation of
a theoretically-driven, culturally-tailored nutrition intervention for Latinas of Mexican
descent. Significant findings from this study suggest that a culturally-tailored nutrition
intervention offers modest assistance for improving dietary fat outcomes. Ongoing
research at multiple systemic levels (e.g., individual, family, neighborhood, community)
is warranted to further understand the mechanisms by which nutrition interventions can
be initiated and sustained to reduce modifiable vascular risk among Latinos and other
high-risk populations.
152

References

Abraham, C. (2008). Beyond stages of change: Multi-determinant continuum models of
action readiness and menu-based interventions. Applied Psychology, 57, 30-
41.
Aggarwal, B., Liao, M., Christian, A., & Mosca, L. (2009). Influence of caregiving on
lifestyle and psychosocial risk factors among family members of patients
hospitalized with cardiovascular disease. Journal of General Internal Medicine,
24, 93-98.
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting
interaction. Newbury Park: Sage.
Alcalay, R., Alvarado, M., Balcazar, H., Newman, E., & Huerta, E. (1999). Salud para
su corazón: A community-based Latino cardiovascular disease prevention and
outreach model. Journal of Community Health, 24, 359-379.
Alonso, A., Jacobs, D. R., Menotti, A., Nissinen, A., Dontas, A., Kafatos, A., et al.
(2009). Cardiovascular risk factors and dementia mortality: 40 years of follow-
up in the Seven Countries study. Journal of the Neurological Sciences, 280, 79-
83.
Alzheimer's Association. (2005). Alzheimer's Association brain health resolution. Paper
presented at the White House Conference on Aging, Washington, D.C.
Alzheimer's Association. (2007). Brain health. Think about your future. Maintain your
brain today. Retrieved January 18, 2007, from
0Hhttp://alz.org/we_can_help_brain_health_maintain_your_brain.asp
American Association of Retired Persons. (2007). Taking control of brain health.
Retrieved January 18, 2007, from
1Hhttp://www.aarp.org/health/brain/takingcontrol/
American Dietetic Association. (2006). Position of the American Dietetic Association:
Food and nutrition misinformation. Journal of the American Dietetic Association,
106, 601-607.
153

Ammerman, A. S., Lindquist, C. H., Lohr, K. N., & Hersey, J. (2002). The efficacy of
behavioral interventions to modify dietary fat and fruit and vegetable intake: A
review of the evidence. Preventive Medicine, 35, 25-41.
Anderson, L. A., Day, K. L., Beard, R. L., Reed, P. S., & Wu, B. (2009). The public's
perception about cognitive health and Alzheimer's disease among the U.S.
population: A national review. The Gerontologist, S1, S3-S11.
Aranda, M. P., & Knight, B. G. (1997). The influence of ethnicity and culture on the
caregiver stress and coping process: A sociocultural review and analysis.
Gerontologist, 37, 342-354.
Armitage, C. J. (2004). Evidence that implementation intentions reduce dietary fat
intake: A randomized trial. Health Psychology, 23, 319-323.
Aronow, W. S., & Ahn, C. (2001). Risk factors for new coronary events in older
Hispanic men and women. American Journal of Cardiology, 88, 901-903.
Atman, C. J., Bostrom, A., Fischhoff, B., & Morgan, M. G. (1994). Designing risk
communications: Completing and correcting mental models of hazardous
processes, part I. Risk Analysis, 14, 779-788.
Ayala, G. X., & Elder, J. P. (2001). Verbal methods in perceived efficacy work. In S.
Sussman (Ed.), Handbook of program development for health behavior
research and practice. Newbury Park: Sage.
Ayala, G. X., Elder, J. P., Campbell, N. R., Engelberg, M., Olson, S., Moreno, C., et al.
(2001). Nutrition communication for a Latino community: Formative research
foundations. Family and Community Health, 24, 72-87.
Ayalon, L., & Arean, P. (2004). Knowledge of Alzheimer's disease in four ethnic groups
of older adults. International Journal of Geriatric Psychiatry, 19, 51-57.
Bakeman, R. (2005). Recommended effect size statistics for repeated measures
designs. Behavior Research Methods, 3, 379-384.
Bandura, A. (1977a). Self-efficacy: Toward a unifying theory of behavioral change.
Psychological Review, 84, 191-215.
154

Bandura, A. (1977b). Social learning theory. Englewood Cliffs: Prentice Hall.
Bandura, A. (1986). Social foundations of thought and action. A social cognitive theory.
Englewood Cliffs: Prentice Hall.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.
Baranowski, T. (1997). Psychosocial and sociocultural factors that influence nutritional
behaviors and interventions: Cardiovascular disease. In C. Garza, J. Haas, H.
J-P. & D. Pelletier (Eds.), Beyond Nutritional Recommendations: Implementing
Science for Healthier Populations (pp. 163–188). Ithaca: Cornell University.
Baranowski, T., Weber Cullen, K., & Baranowski, J. (1999). Psychosocial correlates of
dietary behavior: Advancing dietary intervention. Annual Review of Nutrition,
19, 17-40.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical
considerations. Journal of Personality & Social Psychology, 51, 1173-1182.
Begg, C., Cho, M., Eastwood, S., Horton, R., Moher, D., Olkin, I., et al. (1996).
Improving the quality of reporting of randomized controlled trials. The
CONSORT statement. JAMA, 276, 637-639.
Berg, J. M., Tymoczko, J. L., & Stryer, L. (2006). Biochemistry (6th ed.). New York:
W.H. Freeman.
Bond, D. S., & Wing, R. R. (2009). Behavioral interventions for the treatment of obesity
in women with or at risk for cardiovascular disease. Current Cardiovascular
Risk Reports, 3, 219-225.
Boulware, L. E., Carson, K. A., Troll, M. U., Powe, N. R., & Cooper, L. A. (2009).
Perceived susceptibility to chronic kidney disease among high-risk patients
seen in primary care practices. Journal of General Internal Medicine, 24, 1123-
1129.
Bowen, D. J., & Beresford, S. A. (2002). Dietary interventions to prevent disease.
Annual Review of Public Health, 23, 255-286.
155

Brayne, C., Gill, C., & Huppert, F. A. (1995). Incidence of clinically diagnosed subtypes
of dementia in an elderly population: Cambridge Project for Later Life. British
Journal of Psychiatry, 1995, 255-262.
Bronfenbrenner, U. (1967). Response to pressure from peers vs. adults among Soviet
and American school children. International Journal of Psychology, 2, 199-207.
Brunner, E., Rees, K., Ward, K., Burke, M., & Thorogood, M. (2007). Dietary advice for
reducing cardiovascular risk. Cochrane Database of Systematic Reviews, Art.
No.: CD002128. DOI: 002110.001002/14651858.CD14002128.pub14651853.
Bureau of the Census. (2000). Statistical abstract of the United States: The national
data book. Washington: U.S. Department of Commerce.
Cabassa, L. J. (2003). Measuring acculturation: Where we are and where we need to
go. Hispanic Journal of Behavioral Sciences, 25, 127-146.
Campbell, C., & Desjardins, E. (1989). A model and research approach for studying the
management of limited food resources by low-income families. Journal of
Nutrition Education, 21, 162–171.
Centers for Disease Control and Prevention and the Alzheimer's Association. (2007).
The Healthy Brain Initiative: A national public health road map to maintaining
cognitive health. Chicago: Alzheimer's Association.
Chavez, N. L. S., Persky, V., Langenberg, P., & Pestano-Binghay, E. (1994). Effect of
length of US residence on food group intake in Mexican and Puerto Rican
women. Journal of Nutrition Education, 26, 79-86.
Chmura Kraemer, H., Wilson, T., Fairburn, C. G., & Agras, W. S. (2002). Mediators and
moderators of treatment effects in randomized clinical trials. Archives of
General Psychiatry, 59, 877-883.
Chow, C. K., Jolly, S., Rao-Melacini, P., Fox, K. A. A., Anand, S., & Yusuf, S. (2010).
Association of diet, exercise, and smoking modification With risk of early
cardiovascular events after acute coronary syndromes. Circulation, 121, 750-
758.
156

Chyun, D. A., Amend, A. M., Newlin, K., Langerman, S., & Melkus, G. D. (2003).
Coronary heart disease prevention and lifestyle interventions: Cultural
influences. Journal of Cardiovascular Nursing, 18, 302-318.
Cohen, J. (1987). Statistical power analysis for the behavioral sciences. Hillsdale, NJ,
England: Lawrence Erlbaum Associates, Inc.
Connell, C. M., Scott Roberts, J., McLaughlin, S. J., & Akinleye, D. (2009). Racial
differences in knowledge and beliefs about Alzheimer disease. Alzheimer
Disease & Associated Disorders, 23, 110-116.
Craik, K. (1943). The nature of explanation. Cambridge: Cambridge University Press.
Cuellar, I., Arnold, B., & Maldonado, R. (1995). Acculturation Rating Scale for Mexican
Americans-II: A revision of the original ARSMA scale. Hispanic Journal of
Behavioral Sciences, 17, 275-304.
Cuellar, I., Harris, L., & Jasso, R. (1980). An acculturation scale for Mexican American
normal and clinical populations. Hispanic Journal of Behavioral Sciences, 2,
199-217.
Delgado, J. L., Johnson, C. L., Roy, I., & Treviño, F. M. (1990). Hispanic Health and
Nutrition Examination Survey: Methodological considerations. American Journal
of Public Health, 80 (Suppl), 6-10.
DiLillo, V., Siegfried, N. J., & West, D. S. (2003). Incorporating motivational
interviewing into behavioral obesity treatment. Cognitive and Behavioral
Practice, 10, 120-130.
Duara, R., Barker, W. W., Lopez-Alberola, R., Loewenstein, D. A., Grau, L. B.,
Gilchrist, D., et al. (1996). Interaction of apolipoprotein E genotype, family
history of dementia, gender, education, ethnicity, and age of onset Neurology,
46, 1575-1579.
Elder, J. P., Ayala, G. X., Campbell, N. R., Arredondo, E. M., Slymen, D. J., Baquero,
B., et al. (2006). Long-term effects of a communication intervention for Spanish-
dominant Latinas. American Journal of Preventive Medicine, 31, 159-166.
157

Elder, J. P., Ayala, G. X., Parra-Medina, D., & Talavera, G. A. (2009). Health
communication in the Latino community: Issues and approaches. Annual
Review of Public Health, 30, 227-251.
Elliott, A. C., & Reisch, J. S. Implementing a multiple comparison test for proportions in
a 2xc crosstabulation in SAS [Electronic Version], SAS SUGI Paper 204-31.
Retrieved June 3, 2010 from 2Hhttp://www2.sas.com/proceedings/sugi31/204-
31.pdf.
Espino, D. V., & Maldonado, D. (1990). Hypertension and acculturation in elderly
Mexican Americans: Results from 1982-84 Hispanic HANES. Journal of
Gerontology: Medical Sciences, 45, 209-213.
Evans, D. A., Funkenstein, H. H., Albert, M. S., Scherr, P. A., Cook, N. R., Chown, M.
J., et al. (1989). Prevalence of Alzheimer's disease in a community population
of older persons: Higher than previously reported. JAMA, 262, 2551-2556.
Farrer, L. A., Cupples, L. A., Haines, J. L., Hyman, B., Kukull, W. A., Mayeux, R., et al.
(1997). Effects of age, sex, and ethnicity on the association between
apolipoprotein E genotype and Alzheimer disease. JAMA, 278, 1349-1356.
Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible
statistical power analysis program for the social, behavioral, and biomedical
sciences. Behavior Research Methods, 39, 175-191.
Fernández-Ballesteros, R., & Botella, J. (2008). Self-report measures. In A. M. Nezu &
C. M. Nezu (Eds.), Evidence-based outcome research: A practical guide to
conducting randomized controlled trials for psychosocial interventions (pp. 95-
120). New York: Oxford University Press.
Finch, B., Hummer, R., Reindl, M., & Vega, W. (2002). Validity of self-rated health
among Latino(a)s. American Journal of Epidemiology, 155, 755-759.
Finnegan, J. R., & Viswanath, K. (2008). Communication theory and health behavior
change: The media studies framework. In K. Glanz, B. K. Rimer & K. Viswanath
(Eds.), Health behavior and health education: Theory, research, and practice
(4th ed.). San Francisco: Jossey-Bass.
158

Fitten, L. J., Ortiz, F., & Pontón, M. (2001). Frequency of Alzheimer's disease and other
dementias in a community outreach sample of Hispanics. Journal of the
American Geriatrics Society, 49, 1301-1308.
Fitzgibbon, M. L., Gapstur, S. M., & Knight, S. J. (2004). Results of Mujeres Felices por
ser Saludables: A dietary/breast health randomized clinical trial for Latino
women. Annals of Behavioral Medicine, 28, 95-104.
Flores, G. (2000). Culture and the patient-physician relationship: Achieving cultural
competency in health care. Journal of Pediatrics, 136, 14-23.
Ford, E. S., & Giles, W. H. (2003). Changes in prevalence of nonfatal coronary heart
disease in the United States from 1971-1994. Ethnicity and Disease, 13, 85-93.
Ford, E. S., Giles, W. H., & Dietz, W. H. (2002). Prevalence of the metabolic syndrome
among U.S. adults: Findings from the Third National Health and Nutrition
Examination Survey. JAMA, 287, 356-359.
Fuchs, R., Heath, G. W., & Wheeler, F. C. (1992). Perceived morbidity as a
determinant of health behavior. Health Education Research, 7, 327-334.
Gallivan, J., Brown, C., Greenberg, R., & Clark, C. M. (2009). Predictors of perceived
risk of the development of diabetes. Diabetes Spectrum, 22, 163-169.
Galvin, J. E., Fu, Q., Nguyen, J. T., Glasheen, C., & Scharff, D. (2008). Psychosocial
determinants of intention to screen for Alzheimer's disease. Alzheimers
Dementia, 4, 353-360.
Gillette-Guyonnet, S., Abellan Van Kan, G., Andrieu, S., Barberger-Gateau, P., Berr,
C., Bonnefoy, M., et al. (2007). IANA task force on nutrition and cognitive
decline with aging. The Journal of Nutrition, Health and Aging, 11, 132-152.
Glanz, K., Rimer, B. K., & Viswanath, K. (Eds.). (2008). Health behavior and health
education: Theory, research, and practice. San Francisco: Jossey-Bass
Publishers.
159

Glass, T. A., & McAtee, M. J. (2006). Behavioral science at the crossroads in public
health: Extending horizons, envisioning the future. Social Science & Medicine,
62, 1650-1671.
Globe, D. R., Paz, S., Yu, E., PrestonMartin, S., Azen, S., Varma, R., et al. (2002).
Using focus groups to develop a culturally sensitive methodology for
epidemiological surveys in a Latino population: Findings from the Los Angeles
Latino Eye Study (LALES). Ethnicity & Disease: An International Journal on
Population Differences in Disease Patterns, 12, 259-266.
Goff, D. C., Nichaman, M. Z., & Chan, W. (1997). Greater incidence of hospitalized
myocardial infarction among Mexican-Americans than non-Hispanic Whites:
The Corpus Christi Heart Project, 1988-1992. Circulation, 95, 1433-1440.
Gonzales-Gross, M., Marcos, A., & Pietrzik, K. (2001). Nutrition and cognitive
impairment in the elderly. British Journal of Nutrition, 86, 313-320.
Gray, G. E. (1989). Nutrition and dementia. Journal of the American Dietetic
Association, 89, 1795-1802.
Greene, G. W., Rossi, S. R., Rossi, J. S., Velicer, W. F., Fava, J. L., & Prochaska, J. O.
(1999). Dietary applications of the stages of change model. Journal of the
American Dietetic Association, 99, 673-678.
Griffith, R., Dunwoody, S., & Neuwirth, K. (1999). Proposed model of the relationship of
risk information seeking and processing to the development of preventive
behaviors. Environmental Research, 80, 230-245.
Grundy, S. M., Benjamin, I. J., Burke, G. L., Chait, A., Eckel, R. H., Howard, B. V., et
al. (1999). Diabetes and cardiovascular disease: A statement for healthcare
professionals from the American Heart Association Circulation, 100, 1134-1146.
Haan, M. N., Mungas, D. M., Gonzalez, H. M., Ortiz, T. A., Acharya, A., & Jagust, W. J.
(2003). Prevalence of dementia in older Latinos: The influence of type II
diabetes mellitus, stroke, and genetic factors. Journal of the American
Geriatrics Society, 51, 169-177.
160

Hanlon, P., McEwen, J., Carey, L., Gilmour, H., Tannahill, C., Tannahill, A., et al.
(1995). Health checks and coronary risk: Further evidence from a randomised
controlled trial. BMJ, 16, 1609-1613.
Harwood, D. G., Barker, W. W., Loewenstein, D. A., Ownby, R. L., St. George-Hyslop,
P., Mullan, M., et al. (1999). A cross-ethnic analysis of risk factors for AD in
White Hispanics and White non-Hispanics. Neurology, 52, 551-556.
Hayes, A. M., Laurenceau, J., & Cardaciotto, L. (2008). Methods for capturing the
process of change. In A. M. Nezu & C. M. Nezu (Eds.), Evidence-based
outcome research: A practical guide to conducting randomized controlled trials
for psychosocial interventions (pp. 335-358). New York: Oxford University
Press.
Hendrie, H. C. (1998). Epidemiology of dementia and Alzheimer's disease. American
Journal of Geriatric Psychiatry, 6 (Supplement), S3-S18.
Hendrie, H. C., Albert, M. S., Butters, M. A., Gao, S., Knopman, D. S., Launer, L. J., et
al. (2006). The NIH Cognitive and Emotional Health Project: Report of the
critical evaluation study committee. Alzheimer's & Dementia, 2, 12-32.
Hooper, L., Summerbell, C. D., Higgins, J. P. T., Thompson, R. L., Clements, G.,
Capps, N., et al. (2001). Reduced or modified dietary fat for preventing
cardiovascular disease. Cochrane Database of Systematic Reviews, Art. No.:
CD002137. DOI: 002110.001002/14651858.CD14002137.
Hornik, R., & Kelly, B. (2007). Communication and diet: An overview of experience and
principles. Journal of Nutrition Education Behavior, 39, S5-S12.
Horowitz, C. R., Tuzzio, L., Rojas, M., Monteith, S., & Sisk, J. E. (2004). How do urban
African Americans and Latinos view the influence of diet on hypertension?
Journal of Health Care for the Poor and Underserved, 15, 631-644.
Howard-Pitney, B., Winkleby, M. A., Albright, C. L., Bruce, B., & Fortmann, S. P.
(1997). The Stanford Nutrition Action Program: A dietary fat intervention for low-
literacy adults. American Journal of Public Health, 87, 1971-1976.

161

Howell, D. C. (2008). Mixed models for repeated (longitudinal) data [Electronic
Version]. Retrieved July 23, 2010 from
3Hhttp://www.uvm.edu/~dhowell/StatPages/More_Stuff/Mixed%20Models%20for
%20Repeated%20Measures.pdf.
Hu, F. B., & Willett, W. C. (2002). Optimal diets for prevention of coronary heart
disease. JAMA, 288, 2569-2578.
Hughes, T. F., & Ganguli, M. (2009). Modifiable midlife risk factors for late-life cognitive
impairment and dementia. Current Psychiatry Review, 5, 73-92.
Janz, N. K. (1988). The health belief model in understanding cardiovascular risk factor
reduction behaviors. Cardiovascular Nursing, 24, 39-41.
Janz, N. K., & Becker, M. H. (1984). The health belief model: A decade later. Health
Education Quarterly, 11, 1-47.
Kahlor, L., Dunwoody, S., Griffin, R. J., Neuwirth, K., & Giese, J. (2003). Studying
heuristic-systematic processing of risk communication. Risk Analysis, 23, 355-
368.
Kalmijn, S. (2000). Fatty acid intake and the risk of dementia and cognitive decline: A
review of clinical and epidemiological studies. Journal of Nutrition and Healthy
Aging, 4, 202-207.
Kalmijn, S., Fesken, E. J. M., Launer, L. J., Stijnen, T., & Kromhout, D. (1995).
Glucose-intolerance, hyperinsulinemia and cognitive function in a general-
population of elderly men. Diabetologia, 38, 1096-1102.
Kalmijn, S., Launer, L. J., Ott, A., Witteman, J. C. M., Hofman, A., & Breteler, M. M. B.
(1997). Dietary fat intake and the risk of incident dementia in the Rotterdam
Study. Annals of Neurology, 42, 776-782.
Kazdin, A. E. (2003). Research design in clinical psychology (4th ed.). Boston: Allyn &
Bacon.
Kidd, R. F. (1976). Manipulation checks: Advantage or disadvantage? Representative
Research in Social Psychology, 1976, 160-165.
162

Klein, S., Burke, L. E., Bray, G. A., Blair, S., Allison, D. B., Pi-Sunyer, X., et al. (2004).
Clinical implications of obesity with specific focus on cardiovascular disease: A
statement for professionals from the American Heart Association Council on
Nutrition, Physical Activity, and Metabolism: endorsed by the American College
of Cardiology Foundation. Circulation, 110, 2952-2967.
Kloppenborg, R. P., van den Berg, E., Kappelle, L. J., & Biessels, G. J. (2008).
Diabetes and other vascular risk factors for dementia: Which factor matters
most? A systematic review European Journal of Pharmacology, 585, 97-108.
Knapp, J. A., Hazuda, H. P., Haffner, S. M., Young, E. A., & Stern, M. P. (1988). A
saturated fat/cholesterol avoidance scale: Sex and ethnic differences in a
biethnic population. Journal of the American Dietetic Association, 88, 172-177.
Knight, B. G., Robinson, G. S., Flynn Longmire, C. V., Chun, M., Nakao, K., & Kim, J.
H. (2002). Cross cultural issues in caregiving for persons with dementia: Do
familism values reduce burden and distress? Ageing International, 27, 70-94.
Kreuter, M. W., Lukwago, S. N., Bucholtz, D. C., Clark, E. M., & Sanders-Thompson, V.
(2002). Achieving cultural appropriateness in health promotion programs:
Targeted and tailored approaches. Health Education and Behavior, 30, 133-
136.
Kreuter, M. W., & McClure, S. M. (2004). The role of culture in health communication.
Annual Review of Public Health, 25, 439-455.
Kristal, A. R., Beresford, S. A., & Lazovich, D. (1994). Assessing change in diet-
intervention research. American Journal of Clinical Nutrition, 59, 185S-189S.
Kristal, A. R., Glanz, K., Curry, S. J., & Patterson, R. E. (1999). How can stages of
change be best used in dietary interventions? Journal of the American Dietetic
Association, 99, 679-684.
Kristal, A. R., Shattuck, A. L., & Henry, H. J. (1990). Patterns of dietary behavior
associated with selecting diets low in fat: Reliability and validity of a behavioral
approach to dietary assessment. Journal of the American Dietetic Association,
90, 214-220.
163

Kumanyika, S. K., Obarzanek, E., Stettler, N., Bell, R., Field, A. E., Fortmann, S. P., et
al. (2008). Population-based prevention of obesity: The need for
comprehensive promotion of healthful eating, physical activity, and energy
balance: a scientific statement from American Heart Association Council on
Epidemiology and Prevention, Interdisciplinary Committee for Prevention
(formerly the Expert Panel on Population and Prevention Science). Circulation,
118, 428-464.
Kumanyika, S. K., Van Horn, L., Bowen, D., Perri, M. G., Rolls, B. J., Czajkowski, S.
M., et al. (2000). Maintenance of dietary behavior change. Health Psychology,
19 (Supplement), 42-56.
Kunz, R., Vist, G. E., & Oxman, A. D. (2007). Randomisation to protect against
selection bias in healthcare trials. Cochrane Database of Systematic Reviews,
Art. No.: MR000012. DOI:
000010.001002/14651858.MR14000012.pub14651852.
Laroche, H. H., Hofer, T. P., & Davis, M. M. (2007). Adult fat intake associated with the
presence of children in households: Findings from NHANES III. Journal of the
American Board of Family Medicine, 20, 9-15.
Launer, L. J. (2006). Prevention of AD: The which, when, and on whom? Alzheimer
Disease & Associated Disorders, 20, Supplement 2, S75-S78.
Leander, K., Hallqvist, J., Reuterwall, C., Ahlbom, A., & de Faire, U. (2001). Family
history of coronary heart disease, a strong risk factor for myocardial infarction
interacting with other cardiovascular risk factors: Results from the Stockholm
Heart Epidemiology Program (SHEEP). Epidemiology, 12, 215-221.
Lee, S., Colditz, G. A., Berkman, L. F., & Kawachi, I. (2003). Caregiving and risk of
coronary heart disease in U.S. women: A prospective study. American Journal
of Preventive Medicine, 24, 113-119.
Levine, G., & Parkinson, S. (1994). Experimental methods in psychology. Hillsdale:
Lawrence Erlbaum Associates, Inc.
Lundgren, R. E., & McMakin, A. H. (2009). Risk communication: A handbook for
communicating environmental, safety, and health risks. Hoboken: John Wiley &
Sons.
164

Maestre, G., Ottman, R., Stern, Y., Gurland, B., Chun, M., Tang, M. X., et al. (1995).
Apolipoprotein E and Alzheimer's disease: Ethnic variation in genotypic risks.
Annals of Neurology, 37, 254-259.
Mayeux, R., Stern, Y., Ottman, R., Tatemichi, T. K., Tang, M. X., Maestre, G., et al.
(2004). The apolipoprotein 4 allele in patients with Alzheimer's disease. Annals
of Neurology, 34, 752 - 754.
McCusker, M. E., Yoon, P. W., Gwinn, M., Malarcher, A. M., Neff, L., & Khoury, M. J.
(2004). Family history of heart disease and cardiovascular disease risk-
reducing behaviors. Genetics in Medicine, 6, 153-158.
McIntosh, W. A., & Zey, M. (1989). Women as gatekeepers of food consumption: A
sociological critique. Food and Foodways, 3, 317-332.
McMahon, S., Cathorall, M., & Romero, D. R. (2007). Cardiovascular disease risk
perception and knowledge: A comparison of Hispanic and White college
students in a Hispanic-serving institution. Journal of Hispanic Higher Education,
6, 5-18.
MetLife Foundation. (2006). MetLife Foundation Alzheimer's Survey: What America
thinks. Retrieved April 23, 2007
Mier, N., Ory, M. G., & Medina, A. A. (2010). Anatomy of culturally sensitive
interventions promoting nutrition and exercise in Hispanics: A critical
examination of existing literature Health Promotion Practice, 11, 541-554.
Miles, M. B., & Huberman, M. (1994). Qualitative data analysis: An expanded
sourcebook (2nd ed.). Thousand Oaks: Sage Publications.
Morgan, M. G., Fischhoff, B., Bostrom, A., & Atman, C. J. (2002). Risk communication:
A mental models approach. New York City: Cambridge University Press.
Morris, M. C., Evans, D. A., Bienias, J. L., Tangney, C. C., & Wilson, R. S. (2004).
Dietary fat intake and 6-year cognitive change in an older biracial community
population. Neurology, 62, 1573-1579.
165

Mosca, L., Ferris, A., Fabunmi, R., & Robertson, R. M. f. t. A. H. A. (2004). Tracking
women's awareness of heart disease: An American Heart Association national
study. Circulation, 109, 573-579.
Mozaffarian, D., Katan, M. B., Ascherio, A., Stampfer, M. J., & Willett, W. C. (2006).
Trans fatty acids and cardiovascular disease. New England Journal of
Medicine, 354, 1601-1613.
Muntner, P., Garrett, E., Klag, M. J., & Coresh, J. (2002). Trends in stroke prevalence
between 1973 and 1991 in the U.S. population 25 to 74 years of age. Stroke,
33, 1209-1213.
Negy, C., & Woods, D. J. (1992). The importance of acculturation in understanding
research with Hispanic Americans. Hispanic Journal of Behavioral Sciences,
14, 224-247.
Neuhouser, M. L., Thompson, B., Coronado, G. D., & Solomon, C. C. (2004). Higher
fat intake and lower fruit and vegetables intakes are associated with greater
acculturation among Mexicans living in Washington state. Journal of the
American Dietetic Association, 104, 51-57.
Nezu, A., & Nezu, C. M. (2008). Ensuring treatment integrity. In A. Nezu & C. M. Nezu
(Eds.), Evidence-based outcome research: A practical guide to conducting
randomized controlled trials for psychosocial interventions (pp. 263-281). New
York: Oxford University Press.
Nezu, A. M., & Nezu, C. M. (2008). The "devil is in the details": Recognizing and
dealing with threats to validity in randomized controlled trials. In A. M. Nezu &
C. M. Nezu (Eds.), Evidence-based outcome research: A practical guide to
conducting randomized controlled trials for psychosocial interventions (pp. 3-
24). New York: Oxford University Press.
Notkola, I. L., Sulkava, R., Pekkanen, J., Erkinjuntti, T., Ehnholm, C., & Kivinen, P.
(1998). Serum total cholesterol, apolipoprotein E e4 allele, and Alzheimer's
disease. Neuroepidemiology, 17, 14-20.
Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New York:
McGraw-Hill, Inc.
166

Obayashi, S., Bianchi, L. J., & Song, W. O. (2003). Reliability and validity of nutrition
knowledge, social-psychological factors, and food label use scales from the
1995 Diet and Health Knowledge Survey. Journal of Nutrition Education and
Behavior, 35, 83-92.
Olin, J. T., Dagerman, K. S., Fox, L. S., Bowers, B., & Schneider, L. S. (2002).
Increasing ethnic minority participation in Alzheimer disease research.
Alzheimer Disease & Associated Disorders, 16 Supplement, S82-S85.
Park, Y., Zhu, S., Palaniappan, L., Heshka, S., Carnethon, M. R., & Heymsfield, S. B.
(2003). The metabolic syndrome: Prevalence and associated risk factor findings
in the U.S. population from the Third National Health and Nutrition Examination
Survey, 1988-1994 Archives of Internal Medicine, 4, 427-436.
Paskett, E. D., Reeves, K.W., McLaughlin, J.M., Katz, M.L., McAlearney, A.S., Ruffin,
M.T., Halbert, C.H., Merete, C., Davis, F., Gehlert, S. (2008). Recruitment of
minority and underserved populations in the United States: The centers for
population health and health disparities experience. Contemporary Clinical
Trials, 29, 847-861.
Plassman, B. L., Langa, K. M., Fisher, G. G., Heeringa, S. G., Weir, D. R., Ofstedal, M.
B., et al. (2007). Prevalence of dementia in the United States: The Aging,
Demographics, and Memory Study. Neuroepidemiology, 29, 125-132.
Prochaska, J. O., & DiClemente, C. C. (1986). Towards a comprehensive model of
change. In W. Miller & N. Heather (Eds.), Treating addictive behaviors (pp. 3-
27). New York: Plenum Press.
Rimal, R. N. (2001). Perceived risk and self-efficacy as motivators: Understanding
individuals' long-term use of health information. Journal of Communication, 51,
633-654.
Rimal, R. N., & Real, K. (2003). Perceived risk and efficacy beliefs as motivators of
change: Use of the risk perception attitude (RPA) framework to understand
health behaviors. Human Communication Research, 29, 370-399.
Roberts, J. S., Connell, C. M., Cisewski, D., Hipps, Y. G., Demissie, S., & Green, R. C.
(2003). Differences between African Americans and Whites in their perceptions
of Alzheimer disease. Alzheimer Disease & Associated Disorders, 17, 19-26.
167

Rosenstock, I. M. (1974). Historical origins of the health belief model. Health Education
Monographs, 2, 328-335.
Rosenstock, I. M. (1982). The health belief model and nutrition education. Journal of
the Canadian Dietetic Association, 43, 184-193.
Rosenstock, I. M., Stretcher, V. J., & Becker, M. H. (1988). Social learning theory and
the health belief model. Health Education Quarterly, 15, 175-184.
Safeguarding cognitive health in an ageing population. (2006). Lancet Neurology, 5,
283.
Salerno-Kennedy, R., & Cashman, K. D. (2005). Relationship between dementia and
nutrition-related factors and disorders: An overview. International Journal for
Vitamin and Nutrition Research, 75, 83-95.
Samsa, G. P., Cohen, S. J., Goldstein, L. B., Bonito, A. J., Duncan, P. W., Enarson, C.,
et al. (1997). Knowledge of risk among patients at increased risk for stroke.
Stroke, 28, 916-921.
Sapp, S. G., & Jensen, H. H. (1997). Reliability and validity of nutrition knowledge and
diet-health awareness tests developed from the 1989-1991 Diet and Health
Knowledge Surveys. Journal of Nutrition Education, 29, 63-72.
SAS Institute. (2009). SAS online documentation for Windows (Version release 9.2).
Satia-Abouta, J., Patterson, R. E., Neuhouser, M. L., & Elder, J. P. (2002). Dietary
acculturation: Application to nutrition research and dietetics. Journal of the
American Dietetic Association, 102, 1105-1118.
Satia, J. A., Patterson, R. E., Kristal, A. R., Hislop, T. G., Yasui, Y., & Taylor, V. M.
(2001). Development of dietary acculturation scales among Chinese Americans
and Chinese Canadians. Journal of the American Dietetic Association, 101,
548-553.
Scarmeas, N., Stern, Y., Tang, M.-X., Mayeux, R., & Luchsinger, J. A. (2006).
Mediterranean diet and risk for Alzheimer's disease. Annals of Neurology, 59,
912-921.
168

Schwarzer, R. (2008). Modeling health behavior change: How to predict and modify the
adoption and maintenance of health behaviors. Applied Psychology, 57, 1-29.
Sevush, S., Peruyera, G., Crawford, F., & Mullan, M. (2000). Apolipoprotein-E
[varepsilon]4 allele frequency and conferred risk for Cuban Americans With
Alzheimer's disease. The American Journal of Geriatric Psychiatry, 8, 254-256.
Shannon, J., Kristal, A. R., Curry, S. J., & Beresford, S. A. (1997). Application of a
behavioral approach to measuring dietary change: The fat- and fiber-related
diet behavior questionnaire. Cancer Epidemiology Biomarkers & Prevention, 6,
355-361.
Sharkey, J. R., Sharf, B. F., & St. John, J. A. (2009). "Una persona derechita (staying
right in the mind)": Perceptions of Spanish-speaking Mexican American older
adults in South Texas colonias. The Gerontologist, 49, S79-S85.
Sheeran, P. (2002). Intention-behavior relations: A conceptual and empirical review.
European Review of Social Psychology, 12, 1-36.
Silagy, C., Muir, J., Coulter, A., Thorogood, M., & Roe, L. (1993). Cardiovascular risk
and attitudes to lifestyle: What do patients think? BMJ, 306, 1657-1660.
Solfrizzi, V., Colacicco, A. M., D'Introno, A., Capurso, C., Torres, F., Rizzo, C., et al.
(2006). Dietary intake of unsaturated fatty acids and age-related cognitive
decline: A 8.5-year follow-up of the Italian Longitudinal Study on Aging.
Neurobiology of Aging, 27, 1694-1704.
Solfrizzi, V., D'Introno, A., Colacicco, A. M., Capurso, A., Del Parigi, A., Capurso, S., et
al. (2005). Dietary fatty acid intake: Possible role in cognitive decline and
dementia. Experimental Gerontology, 40, 257-270.
Solomon, R. L. (1949). An extension of control group design. Psychological Bulletin,
46, 137-150.
Sternfeld, B., Cauley, J., Harlow, S., Liu, G., & Lee, M. (2000). Assessment of physical
activity with a single global question in a large, multiethnic sample of midlife
women. American Journal of Epidemiology, 152, 678-687.
169

Stewart, R., & Liolitsa, D. (1999). Type 2 diabetes mellitus, cognitive impairment and
dementia. Diabetic Medicine, 16, 93-112.
Stroup-Benham, C. A., Markides, K. S., Espino, D. V., & Goodwin, J. S. (1999).
Changes in blood pressure and risk factors for cardiovascular disease among
older Mexican Americans from 1982-1984 to 1993-1994. Journal of the
American Geriatrics Society, 47, 804-810.
Sundquist, J., & Winkleby, M. A. (1999). Cardiovascular risk factors in Mexican
American adults: A transcultural analysis of NHANES III, 1988-1994. American
Journal of Public Health, 89, 723-730.
Tang, M. X., Stern, Y., Marder, K., Bell, K., Gurland, B., Lantigua, R., et al. (1998). The
APOE-epsilon4 allele and the risk of Alzheimer disease among African
Americans, Whites, and Hispanics. JAMA, 279, 751-755.
The California Policy Research Center. (1998). The California Latino demographic
databook: Regents of the University of California.
Toft, U. N., Kristoffersen, L. H., Aadahl, M., von Huth Smith, L., Pisinger, C., &
Jørgensen, T. (2007). Diet and exercise intervention in a general population--
mediators of participation and adherence: The Inter99 study. European Journal
of Public Health, 17, 455-463.
Torgerson, D. J., Roberts, C. (1999). Understanding controlled trials: Randomisation
methods: Concealment. BMJ, 1999, 375-376.
Trotter, R. C., II., & Chavira, J. A. (1997). Cuaranderismo: Mexican American folk
healing. Athens: University of Georgia Press.
Turk, M. W., Yang, K., Hravnak, M., Sereika, S. M., Ewing, L. J., & Burke, L. E. (2009).
Randomized clinical trials of weight loss maintenance: A review. Journal of
Cardiovascular Nursing, 24, 58-80.
U.S. Census Bureau. (2004a). U.S. interim projections by age, sex, race, and Hispanic
origin. Retrieved March 17, 2007, from
4Hhttp://www.census.gov/ipc/www/usinterimproj/
170

U.S. Census Bureau. (2004b). We the people: Hispanics in the U.S. Retrieved March
17, 2007, from 5Hhttp://www.census.gov/prod/2004pubs/censr-18.pdf
U.S. Census Bureau. (2007). State & County Quickfacts: Los Angeles County,
California. Retrieved August 16, 2007, from
6Hhttp://quickfacts.census.gov/qfd/states/06/06037.html
U.S. Department of Agriculture, A. R. S., Beltsville Human Nutrition Research Center,
Food Surveys Research Group (1994-1996). Continuing Survey of Food
Intakes by Individuals 1994-96, 1998 and Diet and Health Knowledge Survey
1994-96.
U.S. Department of Health and Human Services. (1998). Call to action: Eliminating
racial and ethnic disparities in health. Washington, DC: Grantmakers in Health.
Varma, R., Paz, S. H., Azen, S. P., Klein, R., Globe, D. R., Torres, M., et al. (2004).
The Los Angeles Latino Eye Study: Design, methods, and baseline data.
Ophthalmology, 111, 1121-1131.
Villarreal, R., Blozis, S. A., & Widaman, K. F. (2005). Factorial invariance of a Pan-
Hispanic Familism Scale. Hispanic Journal of Behavioral Sciences, 27, 409-
425.
Wansink, B. (2002). Changing habits on the home front: Lost lessons from World War
II research. Journal of Public Policy and Marketing, 201, 90-99.
Wansink, B. (2005). Marketing nutrition: Soy, functional foods, biotechnology, and
obesity. Urbana: University of Illinois Press.
Wansink, B. (2006). Nutritional gatekeepers and the 72% solution. Journal of the
American Dietetic Association, 106, 1324-1327.
Wardle, J., Parmenter, K., & Waller, J. (2000). Nutrition knowledge and food intake.
Appetite, 34, 269-275.
Watson, G. S., Peskind, E. R., Asthana, S., Purganan, K., Wait, C., Chapman, D., et al.
(2003). Insulin increases CSF Abeta42 levels in normal older adults. Neurology,
60, 1899-1903.
171

Weiss, B. D., Mays, M. Z., Martz, W., Castro, K. M., Dewalt, D. A., Pignone, M. P., et
al. (2005). Quick assessment of literacy in primary care: The Newest Vital Sign.
Annals of Family Medicine, 3, 514-522.
Whitmer, R. A., Gustafson, D. R., Barrett-Connor, E., Haan, M. N., Gunderson, E. P., &
Yaffe, K. (2008). Central obesity and increased risk of dementia more than
three decades later. Neurology, 71, 1057-1064.
Winkleby, M. A., Flora, J. A., & Kraemer, H. C. (1994). A community-based heart
disease intervention: Predictors of change. American Journal of Public Health,
84, 767-772.
Xu, W., Qiu, C., Gatz, M., Pedersen, N. L., Johansson, B., & Fratiglioni, L. (2009). Mid-
and late-life diabetes in relation to the risk of dementia: A population-based twin
study. Diabetes, 58, 71-77.
Yancey, A. K., Kumanyika, S. K., Ponce, N. A., McCarthy, W. J., Fielding, J. E., Leslie,
J. P., et al. (2004). Population-based interventions engaging communities of
color in healthy eating and active living: A review. Preventing Chronic Disease:
Public Health Research, Practice, and Policy, 1, 1-18.
Yehuda, S., Rabinovitz, S., & Mostofsky, D. I. (1999). Essential fatty acids are
mediators of brain biochemistry and cognitive functions. Journal of
Neuroscience Research, 56, 565-570.
Yeo, G., & Gallagher-Thompson, D. (Eds.). (1996). Ethnicity and the dementias (1 ed.).
San Francisco: Taylor & Francis.
Yoon, P. W., Scheuner, M. T., Peterson-Oehlke, K. L., Gwinn, M., Faucett, A., &
Khoury, M. J. (2002). Can family history be used as a tool for public health and
preventive medicine? Genetics in Medicine, 4, 304-310.
Zar, J. H. (1998). Biostatistical analysis (4th ed.). Englewood Cliffs: Prentice Hall.

172

Appendix A: Certificate of Completion

173

Appendix B: Pilot Test Findings

Key culture-specific communication strategies based on promotoras’ field experiences:

INFORMED CONSENT & RECRUITMENT MATERIALS

ISSUE: Emphasizing healthy foods on the recruitment flyer* will not attract attention,
as healthy
foods are not a priority among Latinos.
SOLUTION: Emphasize healthy foods as tasty foods.

ISSUE: The photos of Latinas on the recruitment flyer were very professionally
dressed Latinas, potentially intimidating community-dwelling participants.
SOLUTION: Replace with photos of casually dressed Latinas.

ISSUE: The recruitment flyer “sounded” too formal.
SOLUTION: Make the flyer more informal (e.g., by using a fun font such as Comic
Sans, applying a colorful border).

ISSUE: It may be especially intimidating for less acculturated Latinas to sign up for the
study on their own.
SOLUTION: Invite the reader to bring a female friend or relative along, if they want.

ISSUE: Latinas, particularly those of Mexican descent, are becoming increasingly
wary to participate in any community offerings due to immigration concerns of being
reported.
SOLUTION: Emphasize: (1) the reasons the identifying information is needed (e.g., to
contact for follow-up, to link surveys across time points); and (2) strict confidentiality of
information will be maintained (will not travel beyond study personnel).

*A recruitment flyer was originally pilot tested. However, since the completion of pilot
testing, it was decided that a recruitment letter would be a better recruitment tool.
Therefore, these findings were incorporated into the construction of the letter.

STUDY MEASURES

ISSUE: Very low literacy levels in study area.
SOLUTION: Administer survey questionnaires in an interview format (in Spanish or
English). This strategy also builds personalismo, or personal rapport between
interviewer and participant, a key aspect of communication in the Latino community.

ISSUE: Participants will have immigration concerns as to why they are asked their
birthdate.
SOLUTION: Ask for their current age (either open-endedly, or with age ranges).

174

ISSUE: Participants may feel uncomfortable mentioning that they had no formal
schooling, they are in an unmarried, long-term relationship, or that they are currently
unemployed due to disability.
SOLUTION: Include these choices as response options.

ISSUE: The demographics item querying for the number of total people living in a
home may not provide a valid response because several families often reside under
one roof.
SOLUTION: Increase item validity by asking ‘Is there more than one family in your
home? If so, how many people are in your immediate family?’. Query by current total
family income rather than total household income.

ISSUE: Measures mentioning the use of margarine or butter in the cooking process
will not be culturally valid.
SOLUTION: Use lard (manteca) or corn oil (aceite de maiz).

INTERVENTION MATERIALS

ISSUE: Maintain a visibly clear connection between Latino food and heart health in
the study logo.
SOLUTION: Incorporate the Latin American Food Pyramid (LAFP) with a heart
design. According to the pilot testers’ experiences, the LAFP is becoming increasingly
familiar to Latinos.

ISSUE: Latinos possess inadequate levels of dietary fat knowledge and healthier fat
behaviors.
SOLUTION: Physically show fat (e.g., fat replicas), clearly explain different types of
fat, explain the fat-disease connection, have participants discuss the positive and
negative consequences of high fat diets (e.g., physical appearance is a large
motivating factor). Present dietary behaviors related to food procurement (e.g., reading
nutrition labels), preparation (e.g., substituting healthier fats).

ISSUE: Often, Latinas take into primary consideration their families’ taste preferences
while purchasing and preparing food.
SOLUTION: Prominently highlight the critical role female family caregivers play in the
maintenance of not only their own, but also their family’s nutritional health and well-
being.

ISSUE: Higher fat products are perceived as containing more nutrients (e.g., whole
milk). Also, lower fat products are not as readily available in Latino communities as
their higher fat counterparts.
SOLUTION: (1) Address the nutritional offerings of products with a range of fat levels
(e.g., milk) and (2) show economical ways of procuring and preparing delicious culture-
specific dishes.

175

Appendix C: Study Logos

Heart Plus Brain Group:

Heart Only Group:

176

Appendix D: Study Timeline

177

Appendix E: Permission to Recruit Letter

178

Appendix F: Motivational Telephone Call Script (in English)

BUENOS HABITOS ALIMENTICIOS PARA UNA BUENA SALUD
MOTIVATIONAL PHONE CALL

Participant Simple ID: #___

Group: A B (circle one)

Name:

Language: Spanish or English (circle
one)

Phone Number:

Posttest Date: __/__/__

Scheduled Call Date: __/__/__

Actual Call Date: __/__/__

If leaving a message:

• Hello! My name is ______________ & I’m calling from the nutrition study
that you are a part of at the LALES Clinic called the Good Eating Habits for
Good Health Study or Buenos Habitos Alimenticios Para Una Buena Salud.

• Two weeks have passed/nearly passed since your last nutrition class and
we were calling to find out how you are doing and to learn about your
current nutrition goals and to schedule your next interview at the clinic. As
you may remember, after you complete this next interview, you will receive a
$25 Target gift card.

• Please call us back at your earliest convenience. We can be reached at the
study office at (213) 740-0864. Thank you very much, and we look forward
to hearing from you soon.

179

If talking to a live person:

Talking points:

• Hello! My name is ______________ & I’m calling from the nutrition study
that you are a part of at the LALES Clinic called the Good Eating Habits for
Good Health Study or Buenos Habitos Alimenticios Para Una Buena Salud.

• Two weeks have passed/nearly passed since your last nutrition class and
we were calling to find out how you are doing. Do you have any nutritional
goals for yourself?

• What are your current nutrition goals?

• How are you doing with these nutrition goals?

• What barriers or obstacles are you facing with these goals?

• What benefits do you see by making these changes?

• Give appropriate level and amount of motivation…something like: You’re
doing a great job/you’re well on your way to making some very important
and healthy changes to your diet…keep up the good effort…what a
difference you are having/will have to the health of you and your family.

• Lastly, we would like to now schedule your 3
rd
interview in 2 weeks. At the
end of this interview, you will receive a $25 Target gift card.

Schedule follow-up date on or very soon thereafter:
______________________________________

180

• Do you have any questions? In the future, please do not hesitate to
contact us if you have any questions. Our study number once again is
(213) 740-0864. Thank you and goodbye.

181

Appendix F: Motivational Telephone Call Script (in Spanish)

BUENOS HABITOS ALIMENTICIOS PARA UNA BUENA SALUD
LLAMADA DE MOTIVACIÓN

Participant Simple ID: #___

Group: A B (circle one)

Name:

Language: Spanish or English (circle
one)

Phone Number:

Posttest Date: __/__/__

Scheduled Call Date: __/__/__

Actual Call Date: __/__/__

If leaving a message:

• ¡Hola! Me llamo ______________ y estoy llamando del estudio de nutrición
Buenos Habitos Alimenticios Para Una Buena Salud.

• Han pasado dos semanas desde que tomo su clase de nutricion y estamos
llamanado para ver como esta y como se encuentra en sus metas de
nutrición. Tambien queremos citar la proxima entrevista en la clinica.
Recuerde que despues
De completar la entrevista final en la clinica, recibira una tarjeta de regalo
de $25.00 de la tienda Target.

• Por favor regrese esta llamada cuando sea conveniente. Puede
contactarnos en la oficina del estudio a (213) 740-0864. Muchas gracias y
esperamos hablar con usted pronto!

182

If talking to a live person:

Talking points:

• Hola! Mi nombre es ____________ y le estoy hablando del estudio de
nutrición Buenos Habitos Alimenticios Para Una Buena Salud.

• Como dos semanas han pasado desde su ultima clase de nutrición le
estamos hablando para ver como esta.

• Cuales son sus metas actuales de nutrición?

• Como va con estas metas?

• Se ha enfrentado con obstáculos o barrerás al completar sus metas?

• Cuales son los beneficios que usted ve al hacer estos cambios?

• Give appropriate level and amount of motivation…something like: Muy buen
trabajo!...Esta en el buen camino de hacer cambios importantes y sanas en
su dieta… Siga con su buen esfuerzo!

• Finalmente, queremos citarla para su tercera entrevista en dos semanas. Al
final de esta entrevista le entregaremos una tarjeta de Target de $25.00.

Schedule follow-up date on or very soon thereafter:
______________________________________

183

• Tiene cualquier pregunta? En el futuro, por favor no se detenga en
hablarnos. El numero del estudio es (213) 740-0864. Gracias y que tenga
buen dia.

184

Appendix G: Waiting List Letter (in English)

June 12, 2008

Re: Good Eating Habits for Good Health Research Project

Dear Participant:

You recently participated in interviews for a nutritional education project for Latinas in
La Puente titled “Buenos Habitos Alimenticios Para Una Buena Salud” (“Good Eating
Habits for Good Health”). At this time, we would like to invite you to participate in the
nutrition workshops that will be conducted at the LALES Clinic in La Puente. During a
two week period, there will be two workshops. Each workshop will last for two hours.
The material covered in these workshops has the potential to help you and your family
make healthier eating habit choices. In these workshops, you will join other Latinas
from LALES to learn about healthy fats, participate in tasty cooking demonstrations,
and learn and share healthy shopping and cooking tips. Your participation is voluntary.
You must be aged 18 or older to participate.

If you are interested in joining a nutrition workshop, learning more about this study or
have any questions after reading this letter, you can contact Ms.
Poorni Otilingam or any of the research assistants by phone at (213) 740-0864.

I appreciate your time in helping us with this research and I can answer any questions
that you might have about it.

Thank you for your time and cooperation!

Sincerely,

Poorni Otilingam, M.P.H., M.A.
Doctoral Candidate
Department of Psychology
University of Southern California

Margaret Gatz, Ph.D.
Professor,
Department of Psychology
University of Southern California

Date of Preparation: December 17, 2007
USC UPIRB # UP-07-00423

185

Appendix G: Waiting List Letter (in Spanish)

Junio 12, 2008

Re: El Proyecto Buenos Hábitos Alimenticios para la Buena Salud

Estimada Participante:

Recientemente usted participó en entrevistas para un proyecto de educación sobre la
nutrición para latinas de La Puente titulado “Buenos Hábitos Alimenticios Para Una
Buena Salud” (“Good Eating Habits for Good Health”). Ahora, quisiéramos invitarla
para participar en los talleres de nutrición que se llevarán a cabo en la Clínica de
LALES de La Puente. Durante un período de dos semanas, se realizarán dos talleres.
Cada taller durará dos horas. El material que se cubrirá en estos talleres tiene el
potencial de ayudarle a usted y a su familia a escoger hábitos alimenticios más
saludables. En estos talleres, usted se unirá a otras latinas de LALES para aprender
de grasas saludables, participar en demostraciones deliciosas de cocina, y compartir y
enterarse de recomendaciones sanas para comprar y preparar comestibles. Su
participación es voluntaria. Debe de tener por lo menos 18 años de edad para
participar.

Si está interesada en participar en un taller de nutrición, aprender más sobre este
estudio o si después de leer esta carta tiene cualquier pregunta al respecto, sírvase
comunicarse con la Srta. Poorni Otilingam o con cualquiera de los asistentes de
investigación al teléfono (213) 740-0864.

Le agradezco mucho el tiempo que dedica a ayudarnos con esta investigación y estoy
dispuesta a contester cualquier pregunta que tenga al respecto.

¡Gracias por su tiempo y cooperación!

Atentamente,

Poorni Otilingam, M.P.H., M.A.
Candidata Doctoral
Departamento de Psicología
Universidad del Sur de California

Margaret Gatz, Ph.D.
Profesora,
Departamento de Psicología
Universidad del Sur de California

Date of Preparation: December 17, 2007
USC UPIRB # UP-07-00423

186

Appendix H: Recruitment Letter (in English)

187

Appendix H: Recruitment Letter (in English), cont.

188

Appendix H: Recruitment Letter (in Spanish)

189

Appendix H: Recruitment Letter (in Spanish), cont.

190

Appendix I: Telephone Screening & Intake (in English)

DATE/TIME CALLED __________________________________

WHEN SPEAKING WITH NEW RESPONDENT: GOOD
(MORNING/AFTERNOON/EVENING). MY NAME IS _____________________ AND I
REPRESENT THE GOOD EATING HABITS FOR GOOD HEALTH STUDY OF USC, THE
UNIVERSITY OF SOUTHERN CALIFORNIA. ARE YOU CALLING TO FIND OUT HOW YOU
CAN TAKE PART IN A RESEARCH STUDY ABOUT NUTRITION EDUCATION IN THE
LATINO COMMUNITY LIVING IN LA PUENTE? WELL, THANK YOU FOR YOUR INTEREST
IN THIS STUDY. AS YOU KNOW, THIS IS A NUTRITION EDUCATION STUDY STRICTLY
FOR PARTICIPANTS IN LALES, THE LOS ANGELES LATINO EYE STUDY. BEFORE WE
GO ON, WE NEED TO MAKE SURE THAT YOU ARE ELIGIBLE TO PARTICIPATE IN THE
NUTRITION EDUCATION STUDY. I WILL ASK YOU FIVE QUICK QUESTIONS. LET’S
BEGIN.

1. ARE YOU THE PERSON WHO WILL BE PARTICIPATING IN THIS STUDY?
1. Yes
a. (DON’T ASK) INTERVIEWER: IS THE CALLER
FEMALE?
i. Yes GO TO Q2.
ii. No GO TO Q1.2.
2. No
a. Will the participant for whom you are calling on behalf
of be female?
i. Yes GO TO Q2.
ii. No* (EXCLUDE FROM STUDY)

2. DO YOU EXPECT TO BE IN THE LOS ANGELES AREA FOR THE NEXT 3
MONTHS?
1. Yes
2. No* (EXCLUDE FROM STUDY)

3. ARE YOU CURRENTLY ON A SPECIAL DIET FOR MEDICAL REASONS?
1. Yes* (EXCLUDE FROM STUDY)
2. No

4. WE ASK THIS QUESTION TO EVERY PARTICIPANT, SO PLEASE BEAR WITH US.
IS THERE ANY CHANCE THAT YOU ARE CURRENTLY PREGNANT, COULD BE
PREGNANT, OR ARE ACTIVELY TRYING TO GET PREGNANT IN THE NEXT 3
MONTHS?
1. Yes* (EXCLUDE FROM STUDY)
2. No

5. WE ALSO ASK THIS QUESTION TO EVERY PARTICIPANT, SO WE APPRECIATE
YOUR PATIENCE. DO YOU BELIEVE OR HAVE YOU IN RECENT TIMES BEEN
TOLD BY A MEDICAL PROFESSIONAL THAT YOU HAVE AN EATING DISORDER
LIKE ANOREXIA OR BULIMIA NERVOSA?
1. Yes* (EXCLUDE FROM STUDY)
2. No

191

GOOD EATING HABITS FOR GOOD HEALTH

TELEPHONE INTAKE

NAME________________________________________________________________

ADDRESS____________________________________________________________

PHONE_______________________________________________________________

DO YOU HAVE ANOTHER PHONE NUMBER AT WHICH YOU CAN BE REACHED?

_____________________________________________________________________

TO HELP US IN COMMUNICATING WITH YOU, DO YOU PREFER TO SPEAK IN:
SPANISH OR ENGLISH?

THE NEXT STEP IS FOR YOU TO MEET ONE OF OUR RESEARCH STAFF AT THE
LALES CLINIC TO HAVE YOU REVIEW AND SIGN THE INFORMED CONSENT
PAPERWORK BEFORE WE CAN ENROLL YOU IN THE STUDY. THE MEETING
WILL TAKE APPROXIMATELY 20 MINUTES TO AN HOUR, DEPENDING ON WHICH
STUDY GROUP YOU WILL BE ASSIGNED TO AFTER YOU ARE ENROLLED IN
THE STUDY.

COULD YOU TELL US WHICH OF THE FOLLOWING TIMES THIS WEEK OR NEXT
WEEK THAT BEST FITS YOUR SCHEDULE TO MEET ONE OF OUR RESEARCH
STAFF AT THE CLINIC?

192

FANTASTIC! WE VERY MUCH LOOK FORWARD TO MEETING YOU THEN. JUST
TO BE SURE, THE LALES CLINIC ADDRESS IS: 15151 FAIRGROVE AVENUE, LA
PUENTE, CA 91744.

ALRIGHT, JUST TO BE SURE, WE HAVE YOU CONFIRMED FOR:
__________________ AT _________. IS THAT CORRECT? ALSO, ONE OF OUR
THE STUDY STAFF WILL BE CALLING TO REMIND YOU OF THE MEETING THE
DAY BEFORE. IF THERE ARE ANY CHANGES, PLEASE CALL US AT 213.740.0864.

DO YOU HAVE ANY QUESTIONS BEFORE WE WRAP UP THIS PHONE CALL?
THANK YOU FOR YOUR INTEREST IN THE BUENOS HABITOS STUDY.
GOODBYE.

For study personnel only:

DO THE FOLLOWING AFTER YOU GET A PARTICIPANT:

1. POST TO ONLINE CALENDAR (DO NOT LIST PARTICIPANT’S NAME, ONLY
THEIR #)
2. EMAIL THE RA’S WHO ARE AVAILABLE ON THAT DAY (LOOK AT WEEKLY
SCHEDULE FOR AVAILABILITIES)
3. WRITE ON OFFICE WALL CALENDAR

IF PARTICIPANT REFUSED OR NO LONGER IS INTERESTED IN THE STUDY,
STATE THE REASON HERE:
_____________________________________________________________

INTAKE TAKEN BY: _________________________ (YOUR NAME)

193

Appendix I: Telephone Screening & Intake (in Spanish)

DATE/TIME CALLED: __________________________________

WHEN SPEAKING WITH NEW RESPONDENT: BUENOS (DIAS/TARDES/NOCHES). MI
NOMBRE ES _____________________. YO SOY UN/UNA REPRESENTANTE DEL
ESTUDIO BUENOS HABITOS ALIMENTICIOS PARA LA BUENA SALUD DE LA
UNIVERSIDAD DEL SUR DE CALIFORNIA. ¿ESTA LLAMANDO EN BUSCA DE
INFORMACION PARA PARTICIPAR EN UN ESTUDIO IMPORTANTE RELACIONADO A LA
EDUCACION NUTRITIVA EN LA COMMUNIDAD LATINA DE LA PUENTE? BUENO,
GRACIAS POR SU INTERES EN ESTE ESTUDIO. COMO YA SABRA, ESTE ES UN
ESTUDIO EDUCATIVO DE NUTRICION SOLAMENTE PARA PARTICIPANTES DE LALES,
EL ESTUDIO DE OJOS DE LATINOS EN LOS ANGELES. ANTES DE CONTINUAR,
TENEMOS QUE VERIFICAR QUE SEA ELEGIBLE PARA PARTICIPAR EN NUESTRO
ESTUDIO EDUCATIVO DE NUTRICION. LE VOY A HACER 5 PREGUNTAS RAPIDAS.
VAMOS A COMENZAR.

1. ¿ ES USTED LA PERSONA QUE PARTICIPARIA EN EL ESTUDIO?
1. Yes
a. (DON’T ASK) INTERVIEWER: IS THE CALLER
FEMALE?
i. Yes GO TO Q2.
ii. No GO TO Q1.2.
2. No
b. ¿La persona por quien llama, es mujer?
i. Yes GO TO Q2.
ii. No* (EXCLUDE FROM STUDY)

2. ¿PIENSA ESTAR EN LOS ANGELES DURANTE LOS PROXIMOS 3 MESES?
1. Yes
2. No* (EXCLUDE FROM STUDY)

3. ¿SIGUE ACTUALMENTE UNA DIETA ESPECIAL POR RAZONES MEDICAS?
1. Yes* (EXCLUDE FROM STUDY)
2. No

4. LE HACEMOS ESTA PREGUNTA A TODAS LAS PARTICIPANTES, LE PEDIMOS
DISCULPAS Y SU COOPERACION. ¿EXISTE LA POSIBILIDAD DE QUE ESTE
ACTUALMENTE EMBARAZADA, QUE PUEDA ESTAR EMBARAZADA O QUE ESTE
ACTIVAMENTE INTENTANDO QUEDAR EMBARAZADA DURANTE LOS PROXIMOS 3
MESES?
1. Yes* (EXCLUDE FROM STUDY)
2. No

5. TAMBIEN LE HACEMOS ESTA PREGUNTA A TODAS LAS PARTICIPANTES.
AGRADECEMOS SU PACIENCIA. ¿CREE QUE TENGA O LE HA DICHO UN
PROFESIONAL DE MEDICINA QUE PUEDE SER QUE TENGA UN TRASTORNO
ALIMENTICIO COMO ANOREXIA O BULIMIA NERVIOSA?
1. Yes* (EXCLUDE FROM STUDY)
2. No

194

GOOD EATING HABITS FOR GOOD HEALTH

TELEPHONE INTAKE

NOMBRE_____________________________________________________________

DIRECCION___________________________________________________________

TELEFONO___________________________________________________________

¿TIENE OTRO NUMERO DE TELEFONO A TRAVES DEL CUAL PODREMOS
COMMUNICARNOS ?

_____________________________________________________________________

PARA PODER COMUNICARNOS CON USTED PREFIERE HABLAR EN:
ESPAÑOL O INGLES?

EL SIGUIENTE PASO ES PARA QUE USTED PUEDA JUNTARSE CON UNO DE
NUESTROS REPRESENTANTES DEL ESTUDIO EN LA CLINICA LALES PARA QUE
USTED PUEDA REVISAR Y FIRMAR EL FORMULARIO DE CONSENTIMIENTO
ANTES DE INSCRIBIRLA. LA JUNTA DURARA APROXIMADAMENTE 20 MINUTOS
A UNA HORA DEPENDIENDO A CUAL GRUPO DE ESTUDIO LA ASIGNAREMOS.

NOS PUDIERA PORFAVOR DECIR CUAL DE ESTOS HORARIOS DE ESTA
SEMANA O LA SIGUIENTE SEMANA MEJOR LE COMBIENE PARA JUNTARSE
CON NUESTRO REPRESENTANTE DEL ESTUDIO EN LA CLINICA LALES:

195

FANTASTICO! ESTAMOS EMOCIONADOS DE CONOCERLA. NADAMAS PARA
RECORDARLE LA DIRECIÓN DE LA CLINICA LALES 15151 FAIRGROVE AVENUE,
LA PUENTE, CA 91744.

BUENO, PARA ASEGURARNOS, YA ESTAMOS CONFIRMADOS PARA
________________ A LAS _______________. ESTA CORRECTO? TAMBIÉN, UNO
DE NOS REPRESENTANTES DEL ESTUDIO LA LLAMARA UN DIA ANTES PARA
RECORDARLE DE LA JUNTA. SI HAY CUALQUIER CAMBIO LLAMENOS AL 213-
740-0864.

TIENE OTRAS PREGUNTAS ANTES DE TERMINAR ESTA LLAMADA? GRACIAS
POR SU INTERES EN EL ESTUDIO BUENOS HABITOS. QUE TENGA UN BUEN
DÍA. ADIOS!

For study personnel only:

DO THE FOLLOWING AFTER YOU GET A PARTICIPANT:

4. POST TO ONLINE CALENDAR (DO NOT LIST PARTICIPANT’S NAME, ONLY
THEIR #)
5. EMAIL THE RA’S WHO ARE AVAILABLE ON THAT DAY (LOOK AT WEEKLY
SCHEDULE FOR AVAILABILITIES)
6. WRITE ON OFFICE WALL CALENDAR

IF PARTICIPANT REFUSED OR NO LONGER IS INTERESTED IN THE STUDY,
STATE THE REASON HERE:
_____________________________________________________________

INTAKE TAKEN BY: _________________________ (YOUR NAME)

196

Appendix J: Information Brochure (in English)

197

Appendix J: Information Brochure (in Spanish)

198

Appendix K: Information Sheet for Non-Medical Research (in English)

University of Southern California
Department of Psychology

INFORMATION SHEET FOR NON-MEDICAL RESEARCH

Habitos Alimenticios Para Una Buena Salud

You are asked to participate in a research study conducted by Poorni Otilingam,
M.P.H., M.A. and her faculty sponsor, Margaret Gatz, Ph.D., from the
Department of Psychology at the University of Southern California. Results will
be contributed to Ms. Otilingam’s doctoral dissertation. You were selected as a
possible participant in this study because you are a female participant in the
Los Angeles Latino Eye Study (LALES). You must be at least 18 years of age to
participate. You should not participate in this study if you are pregnant, on a
special diet for medical reasons, or have or recently have had an eating
disorder such as anorexia or bulimia because this will not reflect regular eating
habits of the general population. You also should not participate in this study if
you are moving or planning to move away permanently from the Los Angeles
area in the next three months. A total of 240 female subjects will be selected
from LALES to participate. Your participation is voluntary. Please take as much
time as you need to read the information sheet. You may also decide to discuss
it with your family or friends. You will be given a copy of this form.

PURPOSE OF THE STUDY
We are asking you to take part in a research study because we are trying to
learn more about the eating habits and nutrition and health knowledge of
Latinas in Los Angeles.

Completion and return of the questionnaire or response to the interview
questions will constitute consent to participate in this research project.

PROCEDURES
You will be asked to participate in the following ways:

Interviews
We request for you to make a good faith effort to complete 2-3 face-to-face
interviews. Research assistants will interview you in your preferred language
(Spanish or English). Questions asked on the interview will be based on your

199

current eating habits, nutrition knowledge, and other survey questions. You
have the right to refuse to answer any question that you do not wish to answer.

Each interview will last between 30 to 45 minutes. The interviews will happen
over a period of one and a half months. Interviews will happen at the LALES
Clinic. In the event that you are not able to come to the LALES Clinic, we will do
our best to administer the interview by telephone or at your home.

Nutrition Education Classes
We request for you to make a good faith effort to participate in two nutrition
education classes. Research assistants will conduct the classes in your
preferred language (Spanish or English). The classes will teach you about good
eating habits for you and your family. The nutrition classes will be presented in
a group format. You are welcome to participate in the class’s games, food
demonstrations, and discussions. You have the right to refuse participation in
any or all parts of the classes at any time should you choose. Each class will
last for two hours. The classes will happen over a period of two weeks. There
will be a total of two classes. Therefore, a total of four hours is needed to
complete the nutrition education classes. Classes will be held at the LALES
Clinic. Both daytime and evening classes will be offered, and participants may
attend classes at the time that is more convenient for them. Free child care will
be available during the classes if needed.

Group Assignment
Some participants will receive the nutrition classes before the end of the
interviews, and some will receive the nutrition classes after the interviews. You
will be randomly assigned to one of four types of project groups that either will
participate in the classes sooner or later in the research project. You will be
randomized into one of the four project groups by random choice; the process
of random choice is like tossing a coin. The outcome is purely by chance alone.

Length of Time Involvement
A maximum of six hours will be needed of your time for the study. Some
participants will require less time to complete the study.

POTENTIAL RISKS AND DISCOMFORTS
There are no anticipated risks to your participation. You initially may experience
mild discomfort as a result of healthy changes that you may be making to your
eating habits as a result of participating in the study. We anticipate that support
from your peer classmates and from the study’s staff will help you with these
changes. You may also experience mild discomfort, mild boredom or fatigue

200

from answering questions and possibly minor anxiety about whether you know
the answers or not. If any of the questions make you uncomfortable, you can
skip that question and not answer it. A no-cook food demonstration will be
conducted in the nutrition education classes. All precautions will be taken to
minimize all food-related risks.

POTENTIAL BENEFITS TO SUBJECTS AND/OR TO SOCIETY
You may not directly benefit from your participation in this research study.
However, you may find the interviews and nutrition education sessions
interesting. You also may learn more about nutrition, health and healthy cooking
as a consequence of your participation. The project has an anticipated benefit
to science by using the results from this study as a means to improve eating
habits and overall health in the society-at-large.

PAYMENT/COMPENSATION FOR PARTICIPATION
Each time you complete an interview you will receive a small token of
appreciation such as two (2) movie tickets and a $25 Target gift card. You will
receive a $25 Target gift card at study completion. Some participants will also
be asked to complete one to two additional interviews. You will be paid each
time you make a good faith effort to complete an interview. At the end of the
first additional interview, you will receive a $10 Target gift card. At the end of
the second additional interview, you will receive two movie tickets.

POTENTIAL CONFLICTS OF INTEREST
The investigators of this research do not have any financial interest in the
sponsor or in the product being studied.

CONFIDENTIALITY
Any information that is obtained in connection with this study and that can be
identified with you will remain confidential and will be disclosed only with your
permission or as required by law. The information collected about you will be
coded using initials and numbers, for example abc-123, etc. The information
which has your identifiable information will be kept separately from the rest of
your data.

Only members of the research team will have access to the data associated
with this study. The data will be stored in the investigator’s office in a locked file
cabinet/password protected computer.

The data will be stored for at least three years after the study has been
completed and then destroyed. After the three year period, the data can be kept

201

indefinitely or the data can be destroyed. The maximum length of time the data
will be maintained is ten (10) years.

When the results of the research are shared with a dissertation committee,
published or discussed in conferences, no information will be included that
would reveal your identity.

PARTICIPATION AND WITHDRAWAL
You can choose whether to be in this study or not. If you volunteer to be in this
study, you may withdraw at any time without consequences of any kind. You
may also refuse to answer any questions you don’t want to answer and still
remain in the study. The investigator may withdraw you from this research if
circumstances arise which warrant doing so.

ALTERNATIVES TO PARTICIPATION
Your alternative is to not participate. Your participation in the Latino Eye Study
will not be affected. The researcher in the Latino Eye Study will not know
whether or not you participate in the nutritional (food) study.

RIGHTS OF RESEARCH SUBJECTS
You may withdraw your consent at any time and discontinue participation
without penalty. You are not waiving any legal claims, rights or remedies
because of your participation in this research study. If you have any questions
about your rights as a study subject or you would like to speak with someone
independent of the research team to obtain answers to questions about the
research, or in the event the research staff can not be reached, please contact
the University Park IRB, Office of the Vice Provost for Research Advancement,
Stonier Hall, Room 224a, Los Angeles, CA 90089-1146, (213) 821-5272 or
7Hupirb@usc.edu

IDENTIFICATION OF INVESTIGATORS
If you have any questions or concerns about the research, please feel free to
contact the research personnel. To speak to research personnel in English,
please call the Faculty Sponsor, Dr. Margaret Gatz at (213) 740-2212 or the
Student Co-Investigator, Ms. Poorni Otilingam at (213) 740-0864. To speak to
research personnel in Spanish, please call the main study number at (213) 740-
0864 to reach a bilingual Research Assistant. The mailing address for all
research personnel is The Good Eating Habits and Good Health Project, c/o
Poorni Otilingam, The University of Southern California, Department of
Psychology, 3620 McClintock Avenue SGM501, Los Angeles, CA 90089-1061.

202

Appendix K: Information Sheet for Non-Medical Research (in Spanish)

University of Southern California
Department of Psychology

HOJA DE INFORMACIÓN PARA INVESTIGACIÓN NO
MÉDICA

Hábitos Alimenticios Para Una Buena Salud

Se le pide que participe en un estudio investigativo llevado a cabo por Poorni
Otilingam, M.P.H., M.A. y su consejera de facultad, Margaret Gatz, Ph.D., del
Departamento de Psicología de la Universidad del Sur de California. Los
resultados se contribuirán a la disertación doctoral de la Srta.Otilingam. Usted
fue escogida como posible participante en este estudio porque es una
participante femenina en El Estudio de Ojos de Latinos de Los Angeles
(LALES). Debe tener por lo menos 18 años de edad para participar. No debe
participar en este estudio si está embarazada, sigue una dieta especial por
razones médicas, o sufre o ha sufrido recientemente de un trastorno alimenticio
como anorexia o bulimia ya que ambos padecimientos no reflejan hábitos
alimenticios normales de la población en general. También no debe participar
en este estudio si está mudándose o piensa mudarse permanentemente de los
alrededores de Los Angeles en los próximos tres meses. Un total de 240
sujetos femeninos serán seleccionados de LALES para participar. Su
participación es voluntaria. Por favor tome todo el tiempo que necesite para
leer esta hoja de información. Tal vez quiera también discutirla con su familia o
amistades. Se le dará una copia de este formulario.

PROPÓSITO DEL ESTUDIO
Le pedimos que tome parte en un estudio investigativo porque estamos
tratando de aprender más acerca de los hábitos alimenticios y los
conocimientos sobre la nutrición y la salud de las latinas de Los Angeles.

Al completer y devolver el cuestionario y responder a las preguntas de la
entrevista, usted de hecho da su consentimiento para participar en este
proyecto investigativo.

PROCEDIMIENTOS
Se le pedirá que participe de las siguientes maneras:

203

Entrevistas
Le pedimos que haga un esfuerzo de buena fe para completar 2-3 entrevistas
de cara a cara. Los asistentes de investigación la entrevistarán en su idioma
preferido (español o inglés). Las preguntas durante la entrevista se basarán en
sus hábitos alimenticios actuales, sus conocimientos acerca de la nutrición, y
otras preguntas de encuesta. Tiene el derecho de negarse a contestar
cualquier pregunta que no desee contestar. Cada entrevista durará entre 30 a
45 minutos. Las entrevistas se realizarán durante el transcurso de mes y
medio. Usted támbien puede ser invitada a participar en dos entrevistas
adicionales en el futuro - una seis meses despues del estudio y otra un ano
despues del studio. Las entrevistas tendrán lugar en la Clínica de LALES. En
caso de que no pueda venir a la Clínica de LALES, haremos todo lo posible
para que las entrevistas se realicen por teléfono o en su hogar.

Clases de Educación Sobre la Nutrición
Le pedimos que haga un esfuerzo de buena fe para participar en dos clases
de educación sobre la nutrición. Los asistentes de investigación darán las
clases en su idioma preferido (español o inglés). Las clases le informarán
acerca de buenos hábitos alimenticios para usted y su familia. Las clases de
nutrición se presentarán en un formato de grupo. En la clase podrá participar
en juegos, demostraciones de cocina, y en discusiones. Tiene el derecho de
negarse a participar en cualquier parte o en toda parte de las clases cuando lo
desee. Cada clase durará dos horas. Las clases transcurrirán durante un
período de dos semanas. Habrá un total de dos clases. De modo que, un total
de cuatro horas se necesitará para completar las clases de educación sobre la
nutrición. Las clases se efectuarán en la Clínica de LALES. Se ofrecerán clases
de día y de noche, y las participantes podrán asistir a las clases que mejor les
convengan. También habrá cuidado de niños gratis para quienes lo necesiten.

Como Se Le Asigna A Un Grupo
Algunas participantes recibirán las clases de nutrición antes de terminarse las
entrevistas, y otras recibirán las clases de nutrición después de las entrevistas.
Usted será escogida al azar para participar en uno de los cuatro tipos de
grupos del proyecto; dicho grupo participará en las clases de nutrición más
temprano o más tarde en el transcurso del proyecto investigativo. El proceso de
hacer la selección al azar es como tirar una moneda. El resultado es pura
suerte.

La Duración De Su Participación
Un máximo de seis horas de su tiempo se necesitará para este estudio.
Algunas participantes necesitarán menos tiempo para completar el estudio.

204

RIESGOS Y MOLESTIAS POSIBLES
No hay riesgos previstos respecto a su participación. Tal vez usted
experimente un leve malestar que resulta de los cambios saludables que esté
haciendo en sus hábitos de comer como resultado de su participación en el
estudio. Esperamos que el apoyo de sus compañeras de clase y del personal
del estudio le ayudará con estos cambios. También quizás experimente una
leve molestia, leve aburrimiento o cansancio de contestar preguntas y tal vez
un poco de ansiedad porque no sepa una respuesta. Si una pregunta la hace
sentirse incómoda,no tiene que contestarla. Habrá una demostración de cocina
sin cocinar en las clases de educación nutritiva. Toda precaución se tomará
para reducir todo riesgo relacionado con la comida.

BENEFICIOS POSIBLES AL SUJETO Y/O LA SOCIEDAD
Tal vez usted no se beneficie directamente de su participación en este estudio
investigativo. Sin embargo, quizás le resulten interesantes las entrevistas y las
clases de nutrición. También puede aprender más acerca de la nutrición, la
salud y la cocina saludable como consecuencia de su participación. El proyecto
tiene un beneficio previsto para la ciencia puesto que los resultados de este
estudio podrán usarse para mejorar los hábitos alimenticios y la salud en
general de la sociedad.

PAGO/COMPENSACIÓN POR PARTICIPAR
Cada vez que termine una entrevista, recibirá una muestra de nuestro
agradecimiento en forma de dos (2) entradas al cine y una tarjeta de regalo
Target de $25. Recibirá una tarjeta de regalo Target de $25 una vez que se
termine el estudio. Se les pedirá a algunas participantes que participen en una
o dos entrevistas adicionales. Se le pagará cada vez que haga un esfuerzo de
buena fe por completar una entrevista. Una vez terminada la primera entrevista
adicional, recibirá una tarjeta de regalo Target de $10. Una vez terminada la
segunda entrevista adicional, recibirá dos entradas al cine.

POSIBLES CONFLICTOS DE INTERESES
Los investigadores de este estudio no tienen ningún interés financiero en el
patrocinador ni en el producto que se estudia.

CONFIDENCIALIDAD
Cualquier información que se obtenga relacionada con este estudio y que
pueda identificarse con usted será confidencial y se divulgará solamente con su
permiso o si lo requiere la ley. La información que se recoge sobre usted será
puesta en clave usando iniciales y números, por ejemplo abc-123, etc. La

205

información que contiene su información identificable se mantendrá aparte del
resto de sus datos.

Solamente los miembros del equipo de investigación tendrán acceso a los
datos relacionados a este estudio. Los datos se archivarán en la oficina del
investigador en un archivo con llave/una computadora protegida por
contraseña.

Los datos serán archivados por tres años por lo menos después de terminarse
el estudio y entonces destruidos. Después de un período de tres años, los
datos pueden mantenerse archivados por tiempo indefinido o pueden
destruirse. El máximo período de tiempo que pueden archivarse los datos es
diez (10) años.

Cuando los resultados de la investigación se compartan con el comité de
disertación, se publiquen o se discutan en charlas, ninguna información se
incluirá que pueda revelar su identidad.

PARTICIPACIÓN Y RETIRADA
Usted puede decidir si quiere participar en este estudio o no. Si se ofrece
voluntariamente a participar en este estudio, puede retirarse en cualquier
momento sin consecuencias de ninguna clase. También puede negarse a
contestar cualquier pregunta y todavía seguir en el estudio. El investigador
puede retirarla de esta investigación si las circunstancias lo exigen.

ALTERNATIVAS A LA PARTICIPACIÓN
Su alternativa u opción es no participar. Su participación en el Estudio Latino
de Ojos no será afectada. El investigador del Estudio Latino de Ojos no se
enterará si ha participado en el estudio nutritivo (de alimentos) o no.

DERECHOS DEL SUJETO DE LA INVESTIGACIÓN
Usted puede retirar su consentimiento en cualquier momento y dejar de
participar en el estudio sin consecuencias adversas. Por su participación en
este estudio investigativo, usted no renuncia su derecho de presentar una
demanda o cualquier otro remedio legal. Si tiene preguntas respecto a sus
derechos como sujeto de una investigación o si quisiera hablar con alguien que
no esté asociado con el equipo investigativo para que le conteste preguntas
sobre el estudio, o si no puede comunicarse con el personal de investigación,
sírvase comunicarse con University Park IRB, Office of the Vice Provost for
Research Advancement, Stonier Hall, Room 224a, Los Angeles, CA 90089-
1146, (213) 821-5272 or 8Hupirb@usc.edu

206

IDENTIFICACIÓN DE LOS INVESTIGADORES
Si tiene preguntas o inquietudes sobre la investigación, sírvase comunicarse
con el personal investigativo. Para hablar con el personal investigativo en
inglés, por favor llame a la Patrocinadora de la Facultad, la Dra. Margaret Gatz
al (213) 740-2212 o a la Co-Investigadora Estudiantil, la Srta. Poorni Otilingam

al (213) 740-0864. Para hablar con el personal investigativo en español, por
favor llame al número principal del estudio (213) 740-0864 y le contestará un
asistente de investigación bilingüe. La dirección para todo personal
investigativo es The Good Eating Habits and Good Health Project, c/o Poorni
Otilingam, The University of Southern California, Department of Psychology,
3620 McClintock Avenue SGM501, Los Angeles, CA 90089-1061.

207

Appendix L: Interview Packet

INTERVIEW

GOOD EATING HABITS FOR GOOD HEALTH

START TIME

:

INTERVIEWER, PLEASE COMPLETE:

PARTICIPANT FULL ID: ____ ____ ____

DATE: ____/____/____

INTERVIEWER INITIALS: _____________

208

FOR EVERY RESPONDENT: THANK YOU FOR PARTICIPATING IN THE GOOD
EATING HABITS AND GOOD HEALTH STUDY. I WOULD LIKE TO ASK YOU
SOME QUESTIONS ABOUT YOUR DIET, YOUR GENERAL HEALTH, AND YOUR
EATING, FOOD SHOPPING, AND FOOD PREPARATION PATTERNS. THE
INTERVIEW WILL TAKE ABOUT 45 MINUTES. AFTER WE FINISH THE
INTERVIEW, I WILL GIVE YOU A GIFT TO SHOW OUR APPRECIATION FOR
YOUR HELP. YOUR PARTICIPATION IS VOLUNTARY. ALL THE INFORMATION
YOU TELL ME WILL BE KEPT STRICTLY CONFIDENTIAL, AND YOU CAN
REFUSE TO ANSWER ANY QUESTIONS THAT YOU DO NOT WANT TO ANSWER.
YOU MAY ALSO REFUSE ANY PART OF THIS INTERVIEW AND IT WON’T
AFFECT YOUR PARTICIPATION IN THE STUDY.

OK, LET’S BEGIN.

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

RESPONSE OPTION CARD H

229

230

231

232

233

234

235

236

237

238

239

240

Appendix M: Facilitator Adherence

To: Facilitator _______________

Please complete this brief survey by checking off the material/instruction that
you and/or your co-facilitator covered in this workshop. Be honest.

Material Covered?

Fat versus Muscle……………………………………………………
Benefits & Barriers of Eating Less Fat……………………………..
The Chain Game (carabiners)………………………………………
Types of Fats (plants versus animals)……………………………..
Bottle Cap Home Activity……………………………………………
Heart Disease Discussion…………………………………………...
The Brain Connection………………………………………………..
Achieve Your Goals by Taking Small Steps………………………
Local Food Directory for La Puente, CA & Surrounding Areas….
Buy Healthy Foods on a Budget……………………………………
Supermarket Map…………………………………………………….
Buying Snacks (Snack Attack Poster)……………………………..
Food Pyramid…………………………………………………………
Lotería Game…………………………………………………………
Reading Food Labels………………………………………………..
Cooking with Less Fat……………………………………………….
Cooking Demo………………………………………………………..
Portion Size…………………………………………………………...
Serving Size…………………………………………………………..
Balanced Plate………………………………………………………..
Eating During Holidays & Celebrations…………………………….
Eating Out……………………………………………………………..

241

Appendix N: Facilitator Competence

To: Facilitator _______________

Please complete this brief survey by indicating your beliefs about today’s
workshop. Be honest.

1. How effective do you believe today’s workshop was in helping participants
obtain better eating habits?

1 2 3 4 5
Strongly
ineffective
Ineffective Neutral Effective Strongly
effective

242

Appendix O: Participant Satisfaction (in English)

Before you leave today, please complete a brief satisfaction survey. It is
completely anonymous, so please be as honest as you’d like.

1. Below are three faces – a happy face, a neutral face, and a sad face. Please rate
whether you enjoyed this workshop by circling the face that is most similar to how
you feel.

☺

2. Please rate whether you believe this workshop will help you in meeting your
nutritional goals by circling the face that is most similar to how you feel.

☺

3. Please rate the likelihood that you will recommend this workshop to a friend by
circling the face that is most similar to how you feel.

☺

4. Please rate the likelihood that you will participate in another workshop in the
future by circling the face that is most similar to how you feel.

☺

5. Please rate any overall nutritional changes that you have seen in yourself after
participating in this workshop by circling the face that most similar to how you
feel.

☺

243

Appendix O: Participant Satisfaction (in Spanish)

Antes de irse hoy, por favor complete esta breve encuesta de satisfacción. Por
favor conteste honestamente, y recuérde que sus opiniones son completamente
anónimas!

1. Debajo hay tres caras - una cara contenta, una cara neutral, y una cara triste.
Por favor indique si disfruto esta clase con circulando la cara que es mas
semejante a como se siente.

☺

2. Por favor indique si usted cree que estas clases de nutrición le van a ayudar a
lograr sus metas de nutrición con circulando la cara que es mas semejante a como
se siente.

☺

3. Por favor indique cual es la probabilidad de que usted recomendara estas clases
de nutrición a una amiga con circulando la cara que es mas semejante a como se
siente.

☺

4. Por favor indique la probabilidad de que usted participe en otra clase de
nutrición en el futuro con circulando la cara que es mas semejante a como se
siente.

☺

5. Por favor indique si ha visto cambios nutricionales en si misma después de
participar en estas clases de nutrición con circulando la cara que es mas semejante
a como se siente.

☺

Abstract (if available)

Abstract This study assessed the effectiveness of a nutrition intervention that emphasized the connection between heart health and brain health among Latinas. Public health researchers and practitioners have noted the rise in the number of Latinos being diagnosed with diseases involving the cardiovascular system and with risk factors that might portend future vascular disease. Vascular risk factors are important modifiable conditions for both vascular dementia and Alzheimer’s disease. As diets high in saturated fatty acids contribute to vascular risk, change in diet could lower risk for vascular disease and dementia. A randomized controlled study with four conditions was conducted: two intervention conditions (nutrition workshops that emphasized heart health and brain health, or “heart plus brain”, and workshops that emphasized heart health, or “heart only”) and two wait list conditions (“pretest-posttest” and “posttest only”). Intervention conditions and the first wait list condition were assessed at pretest, posttest, and one-month follow-up. The heart plus brain condition received materials that emphasized the importance of vascular health in reducing dementia risk as well as information about the relationship of nutrition to heart health.The heart only condition received identical material about the relationship of nutrition to heart health

Linked assets

University of Southern California Dissertations and Theses

Asset Metadata

Creator Otilingam, Poorni G. (author)

Core Title Effectiveness of a heart plus brain health-focused nutrition intervention among Latinas: a randomized controlled trial of Buenos Hábitos Alimenticios Para Una Buena Salud (Good Eating Habits for ...

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Publication Date 03/28/2011

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

Tag Clinical Psychology,Gerontology,health disparities,Nutrition education,OAI-PMH Harvest

Place Name California (states), La Puente (city or populated place)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Gatz, Margaret (committee chair), John, Richard S. (committee member), Jordan-Marsh, Maryalice (committee member), Meyerowitz, Beth (committee member)

Creator Email otilinga@gmail.com,otilinga@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3484

Unique identifier UC153634

Identifier etd-Otilingam-4147 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-422994 (legacy record id),usctheses-m3484 (legacy record id)

Legacy Identifier etd-Otilingam-4147.pdf

Dmrecord 422994

Document Type Dissertation

Rights Otilingam, Poorni G.

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu