University of Southern California Dissertations and Theses

Predictors of mammography use among California teachers

(USC Thesis Other)

Predictors of mammography use among California teachers

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UM I a complete manuscript and there are missing pages, these w ill be noted. Also, if unauthorized copyright material had to be removed, a note w ill indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. ProQuest Information and Learning 300 North Zeeb Road, Ann Arbor, M l 48106-1346 USA 800-521-0600 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PREDICTORS OF MAMMOGRAPHY USE AMONG CALIFORNIA TEACHERS by Doojduen Villaluna A Thesis Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment o f the Requirements for the Degree MASTER OF SCIENCE (APPLIED BIOSTATISTICS AND EPIDEMIOLOGY) May 2002 Copyright 2002 Doojduen Villaluna Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 1411812 ___ ® UMI UMI Microform 1411812 Copyright 2003 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UNIVERSITY O F SO U T H E R N CALIFORNIA THE GRADUATE SC H O O L UNIVERSITY PA RK LOS A N G ELES. CA LIFO R N IA § 0 0 0 7 This thesis, written by J)coidu tn Villa kin a under the direction of h &C. Thesis Committee, and approved by all its members, has been pre sented to and accepted by the Dean of The Graduate School, in partial fulfillment of the requirements for the degree of o f S cien ce- Apphecf BIcir>?hy o't-j > a a a a a a a ^ a a a a a a | a l a a ^ « S > y a a a a a a a a a a a a a * a y a a a » a a a a l / Date..„&yM*2Q&L THESIS COMMITTEE / Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DEDICATION To my husband, Fernando, for all his support and patience and my parents for their unconditional love. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS Leslie Bernstein, Ph.D. and Jane Sullivan-Halley for all their help and guidance. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS Dedication ii Acknowledgments iii List of Tables V Abstract vi Chapter I: Introduction I Chapter 2: Methods 3 A. Sample 3 B. Variables 4 C. Statistical Analysis 6 Chapter 3: Results 8 Table I 9 Table 2 11 Table 3 16 Table 4 18 Chapter 4: Discussion 19 Bibliography 23 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES TABLE 1: Frequency of mammography within age group by race/ethnicity in the CTS TABLE 2: Predictors of having a mammogram in the past 2 years among four age groups in the CTS TABLE 3: Predictors o f ever having a mammogram among four age groups in the CTS TABLE 4: Average annual age-adjusted incidence rates among women in the CTS diagnosed with a first primary breast cancer by recency of mammogram Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT Screening mammography is widely recommended for women between the ages of 50 and 69, but its utility for younger or older women has been questioned. We used baseline data collected for the California Teachers Study (CTS), a prospective cohort study of 133,479 women, to determine predictors of mammography in four age groups: 30-39 years, 40-49 years, 70-79 years and 80 years or older. Odds ratios and 95% confidence intervals were estimated by multivariate logistic regression. The key predictors of having had a mammogram within the past 2 years were a positive family history of breast cancer, having used hormones, regular vitamin use, recent exercise, and not smoking. Our findings suggest that women who are health-oriented are likely to have mammograms. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1: INTRODUCTION Breast cancer is the most common cancer diagnosed among women in the United States. Each year it is estimated that more than 180,000 women will be diagnosed with the disease and about 45,000 will die from it.1 Studies have shown that early detection can substantially decrease breast cancer mortality.2 Increasing mammography use and breast exams are widely recommended for improving early detection of this disease. The American Cancer Society (ACS), National Cancer Institute (NCI), United States Preventive Services Task Force, and several other organizations agree that women in their 50s and 60s should receive mammograms either annually or every 1-2 years.1 ’ 3 ,4 What remains controversial is whether younger or older women benefit from similar recommendations for mammography.3 * 1 0 Although the number of women who are screened has risen over time, it is important to gain insight into characteristics of women under 50 years or 70+ years who obtain mammograms. Studies have reported that older women, 75 years or older, are less likely to receive mammograms.1 '*1 3 Black or Hispanic ethnicity, low income and education levels 1214, lack of insurance l2’1 5 , and lack o f a primary care physician1 2 have also been documented as factors associated with a decreased likelihood for screening mammography. Conversely, hormone replacement therapy use l5, personal or 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. family history of cancer l4, and previous biopsies have been found to increase screening behavior. The highest screening rates have been reported in nonsmokers compared to former or current smokers l6, while obesity or overweight women have shown the lowest screening rates compared to normal-weight women.1 7 Two additional predictors o f increased screening mammography are PAP smears and clinical breast exams.1 2 1 3 ,1 8 More efforts must be made toward understanding these different effects in a multivariate context. The purpose o f this paper is to utilize information collected at baseline from members of the California Teachers Study (CTS) cohort (1) to present screening rates of teachers by age group and race, (2) to determine factors that predict screening mammography in the youngest and oldest age groups of participants (30- 39, 40-49, 70-79, 80+ years), and (3) to assess breast cancer incidence during follow-up according to mammography history. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2: METHODS A. Sample The California Teachers Study (CTS) is a prospective cohort study of 133,479 women that was initiated in 1995. Participants include current and retired female teachers and professional public school employees. The methods used to establish and maintain the cohort have been described previously (Bernstein et al, submitted 2001). Women completed a 16-page baseline questionnaire which included questions on demographics, reproductive history, health history, personal and family medical history, physical activity, diet, alcohol and tobacco use along with providing mammogram screening practices. We restricted our analyses to women who were California residents at the start of the study (93 % o f the cohort), who provided valid information on personal and family history of breast cancer and were not adopted, and who had no prior history of breast cancer (n=l 16,790). Breast cancer diagnoses prior to entry into the cohort were determined by linkage with the California Cancer Registry (CCR), the state’s population-based statewide cancer registry system established in 1988, and through self-reports on the baseline questionnaire. Our analysis was further limited to women who did not report a previous breast biopsy so that we could focus on screening practices (n=98,660). We further restricted our analyses of predictors of mammography screening to women 30 years o f age or older 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (n=93,560), since it is unlikely that younger women would undergo a screening mammogram. B. Variables Mammography. Cohort members provided information on whether they had “Ever had a mammogram” (no/yes) and “Number of years since last exam” (less than 1 year, 1 to 2 years, or 3 years/more). We compared women who never had a mammogram to those who had at least one mammogram and evaluated how recently they had had their last mammogram (<2 years ago, >2 years ago, and recency unknown). We considered that women who had a prior mammogram in the past 2 years were adhering to a recommended frequency and compared these women to those who received it less frequently (>2 years ago) or who had never had a mammogram. Potential Predictors. We examined a number of characteristics as possible predictors of mammography history in 4 age groups (30-39, 40-49, 50-59,60-69, 70-79, 80+ years). Race/ethnicity was classified as non-Hispanic white, African- American, Hispanic whites only, Asian/Pacific Islander or Other which included women who indicated mixed race or did not provide this information. Among older women (70+ years), the race/ethnicity categories were combined due to the low number of women in racial/ethnic groups other than non-Hispanic white. Body mass index (kg/M2) was categorized as <20,20.0-21.9, 22.0-24.9, 25.0, 27.4, and 27.5+ kg/M2 for younger women (30-49 years). Among older 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. women, the referent group for body mass index (BMI) was established as <22 kg/M2. We excluded or considered as unknown women whose BMI was either <16 kg/M2 or >50 kg/M2. Exercise history over the past 3 years was determined by combining the reports of the hours per week spent in strenuous and moderate activities (<l, 1-3, 4+ hrs/wk). Strenuous activities included running, jogging, basketball, cycling on hills, racquetball, swimming laps, aerobics, and calisthenics. Moderate activities included brisk walking, golf, volleyball, cycling on level streets, recreational tennis, and softball. We also considered whether cohort members used vitamin or mineral supplements regularly (at least once a week). Information was gathered on multiple vitamins such as Regular One-A-Day, Centrum, or Thera type; as well as single vitamin or mineral supplements such as vitamin A, beta-carotene, vitamin C, vitamin E, and selenium. We determined intake of alcoholic beverages over the past year, converting intake of wine, beer and spirits into total grams consumed per day (0, <5, 5-9, 10-14, 15+ g/day). Due to limited intake of alcoholic beverages by older women (70+ years), we combined alcohol consumption categories into 0, 1-9, and 10+ g/day. We classified the smoking history of cohort members into nonsmoker, former smoker, and current smoker. Respondents were considered as having ever smoked if they had smoked at least 100 cigarettes during their lifetime. We established each woman’s menopausal status (premenopausal, perimenopausal, postmenopausal and unknown) after considering at baseline her age, age at last menstrual periods, reason her periods stopped, and reported use o f 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. estrogen replacement therapy with or without addition of a progestin. Perimenopausal women were combined with postmenopausal and compared to premenopausal women in our analyses of younger women. Information on oral contraceptive (OC) use was provided and women were classified as users if they had used OCs for at least one month. OC use was classified into no-use, unknown use, or use for <5 years, 5-9 years, or at least 10 years. Where information on OC was not provided, we considered that respondents bom prior to 1916 were non users. Parous women (those who had had at least one live birth or still birth) were compared to nulliparous women. We refer to use of estrogen replacement therapy (with or without a progestin) as hormone replacement therapy (HRT) and classified women as nonusers, past users, or current users. Family history of breast cancer and ovarian cancer was based on self-report. We classified women as having a positive family history if the cancer was reported in a first-degree relative (mother, any sister, or any daughter). C. Statistical Analysis We used SAS statistical software (version 6.12) to assess predictors of mammography using multivariate logistic regression methods. For each predictor, descriptive statistics and frequency distributions were calculated. Initially, individual predictors were considered after adjustment for single year of age within the age group of interest and race/ethnicity. Likelihood ratio tests were performed to determine whether a predictor made a statistically significant contribution to the 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. multivariate model. The odds ratio (OR) was used to provide a measure of the magnitude of risk. Odds ratio estimates that are less than 1.0 represent factors that predict not having had a mammogram (or not having had a mammogram in the past 2 years) and those greater than 1.0 predict having had a mammogram (or having had a mammogram in the past 2 years). We calculated 95% confidence intervals for the OR using the standard error of the loge OR estimate. For analyses of older women, we included missing responses as a category for some variables in order to maximize the number of respondents contributing to the analysis. Trend tests were performed on ordinal variables (OC use, exercise, BMI and alcohol consumption) by fitting the median of each category. Two-sided p-values are reported for these tests for trend with values less than 0.05 are considered statistically significant. We calculated the age-adjusted incidence rates (AAIR) of breast cancer among women 30 years or older by history of mammography standardized to the 1970 US population. Annual linkages with the California Cancer Registry database have identified all cohort participants living in California who were diagnosed with their first primary breast cancer during the follow-up period (through December 31, 1998). Person months were calculated for all cohort members from the date of submission of their baseline questionnaire to the first of the following events: death, diagnosis of breast cancer, move of residence to another state, or December 31, 1998. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3: RESULTS Table 1 presents the distribution of mammography within age group by race/ethnicity in the CTS cohort. A high percentage (93%) of women aged 40-49 years have had a mammogram. The percentage remains extremely high through the 50-59, 60-69, and 70-79 year old age groups (97%, 97%, 96%, respectively). In the oldest age group (80 years or older), the percentage of women who have ever had a mammogram is lower (91%), but still remains above 90%. As might be expected, we find a low percentage of 30-39 year old women who have ever had a mammogram (35%). The proportions of women who reported a mammogram within 2 years of completing our baseline questionnaire ranged from 22% of those 30 to 39 years of age to more than 90% of those in the 50 to 69 year age range. The percentages of women within an age group who ever had a mammogram varied somewhat by race/ethnicity. The largest differences were observed for women 80 years or older where 93% of African-American women reported having had a mammogram and only 82% of the Hispanic women reported having had one. African-American women under age 50 years and in the 60 to 69 year old age group were also more likely than other racial/ethnic groups to report having had a mammogram, although the variation in mammography rates was smaller than for older women. 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE 1. Frequency of mammography within age group by race/ethnicity in the CTS C haracteristics Never Had N (%) < = 2 yrs N (%) Ever Had > 2 yrs N (%) Yes. recency unknown N (%) A G E (years) 30-39 White 7463 (65.0) 2589 (22.6) 1415(12.3) 11 (0.10) African-American 191 (63.3) 72 (23.8) 39(12.2) 0 (0 ) Hispanic 807 (67.1) 256(21.3) 137(11.4) 3 (0.3) Asian/PI 426 (70.2) 106(17.5) 74(12.2) I (0.2) Other/Mixed/None specified 296 (63.7) 100(21.5) 69(14.8) 0 (0 ) Total 9183 (65.3) 3123(22.2) 1734(12.3) 15(0.1) 40-49 White 1436(6.3) 18560(81.2) 2815(12.3) 52(0.2) African-American 43 (6.1) 540 (76.4) 120(17.0) 4 (0.6) Hispanic 137 (9.9) 1014(73.0) 229(16.5) 9(0.7) Asian/PI 97 (8.1) 916(76.8) 177(14.8) 3 (0.3) Other/Mixed/None specified 70 (9.3) 542 (72.2) 136(18.1) 3 (0.4) Total 1783(6.6) 21572 (80.2) 3477(12.9) 71 (0.3) 50-59 White 542 (2.7) 18260(91.9) 1025 (5.2) 51 (0.3) African-American 14(2.2) 601 (92.6) 30 (4.6) 4 (0.6) Hispanic 14(1.9) 658 (90.6) 51 (7.0) 3 (0.4) Asian/PI 22(2.4) 836(91.8) 50(5.5) 3 (0.3) Other/M ixed/None specified 17(3.4) 431 (86.2) 48 (9.6) 4 (0.8) Total 609 (2.7) 20786(91.7) 1204 (5.3) 65 (0.3) 60-69 White 359 (2.8) 11734 (91.5) 681 (5.3) 47 (0.4) African-American 11(2.1) 490(94.1) 19(3.7) I (0.2) Hispanic 10(2.8) 320 (90.7) 20 (5.7) 3 (0.9) Asian/PI 16(3.2) 443 (89.7) 34 (6.9) 1 (0.2) Other/Mixed/None specified 16(4.4) 321 (88.2) 22 (6.0) 5(1.4) Total 412 (2.8) 13308 (91.5) 776(5.3) 57(0.4) 70-79 White 314(3.7) 7628 (89.1) 559 (6.5) 56 (0.7) African-American 7(3.1) 197 (87.2) 17(7.5) 5(2.2) Hispanic 1 ( l.I ) 89 (95.7) 3 (3.2) 0 (0 ) Asian/PI 3(1.8) 153 (90.5) 12(7.1) 1 (0.6) Other/Mixed/None specified 17(5.1) 283 (85.2) 28 (8.4) 4 (1.2) Total 342 (3.7) 8350(89.1) 619 (6.6) 66(0.7) 80+ White 327 (9.2) 2605 (73.0) 581 (16.3) 56(1.3) African-American 3 (6.7) 32 (71.1) 10 (22.2) 0 (0 ) Hispanic 3 (17.7) 12(70.6) 2(11.8) 0 (0 ) Asian/PI 3 (10.7) 24 (85.7) I (3.6) 0 (0 ) Other/Mixed/None specified 32(11.8) 185(68.3) 43 (15.9) 11(4.1) Total 368 (9.4) 2858 (72.7) 637(16.2) 67(1.7) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. We examined predictors of having had a mammogram in the past 2 years within four age groups, 30-39 years, 40-49 years, 70-79 years and 80 years or older (Table 2). Each of the assessments adjusted for age within the age group. Age 30-39 years. Among women aged 30-39 years, African American and Hispanic white women were more likely than non-Hispanic white women to have had a recent mammogram, although none of the confidence intervals or the OR estimates excluded 1.0. Asian/Pacific Island women were less likely than non- Hispanic white women to have had a recent mammogram. The odds of having a recent screening mammogram were nearly 5-fold greater for women with a family history of breast cancer than for those without a family history. Other predictors of a recent screening mammogram in this age group were menopausal status, family history of ovarian cancer, OC use, and parity. Perimenopausal or postmenopausal women were 70% more likely to have had a mammogram in the past 2 years than premenopausal women. Young women with a family history of ovarian cancer were 1.6 times as likely to have had a recent mammogram as women without this history. Women who had used OCs were more likely to have had a recent mammogram than those who had not, and the relative odds of having had a screening mammogram increased with increasing duration of OC use (p trend <0.0001). Heavier women were less likely to have had a recent mammogram than thinner women (p tre n d =0.07), although none of the confidence intervals for the OR estimates excluded 1.0. Parous women were less likely to have had a recent mammogram. Factors that did not predict 10 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE 2 . Predictors o f having a mammogram i n the past 2 years among four age groups i n the CTS B ~ S.O o 3 p § Z Z — O z z r * » p- r* r * * < < < < Z Z Z 2 < < < < 2 2 2 2 q O o o O' C*3 c. pm w v m o oo © * Q O — p m «r 06 O' d p - r* <o r> oo C j s? I S 2 ® l I a k g I® ' < < 2 2 < < Z 2 ( N ( N O O 0 0 O Q O O ' < < < < Z Z 2 2 < < < < 2 2 2 2 r» m d o QO O ' 2 ^ § | C-w C L - ro O » o o < > ~ ~ oo e % O' 2 £ 8 a S <?<= T r>- m o pm o o — r* ■ •o -f o oo d iA oo o o so *r 0^0 0 0 w * i o oo c* oo oo O' v> d d © o f « n o «n — p m 7 7 ? § 3 5 0 — 0 — ’ ■ 1 1 II 3 2 a C; n n p* — O n p i p * * oc oo oo — o © d 00 oo 5 ? I ^ V “ r 1 “ o - T P M 0 0 0 0 « r t S ' P I * 7 "7 ^ T p ^ > © d © r - » c* «o r * » © o o o S O t T © r- o * — — do O oo m « r» * n © C l “* f * o -r — o o p m m — © P M P M p m m oo o <n ^ O pi m — oo *r o P M p m d -f P * M < N P M P M s s : x oo q ’ d d i PM PM PM 3 > > ■ ! » * S a l s ■ H S I » S 2 5 E •§ S ® s a 2 . < * s .5 — •j£ e -g £ S 3 ■ r 8 i l l * I O j , o v t ! ! < £* S i t O z <: . 5 2 * 3 o 9 S£< J > V S r t ^ ( A o s o u 3 a v _ . E 2 > E 2 > • * 5 o - * 1 r* £ ib ( b o E _ W « n * £ o T ± V ^ 2 > a 0. e o X & 5 " 2 , r - i _ y 11 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Table 2 (continued) Body Mass Index (kg/m1 ) 1 <20 21.9 1.0 80.6 1.0 } 88.7 } 1.0 } 70.1 } 1.0 20.0-21,9 22.1 0.92 (0.77-1.11) 82.3 1.07(0.92-1.24) 22.0-24.9 22.1 0.89(0.74-1.07) 81.6 1.01 (0.88-1.17) 90.5 1.18(0.96-1.46) 75.5 1.22 (0.98-1.52) 25.0-27.4 22.6 0.83 (0.68-1.02) 80.5 0.97(0.83-1.13) 91.0 1.33 (1.06-1.68) 76.7 1.27(0,99-1.63) 27.5+ 22.5 0.84(0.69-1.02) 77.3 0.80(0.69-0.93) 90.0 1.32 (1.06-1.66) 79.1 1.47(1.11-1.95) Trend test p=0.07 p<0.000l p=0.07 p=0.02 Alcohol Consumption in past yr (g/day)4 0 21.6 1.0 77.8 1.0 87.4 1.0 70.7 1.0 <5 24.5 1.11(0.95-1.29) 81.9 1.27(1.14-1.41) ) i ) } 1.14(0.92-1.40) 5-9 23.1 1.02(0.89-1.17) 79.8 1.15(1.04-1.26) } 91.5 } 1.48(1.23-1.77) } 75,7 10-14 22.9 1.07(0.92-1.26) 82.3 1.28(1.14-1.43) } 91.8 } 80.3 } 1.38(1.09-1.76) 15+ 24.8 1.05(0.88-1.26) 83.9 1.36(1.21-1.53) } 1.59(1.31-1.93) Trend test p=0.61 p<0.0001 p=0,0003 p=0,26 Vitamin User No 22.4 1.0 77.6 1.0 85.8 1.0 67.8 1.0 Yes 22.1 1.04(0.94-1.16) 82.4 1.21(1.13-1.30) 91.0 1.32(1.14-1.54) 76.8 1.32(1.11-1.57) Smoking History Nonsmokcr 21.4 1.0 80.0 1.0 89.9 1.0 73.2 1.0 Former smoker 25.3 1.00(0.87-1.14) 82.0 0.92 (0.84-1.00) 90.8 0.93(0.79-1.09) 79.3 1.11(0.91-1.34) Current smoker 25.4 1.13(0.87-1.47) 76.7 0.68(0.58-0.81) 79.4 0.45 (0.35-0.58) 63.0 0.54(0.35-0.85) Hormone Replacement Therapy Nonuser N/A N/A N/A N/A 81.6 1.0 64.2 1.0 Fast user N/A N/A N/A N/A 88.6 1.72(1.47-2.03) 76.2 1.62(1.36-1.93) Current user 95.9 4.75(3.94-5,73) 88.6 3.54(2.81-4.46) Logistic models for each age group include all variables listed in the table for that age group < £ are adjusted within age group for age; OR =odds ratio, C/= confidence interval. 1= For women in the 70-79 & 80+ age groups, all other races were compared to non-ilispanic whites. 2= Combines strenuous and moderate exercise activities 3= For women in the 70-79 & 80* age groups, body mass categories <20 and 20-21 .9 have been combined to form the baseline group. 4s For women in the 70-79 & 80+ age groups, alcohol consumption categories used were 0,1-9, & lO r- g/day. F J having a recent mammogram in this age group were recent exercise history, alcohol consumption and smoking history. Age 40-49 years. Among cohort members who were aged 40 to 49 years at baseline, a first-degree family history of breast cancer was an important predictor of recent screening mammography, while a family history of ovarian cancer was not. OC use, vitamin use, and alcohol consumption also predicted recent screening mammography. The odds of having had a recent mammogram increased with increasing duration of OC use (p t r e m i<0.000l). Vitamin users were 21% more likely to have had a mammogram in the past 2 years compared to nonusers. As observed for women aged 30-39 years at baseline, heavier women were less likely to have had a recent mammogram than thinner women (p tr c n d <0.0001). In contrast to women aged 30-39 years, we found that the relative odds of having had a recent mammogram increased with increasing alcohol consumption (p trend<0.000l) and that current smokers were less likely to have had one than nonsmokers. Women who participated in at least 4 hours per week of exercise in the past 3 years were also more likely to have had a recent mammogram. African American women had lesser relative odds o f having had a recent mammogram than non-Hispanic white women (OR=0.8). Menopausal status and parity were not important predictors of having a recent mammogram among women 40-49 years of age. Age 70-79 years. Race/ethnicity (comparing all other races/ethnicities to non- Hispanic whites) was not an important predictor of recent mammography among 13 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. women 70-79 years of age. Within this age group, use of hormone replacement therapy (HRT) appears to be the most important predictor of having a recent mammogram. The percentage of current users of HRT who have had a mammogram in the past 2 years was 95.9% and the relative odds of having had a recent mammogram was almost 5-fold greater among current HRT users than among women who had never used HRT. Family history of breast and ovarian cancer, parity, recent exercise history, alcohol consumption, BMI, vitamin use and smoking history also were predictors of having a recent mammogram. The magnitude of the OR for family history of breast cancer was smaller for this age group than for younger women (OR=T.8). Current smokers were least likely to have had a recent mammogram, whereas vitamin users and parous women were more likely have had one recently. The likelihood o f having had a prior mammogram recently increased with increasing alcohol consumption (p trcnd =0.0003). Women who exercise more than one hour per week, on average, over the past 3 years were also somewhat more likely than more sedentary women to have had a recent mammogram. Heavier women were more likely to have had a recent mammogram than thinner women, contrary to the pattern observed in the 30- 49 year olds. Age 80+ years. Within the oldest group of women (80+ years), race/ethnicity, family history of breast cancer, and parity were unrelated to mammography history. As observed for women 70 to 79 years o f age, HRT use was most strongly related to having had a recent screening mammogram. Women who currently used HRT 14 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. were 3.5 times as likely to have had a mammogram in the past 2 years, while past users were 1.6 times as likely to have had one compared to non-users of HRT. The odds of having a recent mammogram were substantially lower among current smokers than never smokers, whereas no difference between former smokers and never smokers was observed. Women who participated in recent exercise had greater odds of having had a recent mammogram. Heavier women were more likely than thinner women to have had a recent mammogram (p trend =0.02). Women who consumed at least 15 g/day of alcohol in the past year were also more likely to have had a recent mammogram than non-drinkers (OR=1.4). Vitamin users were 32% more likely to have had a mammogram in the past 2 years than nonusers. Table 3 reports predictors of ever having had a mammogram in the same four age groups. Our results were very similar to those for having had a mammogram in the past 2 years, with a few exceptions. Vitamin use was predictive of ever having had a mammogram in the 30-39 year olds. Recent exercise was no longer predictive for ever having had a mammogram in the 40-49 and 80+ year olds. Alcohol consumption was also not predictive of mammogram use in the 80+ year olds. Table 4 shows the AAIR for women diagnosed with a primary breast cancer during follow-up by recency o f mammography. The AAIR for women diagnosed with in situ breast cancer was highest among those having had a mammogram within the 2 years prior to entry into the study, 30.5 per 100,000 woman years of 15 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. TABLE 3. Predictors of ever having a mammogram among four age groups in the CTS ss& xg sti# 30-39 years 40-49 years 70-79 years 80+ years CHARACTERISTICS % screened OR (95% Cl) IS screened OR (95% Cl) % screened OR (95% Cl) % screened OR (95% Cl) Race1 • Non-llispanic whiles 35.0 1.0 93.7 1.0 96.3 1.0 90.8 1.0 African-American 36.8 1.11(0.80-1.53) 93.9 1.35(0.89-2.03) ~ v Hispanic whiles only 32.9 0.99(0.83-1.18) 90.1 0.84(0.67-1.05) Asian/Pacilic Islander 29.8 0.87(0.68-1.10) 91.9 1.02(0.78-1.33) | 96.6 1.64(1.05-2.56) | 88.6 0.87(0.59-1.28) Other/Mixed/None specified 36.3 1.06(0.81-1.39) 90.7 0.68(0.51-0.92) ) } Menopausal Status N/A N/A N/A Pre- 33.2 1.0 91.6 1.0 N/A Pcri-/Posi- 56.8 1.78(1.33-2.38) 96.8 1.08(0.86-1.35) N/A N/A N/A N/A Family History of Breast Cancer 96.2 1.0 90.4 1.0 No 32.1 1.0 93.0 1.0 Yes 67.2 5.90(4.92-7.08) 96.8 2.43(1.88-3.12) 98.1 2.20(1.44-3.37) 91.6 1.14 (0.80-1.62) Family History of Ovarian Cancer 96.4 1.0 90.5 1.0 No 34.6 1.0 92.7 1.0 Yes 43.8 1.61(1.11-2.33) 93.6 0.96(0.65-1.44) 98.2 2.00(0.87-4,60) 95.2 1.68 (0.72-3.91) Oral Contraceptive Use (yrs) N/A N/A 1) 35.1 1.0 87.5 1.0 N/A N/A <5 35.7 1.12 (0.98-1.28) 93.6 1.72 (1.48-2.00) N/A N/A N/A N/A 5-9 32.0 1.20(1.04-1.39) 95.0 2.03(1.71-2.40) N/A N/A N/A N/A IO C 36.5 1.29(1.11-1,51) 94.7 2.21 (1.82-2.68) N/A N/A N/A N/A Trend test p=0.0004 p<0.000l Full-term Pregnancy 89.5 1.0 No 31.4 1.0 92.7 1.0 94.9 1.0 Yes 36.7 0.91 (0.82-1.00) 93.6 1.02(0.89-1.16) 96.7 1.41(1.08-1.84) 91,2 1.00(0.78-1.29) Exercise History in Past 3 yrs (hrs/wk)' 1.0 <1 32.9 1.0 92.1 1.0 95.1 1.0 89.4 1-3 40.0 1.10(0.97-1.24) 94.5 1.12 (0.98-1.28) 97.0 1.39(1.05-1.84) 92.0 1.04 (0.79-1.37) 4+ 36.8 0.96(0.84-1.09) 94.3 0.98 (0.85-1.13) 96.9 1.31(0.99-1.72) 92.6 0.99(0.71-1.39) Trend lesl p=0.58 p=0.88 p=0.49 p-0,91 O N Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Table 3 (continued) Body Mass Index (kg/m1 ) 1 <20 33.4 1.0 92.5 1.0 } 95.9 } 1.0 } 87.3 } 1.0 20.0-21.9 35.2 1.03(0.87-1.23) 94.3 1.25(0.99-1.58) 22,0-24.9 34.3 0.93(0.79-1.10) 93.9 1.11(0.90-1.38) 96.6 1.17(0.84-1.64) 91.6 1.42(1.03-1.94) 25.0-27.4 34.9 0.87(0.72-1.05) 93.5 1.07(0.85-1.36) 96.6 1.24(0.86-1.78) 93.0 1.63 (1.12-2.37) 27.5+ 34.5 0.87 (0.73-1.04) 92.1 0.90 (0.72-1.12) 96.4 1.39(0.97-2.00) 94.2 1.94(1.24-3.04) Trend test p=0.0l p=0.0008 p=0.07 p=0,02 Alcohol Consumption in past yr (g/day)4 '0 35.0 1.0 91.9 1.0 95.2 1.0 88.5 1.0 <5 36.9 1.04 (0.91-1.20) 94.1 1.40(1.19-1.66) } 1.60(1.19-2.16) } 92.4 } 1.38(1.01-1.88) 5-9 34.9 0.95(0.84-1.08) 93.4 1.25 (1.08-1.46) } 97,1 10-14 35.9 1.03 (0.89-1,19) 94.4 1.40(1.17-1.69) > } 1.90(1.37-2.63) > } 1.22(0.86-1,74) 15+ 37.5 0.99(0.84-1.16) 95.0 1.48(1.22-1.80) } 97.4 } 92.4 Trend test p=0.83 p<0.000l p=0.0003 p=0.26 Vitamin User No 34.5 1.0 92.0 1.0 94.0 1.0 86.9 1.0 Yes 34.7 1.11(1.01-1.22) 94.3 1.26(1.12-1.41) 97.1 1.51(1.19-1.91) 92.2 1.42(1.11-1.81) Smoking History Nonsmoker 33,7 1.0 93.1 1.0 96.5 1.0 90.2 1.0 Former smoker 38.7 1.04(0.91-1.18) 94.3 0.94(0.82-1.08) 96.9 0.88(0.67-1.15) 93.3 1.18(0.87-1.60) Current smoker 37.5 1.02(0.79-1.31) 92.0 0.63 (0.49-0.82) 90.5 0.37(0.26-0.53) 78.2 0.35 (0.20-0.59) Hormone Replacement Therapy Nonuser N/A N/A N/A N/A 91.9 1.0 84.5 1.0 Past user N/A N/A N/A N/A 96.8 2.62(2.00-3.43) 92.9 2.12(1.63-2.77) Current user 99.2 9.33 (6.42-13.56) 98.1 7.40(4.58-11.95) Logistic models for each age group include all variables listed in the table for that age group < £ are adjusted within age group for age; Oh=odds ratio. ('!•= confidence interval. 1= For women in Ihc 70-79 & 80+ age groups, all olher races were compared io non-llispanic whiles. 2= Combines strenuous and moderate exercise activities 3s For women in the 70-79 & 80+ age groups, body moss categories <20 and 20-21.9 have been combined to form the baseline group. 4= For women in the 70-79 & 80+ age groups, alcohol consumption categories used were 0,1-9, & 10+ g/day. "4 TABLE 4. Average annual age-adjusted incidence rates among women in CTS diagnosed with a first primary breast cancer by recency of mammogram Type o f breast cancer Never/ >2 yrs ago AAIR (it cases) <= 2 yrs AAIR (# cases) In situ 14.1 (13) 30.5 (205) Invasive 121.1 (109) 158.4(1012) Localized 76.8 (64) 114.2 (735) Non-localized 44.3 (45) 44.2 (277) AAfR= Age-adjusted incidence rate p er i 00,000 woman years. Rates standardized to the 1970 US population. * We excluded women who did not indicate the recency o f their lost mammogram follow-up. The same appeared to be true for invasive breast cancer with an AAIR of 158.4 per 100,000 woman years. Women who were screened within the 2 years prior to study entry were 2.2 times as likely to be diagnosed with in situ breast cancer in the subsequent 3 years compared to those never screened or screened more 2 years ago (rate ratio=30.5/14.1). They were only 1.3 times as likely to be diagnosed with invasive breast cancer if screened within the 2 years prior to study entry compared to those never screened or screened more than 2 years ago (rate ratio=l 58.4/121.1). 18 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 4: DISCUSSION The purposes of this study were to estimate the frequencies of mammographic screening by age and race/ethnicity among women with uniform access to health care and to utilize information on mammographic screening to understand better the different factors that predict mammography use in age groups were controversy exists as to the advisability of obtaining screening mammograms. In 1998, the National Center for Health Statistics reported that 66.9% of women over 40 years had had a mammogram within the past 2 years. They also reported Hispanic women over 40 as having the lowest screening rates within the past 2 years (60.2%) and White, Non-Hispanic women over 40 as having the highest screening rates within the past 2 years (68%).1 9 In contrast, in this cohort of women with reliable access to health care 86% of them 40 years or older reported having had a mammogram in the past 2 years. In addition, the variation in screening rates by race/ethnicity was not substantial. Although previous studies have reported predictors of mammography, many of them have not incorporated such a large number of lifestyle and reproductive factors into their multivariate models. We evaluated a number of factors to determine which would best predict mammography use in the youngest and oldest age groups. Our findings suggest that the key predictors of having had a recent 19 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mammogram across the different age groups studied are a positive family history of breast cancer, having used hormones (OCs among younger women and HRT among older women), regular vitamin use, recent exercise, and not smoking. The key predictors o f ever having a mammogram are similar to those for recent mammography use, except that current exercise becomes less predictive. The most important predictor of recent mammography in younger women, 30-49 years old, is having a family history of breast cancer. This means that women who are getting mammograms before the recommended age are doing so primarily because of concern that they have high risk of breast cancer. Although family history is not the most important predictor in 70-79 year old women, the odds of having a screening mammogram are almost 2-fold greater for women with a family history of breast cancer than for those without a family history. These results are consistent with a prior study that found an increased likelihood for mammography use in women with a family history of breast cancer.1 4 Our results also show that women who take hormones (OC/HRT) have greater odds of having had a recent mammogram. The odds o f having had a recent screening mammogram increased significantly with increasing duration of OC use among the young women. The odds o f having a recent mammogram were approximately 3 to 5 fold-greater among the older women who were using HRT at baseline. These findings agree with a previous study that reported that women on HRT were 2.5 times as likely to have had a mammogram in the last 2 years as women who were not using HRT.1 3 Therefore, it appears that women who are 20 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. visiting the doctor regularly for hormone prescriptions are being advised to obtain a mammogram. Women who consume vitamins regularly were 21% to 32% more likely to have had a mammogram in the past 2 years. Women who recently participated in some exercise were also 12% to 52% more likely to have had a recent mammogram. Those who were current smokers were 32% to 55% less likely to have had a recent mammogram. These findings suggest that women with healthy lifestyle practices are more likely to engage in screening. In our study, we also report AAIR of women diagnosed with their first primary breast cancer during the follow-up period. We found that women were 2.1 times as likely to be diagnosed with in situ breast cancer if screened within the 2 years prior to study entry than women who were never screened or screened more than 2 years previously. The majority of in situ cancers are ductal carcinomas, which are diagnosed almost exclusively by mammography. A potential limitation of this study is its temporal ambiguity. Since this was a cross-sectional study, some predictors of mammography {i.e. alcohol intake, vitamin use, exercise history, BMI, smoking status) and mammogram screening patterns were obtained concurrently making it uncertain in some instances to know which occurred first. Another possible limitation o f this study is related to the high screening rates found in our CTS cohort. It is possible that the cohort represents a biased group of teachers, those who were motivated to join the cohort and may have been concerned about breast cancer and thus obtained mammograms. 21 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. However, the predictors found to be the most significant were consistent with past studies.1 4 ' 1 6 Self-reported data may also be a potential disadvantage to the study. However, prior studies have found that self-reported mammography proved accurate and reliable when verified with medical records.1 2 ,2 0 Advantages of this study include its large sample size allowing us to have adequate power in our analysis. We were able to evaluate factors in the CTS ranging from demographics, health status, lifestyle factors, reproductive factors, to family history of cancers. This is the only study to incorporate a large number of lifestyle and reproductive factors into a multivariate context in evaluating predictors of mammography. This study also allowed us to focus on other factors of interest besides access to health care. As a result, it provided us with insight into the types of women getting screened for breast cancer in age groups seldom studied. 22 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. BIBLIOGRAPHY 1. National Cancer Institute. What You Need to Know About Breast Cancer (CancerNet). December 12, 2000. Available at: http://www.cancemet.nci. nih. sov. Accessed November 20, 2001. 2. Van Dijck J, Verbeek A, Beex L, Hendriks J, Holland R, Mravunac M, Straatman H, Were J. Mammographic Screening After the Age of 65 Years: Evidence for a Reduction in Breast Cancer Mortality. Cancer 1996; 66:727-731. 3. American Cancer Society. Mammography and Other Breast Imaging Procedures. September 28, 1999. Available at: http://www.cancer.ore. 4. Leitch AM. Controversies in Breast Cancer Screening. Cancer (Supplement) 1995; 76(10): 2064-2069. 5. Berry DA. Benefits and Risks of Screening Mammography for Women in their Forties: a Statistical Appraisal. Journal o f the National Cancer Institute 1998; 90(19): 1431-1439. 6. Esserman L, Kerlikowske K. Should We Recommend Screening Mammography for Women Aged 40-49? Oncology 1996; 10(3): 357-396. 7. Kerlikowske K. Efficacy of Screening Mammography Among Women Aged 40 to 49 Years and 50 to 69 Years: Comparison of Relative and Absolute Benefit. Journal o f the National Cancer Institute Monographs 1997; 22:79-86. 8. Kerlikowske K, Salzmann P, Phillips KA, Cauley JA, Cummings SR. Continuing Screening Mammography in Women Aged 70 to 79 Years: Impact on Life Expectancy and Cost-effectiveness. JAMA 1999; 282(22): 2156-2163. 9. Smith-Bindman R, Kerlikowske K. Is There a Downside to Elderly Women Undergoing Screening Mammography? Journal o f the National Cancer Institute 1998; 90(18): 1322-1323. 10. Smith-Bindman R, Kerlikowske K, Gebretsadik T, Newman J. Is Screening Mammography Effective in Elderly Women? Am J Med 2000; 108(2): 112-119. 11. Bums RB, McCarthy EP, Freund KM, Marwill SL, Shwartz M, Ash A, Moskowitz MA. Variability in Mammography Use Among Older Women. JAGS 1996; 44: 922-926. 23 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12. Cummings DM, Whetstone L, Shende A, Weismiller D. Predictors of Screening Mammography: Implications for Office Practice. Arch Fam Med 2000; 9(9): 870-875. 13. Mayer-Oakes SA, Atchison KA, Matthias RE, De Jong FJ, Lubben J, Schweitzer SO. Mammography Use in Older Women With Regular Physicians: What are the Predictors? Am JP rev Med 1996; 12(1): 44-50. 14. Lee JR, Vogel VG. Who Uses Screening Mammography Regularly? Cancer Epidemiology, Biomarkers & Prevention 1995; 4: 901-906. 15. Ruffin MT, Gorenflo DW, Woodman B. Predictors of Screening for Breast, Cervical, Colorectal, and Prostatic Cancer Among Community-Based Primary Care Practices. J Am Board Fam Pract 2000; 13(1): 1-12. 16. Clark MA, Rakowski W, Ehrich B. Breast and Cervical Cancer Screening: Associations with Personal, Spouse’s, and Combined Smoking Status. Cancer Epidemiology, Biomarkers & Prevention 2000; 9: 513-516. 17. Wee CC, McCarthy EP, Davis RB, Phillips RS. Screening for Cervical and Breast Cancer: Is Obesity an Unrecognized Barrier to Preventive Care? Ann Intern Med 2000; 132(9): 697-704. 18. Phillips KA, Kerlikowske K, Baker LC, Chang SW, Brown ML. Factors Associated with Women’s Adherence to Mammography Screening Guidelines. Health Services Research 1998; 33(1): 29-53. 19. Center for Disease Control. Fast stats A to Z (National Center for Health Statistics). July 10,2001. Available at: http://www.cdc.20v/nchs/fastats/mamm02ram.htm. Accessed November 20, 2001. 20. King E, Rimer BK, Trock B, Balshem A, Engstrom P. How reliable are mammography self reports? Am J Public Health 1990; 80:1386-1388. 21. Blustein J. Medicare Coverage, Supplemental Insurance, and the Use of Mammography by Older Women. M Engl J Med 1995; 332(17): 1138-1143. 22. Bums RB, McCarthy EP, Freund K, Marwill SL, Shwartz M, Ash A, Moskowitz MA. Black Women Receive Less Mammography Even with Similar Use of Primary Care. Ann Intern Med 1996; 125(3): 173-182. 23. O’Malley MS, Earp JL, Hawley ST, Schell MJ, Mathews HF, Mitchell J. The Association of Race/Ethnicity, Socioeconomic Status, and Physician 24 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Recommendation for Mammography: Who Gets the Message About Breast Cancer Screening? Am J Public Health 2001; 91(1): 49-54. 24. Parker JD, Sabogal F, Gebretsadik T. Relationship Between Earlier and Later Mammography Screening-Califomia Medicare, 1992 Through 1994. West J Med 1999; 170: 25-27. 25. Sherman JJ, Abel E, Tavakoli A. What’s Happening: Demographic Predictors of Clinical Breast Examination, Mammography, and Pap Test Screening Among Older Women. J Amer Acad Nurse Practitioners 1996; 8(5): 231-236. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Predictive value of CT coronary artery calcium scanning for coronary heart disease in asymptomatic subjects with diabetes mellitus

PDF

Recreational physical activity and risk of breast cancer: The California Teachers Study

PDF

The influence of family structures on adolescent smoking among multicultural adolescents in Hawaii

PDF

The effects of tobacco smoke on respiratory symptoms among young adults in the Children's Health Study II

PDF

Validation of serum cotinine as a biomarker of environmental tobacco smoke exposure: Validation with self-report and association with subclinical atherosclerosis in non-smokers

PDF

Post-intensive care unit mechanical ventilation: Relationship of infections to outcomes of weaning from prolonged mechanical ventilation

PDF

The association between recreational physical activity and mammographic density

PDF

Progression of carotid intima-media thickness and plasma antioxidants: The Los Angeles Atherosclerosis Study

PDF

P53 and bladder cancer outcome: A combined analysis from the Keck School of Medicine

PDF

Selective laser trabeculoplasty for the treatment of glaucoma

PDF

Polymorphisms in genes involved in steroid hormone metabolism and mammographic density changes in women randomized to menopausal estrogen and progesterone therapy

PDF

Metabolic effects of magnesium supplementation in women with a history of gestational diabetes mellitus

PDF

Cigarettes and alcohol in relation to colorectal cancer within the Singapore Chinese Health Study

PDF

A case-control study of passive smoking and bladder cancer risk in Los Angeles

PDF

Randomized comparison of oral mercaptopurine versus oral thioguanine for standard risk acute lymphoblastic leukemia

PDF

Comparisons of metabolic factors among gestational diabetes mellitus probands, siblings and cousins

PDF

Agreement between administrative and clinical data for term newborns with congenital malformations

PDF

Risk factors for diabetic retinopathy in Latinos: Los Angeles Latino eye study

PDF

Lymphedema: Impact on breast cancer survivors

PDF

Association between body mass and benign prostatic hyperplasia in Hispanics: Role of steroid 5-alpha reductase type 2 (SRD5A2) gene

Asset Metadata

Creator Villaluna, Doojduen (author)

Core Title Predictors of mammography use among California teachers

Degree Master of Science

Degree Program Applied Biostatistics and Epidemiology

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

Tag biology, biostatistics,health sciences, public health,health sciences, radiology,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-296484

Unique identifier UC11337052

Identifier 1411812.pdf (filename),usctheses-c16-296484 (legacy record id)

Legacy Identifier 1411812.pdf

Dmrecord 296484

Document Type Thesis

Rights Villaluna, Doojduen

Type texts

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

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

Repository Name University of Southern California Digital Library

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