Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Do reproductive factors have the same effect on breast cancer risk in African-American and White women?
(USC Thesis Other)
Do reproductive factors have the same effect on breast cancer risk in African-American and White women?
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UM I 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 will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and continuing
from left to right in equal sections with small overlaps.
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.
DO REPRODUCTIVE FACTORS HAVE THE SAME EFFECT ON BREAST
CANCER RISK IN AFRICAN-AMERICAN AND W HITE WOMEN?
b y
Yaping W ang
A Thesis Presented to the
FACULTY O F TH E GRADUATE SCHOOL
UNIVERSITY O F SOUTHERN CALIFORNIA
In P artial Fulfillment of the
Requirem ents for the Degree
M ASTER OF SCIENCE
(APPLIED BIOM ETRY AND EPIDEM IOLOGY)
December 2001
Copyright 2001 Yaping Wang
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 1411040
Copyright 2001 by
Wang, Yaping
All rights reserved.
___ ®
UMI
UMI Microform 1411040
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
t h e g r a d u a t e s c h o o l
UNIVERSITY RARK
LOS ANGELES. CALIFORNIA S 0 0 0 7
This thesis, written by
under the direction of h £ J z 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
W '
D tta
THESJ&£pM M ITTEE
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGMENTS
I am very grateful to Dr. Giske Ursin for her valuable guidance, suggestions,
comments, and editing throughout the course o f my research and in the preparation o f
this manuscript. I would also like to express my deeply gratitude to my thesis
committee, for without their helpful guidance I could not have completed this portion
o f my education.
Giske Ursin, MD, PhD
Leslie Bernstein, PhD
James Gauderman, PhD
And, of course, my deepest thanks go to my family for their support. Thank you.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ii
LIST OF TABLES iv
ABSTRACT v
INTRODUCTION 1
MATERIAL AND METHODS 1
Study subjects 2
Data collection 2
Data analysis 3
RESULTS 6
Demographic characteristics 7
Logistic regression analysis for reproductive variables 8
Reproductive variables by age group 13
DISCUSSION 19
REFERENCES 29
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Table 1. Demographic characteristics in white and African-American women. 7
Table 2. Odds ratios (ORs) and 95% confidence intervals (CIs) for breast 9-11
cancer associated with various reproductive variables.
Table 3. Characteristics o f reproductive variables in ever pregnant white and 15
African-American women according to age group.
Table 4. Odds ratios (ORs) and 95% confidence intervals (CIs) of breast 16
cancer associated with number of full-term pregnancies according to age
group.
Table 5. Odds ratios (ORs) and 95% confidence intervals (CIs) of breast 18
cancer associated with breastfeeding in ever pregnant women according to
age group.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
Data for this thesis came from a population-based case-control study o f 4575 newly
diagnosed female breast cancer patients and 4682 controls identified by random digit
dialing, lived in the same region as the patients, and frequency matched to patients
on age and ethnicity. Having at least one full-term pregnancy lowered breast cancer
risk 33% for white women and 5% for African-Americans. Breastfeeding was
associated with a 20% reduced risk in both ethnic groups. An inverse dose-response
relationship was observed between breast cancer risk and duration of lactation in
both ethnic groups. Overall, the magnitude of the associations observed between
breast cancer risk and pregnancy history was lower in African-Americans than
whites, but these differences were not statistically significant. Effects o f lactation on
breast cancer risk were similar in the two ethnic groups.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INTRODUCTION:
Overall incidence o f breast cancer is lower in African-American women than in
white women. The age-adjusted incidence rates from 1985 to 1996 increased from
92.7 to 100.3 per 100,000 African-American women and from 107.3 to 113.3 per
100,000 white women (National Institutes o f Health, 2000). Reports documenting
incidence rates during the 1970s showed that African-American women under age 40
had higher incidence rates o f breast cancer than white women, whereas white women
40 years or older had lower incidence rates (Gray et al., 1980). This crossover
phenomenon has remained stable over time, except that the age at which the
crossover occurred has increased to age 45 years (Trock, 1996). Few epidemiologic
studies have examined why the overall incidence rates are higher in white than in
African-American women and why the ethnic crossover in incidence rates occurs. In
this study we examined whether white and African-American women differ with
regard to the effects of reproductive variables on breast cancer risk.
MATERIAL AND METHODS:
We used data from the Women’s Contraceptive and Reproductive Experiences
(CARE) Study, a population-based case-control study, designed to examine risk
factors for breast cancer among white and African-American women in five U.S.
regions (Atlanta, Seattle, Detroit, Philadelphia, and Los Angeles) (Marchbanks et al.,
l
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2001 (in press)). Cases were women newly diagnosed with pathologically confirmed
invasive breast cancer from July 1994 through April 1998. Controls were identified
through random digit dialing (RDD). Both cases and controls were randomly
selected from eligible women within the same geographic region. Sampling
fractions for cases were established to provide an equal distribution o f cases in each
5 year age group between 35 and 64 years, when possible. Controls were frequency
matched to cases by strata o f study site, race, and age. Eligible African-American or
white cases and controls were US bom, resided in one o f the above five regions,
were aged 35-64 years at time o f diagnosis/RDD contact, were a member of a
household with a telephone at time of diagnosis/RDD contact, and spoke English
well enough to be interviewed. An estimated 82% of households were screened
successfully through RDD. Interviews were completed for 4575 cases (2953 whites
and 1622 African-Americans) and 4682 controls (3021 whites and 1661 African-
Americans). Overall interview response rates were 76.5% and 78.6% for cases and
controls, respectively, with lower rates for African-American women and older
women.
In-person interviews were conducted from August 1994 through December 1998.
Information on demographic characteristics, reproductive history including
breastfeeding and other known or suspected breast cancer risk factors was obtained
through an in-person interview using a structured questionnaire and a life event
calendar. A reference date was created for each study subject, which was the date o f
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
diagnosis (month and year) for cases and the date on which the screening
questionnaire (RDD) was completed for controls. A detailed description o f the study
design and conduct is provided by Marchbanks et al. (Marchbanks et al., 2001 (in
press)).
The reproductive variables considered in this analysis are the following: gravidity
(ever pregnant vs. never pregnant); parity (no full-term pregnancy vs. full-term
pregnancy); number o f pregnancies; number o f full-term pregnancies; age at first
full-term pregnancy; and years since last full-term pregnancy. Any pregnancy was
considered to be a full-term pregnancy if the gestational length o f that pregnancy was
at least 26 weeks. Age at first full-term pregnancy was the age o f the women at the
ending date o f her pregnancy lasting more than 26 weeks. Detailed histories of
breastfeeding were collected; we examined whether women ever breastfed, whether
they breastfed for at least 2 weeks, and the total months of breastfeeding across all
pregnancies.
O f4575 cases and 4682 controls, six subjects (1 African-American case and 5 white
controls) were excluded from these analyses since it was unknown whether they had
ever had a pregnancy. Four subjects (3 white controls, 1 African-American control)
who were currently pregnant with their first pregnancy at reference date were
excluded from the analyses o f full-term pregnancy. One African-American case
refused to answer questions about her last three pregnancies and two more subjects
i
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(both African-Americans, one case and one control) reported unknown gestational
lengths for at least one pregnancy, and these women were also excluded from
analyses o f the number of full-term pregnancies.
Odds ratios (ORs) and 95% confidence intervals (CIs) for the OR were estimated
using unconditional logistic regression methods (Breslow and Day, 1980).
Multivariate logistic regression analysis method was used to simultaneously control
for several possible confounding factors and for the design variables, age, ethnicity,
and study site. We included the following variables in all models: age (35-39,40-44,
45-49, 50-54, 55-59,60-64 years), ethnicity (white and African-American), family
history of breast cancer (classified as first degree (mother or sister), second degree
(grandmother or aunt), unknown, and no family history); age at menarche (<12, 12-
13, >13 years), body mass index (BMI) at age 18 (<=19,20, 21-22, >22 kg/m2),
study site (5 study sites), education (< high school, high school, technical school,
some college, college graduate, and post graduate), and poverty index (1,2, 3,4, and
5). The poverty index was based on classifying the woman’s household income for
the calendar year preceding the year of interview as a multiple o f the federal poverty
guideline cutoff for households o f the same size as the woman’s household. For
instance, for a woman living in a household of 4 persons, a poverty index of 5 means
that she lived in a household with a household income that was 5 times the poverty
threshold for households of 4 persons (Dalker and Proctor, 2000). Depending on the
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
variables of interest, the multivariate models also included the possible confounding
variables of age at first full-term pregnancy (<=19, 20-24, 25-29, >=30 years),
number of full-term pregnancies (1 ,2 ,3 ,4 , 5+), and total duration of breastfeeding
(0, 1-6, 7-23, >=24 months). All confounding variables were included as categorical
variables in the model. When testing for linear trend, we analyzed the variable of
interest as a continuous variable in the model. Women who were never pregnant and
only ever non full-term pregnant were assigned as reference value for trend test and
effect modification test. All p-values are based on a 2-sided alternative hypothesis.
To test for possible effect modification by age on the relationship between breast
cancer risk and pregnancy-related factors, we further analyzed data in three
subgroups: age <40 years, 40-55 years, and >55 years. These represent,
approximately, premenopausal, perimenopausal, and postmenopausal women,
respectively.
To examine whether ethnicity was an effect modifier o f the association between
breast cancer risk and another reproductive variable, we used a Likelihood Ratio Test
(LRT) to compare a restricted model (fitting only one pregnancy variable) to that of
an expanded model (fitting two ethnicity specific pregnancy variables). For instance,
to test whether ethnic group modified the effect o f number o f full-term pregnancies
(FTP) on breast cancer risk, we created two new variables, white FTP and African-
American FTP. In white women the white FTP variable was equal to total number
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of FTP, while African-American women were assigned the reference value. The
African-American FTP variable was equal to total number of FTP in African-
American women and white women were assigned the reference value o f this
variable. The adjusted models with only the main FTP variable and with the two
ethnicity specific FTP variables were compared. The LRT, calculated as twice the
difference in log-likelihoods (-21n(L)) between the two models, is treated as a Chi-
square statistic with degrees o f freedom equal to the difference in the number of
parameters being estimated under the two models. Thus, we are able to determine
whether the model with two terms statistically provides a better fit to the data than
the model with one term. A p-value o f less than 0.05 indicated effect modification
on a multiplicative odds ratio scale. All analyses were performed using EPILOG
(Software, 1999).
RESULTS:
About 65% of the participants were white (N=5974) and 35% were African-
American (N=3283) in our study (Table 1). Age distributions were similar among
cases and controls in both white and African-American women. Cases and controls
were comparable within ethnic groups in five study sites respectively, although in
Seattle, the majority of cases and controls were white (about 94% white and 6%
African-American). Cases and controls had similar distributions for education levels
and poverty index in both white and African-American women respectively.
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 1. Demographic characteristics in white and African-American women
Whites
Cases Controls
African-Americans
Cases Controls
No. % No. % No. % No. %
Age (year)
35-39 465 15.7 450 14.9 223 13.7 216 13.0
40-44 469 15.9 526 17.4 289 17.8 306 18.4
4549 484 16.4 542 17.9 297 18.3 315 19.0
50-54 540 18.3 523 17.3 305 18.8 302 18.2
55-59 498 16.9 515 17.0 271 16.7 286 17.2
60-64 497 16.8 465 15.4 237 14.6 236 14.2
P value * 0.18 0.96
Site
Atlanta 559 63.5 552 61.7 322 36.5 343 38.3
Seattle 1004 94.2 963 94.7 62 5.8 54 5.3
Detroit 401 59.1 446 57.3 278 40.9 333 42.7
Philadelphia 318 45.0 345 46.9 389 55.0 391 53.1
Los Angeles 671 54.0 715 57.0 571 46.0 540 43.0
P value 0.29 0.16
Education
< High school 164 5.6 188 6.2 235 14.5 256 15.4
High school 788 26.7 837 27.7 547 33.7 513 30.9
Tech school 139 4.7 127 4.2 78 4.8 99 6.0
Some college 813 27.5 800 26.5 453 27.9 469 28.2
College graduate 569 19.3 588 19.5 159 9.8 182 11.0
Post graduate 480 16.3 481 15.9 149 9.2 142 8.5
P value 0.65 0.31
Poverty index
1 357 12.5 367 12.5 596 37.9 649 40.1
2 442 15.5 489 16.6 329 20.9 331 20.4
3 709 24.9 731 24.9 333 21.2 306 18.9
4 722 25.3 765 26.0 180 11.4 211 13.0
5 620 21.8 587 20.0 135 8.6 123 7.6
P value 0.46 0.22
Family history
1st degree 549 18.6 318 10.5 253 15.6 164 9.9
2nd degree 656 22.2 566 18.7 227 14.0 183 11.0
Unknown 93 3.1 121 4.0 98 6.0 113 6.8
No 1655 56.0 2016 66.7 1044 64.4 1201 72.3
P value <0.0001 <0.0001
BMi at 18 years
<=19 992 33.7 937 31.1 497 31.0 510 30.9
20 597 20.3 603 20.0 248 15.5 220 13.3
21-22 813 27.6 842 27.9 417 26.0 433 26.2
>22 545 18.5 631 20.9 443 27.6 490 29.6
P value 0.054 0.28
* P value: for Chi-square test.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
However, white women tended to have higher levels o f education and poverty index
than African-Americans. Cases more frequently reported a first degree or a second
degree family history o f breast cancer than controls in both ethnic groups (18.6% and
22.2% vs. 10.5% and 18.7% for cases vs. controls o f first degree and second degree
family history o f breast cancer in whites; 15.6% and 14.0% vs. 9.9% and 11.0% for
cases vs. controls in African-Americans). A greater proportion o f African-American
cases (27.6%) and controls (29.6%) reported a BMI > 22 kg/m2 at age 18 years than
did whites (18.5% and 20.9% for cases and controls) (P=0.034). Since these
characteristics were also related to reproductive variables, we controlled for them in
all subsequent analyses.
Table 2 shows the ORs and 95% CIs for breast cancer associated with various
reproductive variables. Among white women, those who had been pregnant had
lower breast cancer risk than those who had not, and this reduction in risk was
observed both for women who had a full-term pregnancy (OR: 0.67, 95% Cl: 0.54-
0.84) and those who had only had other pregnancy outcomes (OR: 0.75, 95% Cl:
0.59-0.96). No similar reduction in risk was observed among African-American
women, although risk was reduced among those reporting only pregnancy outcomes
that were not more than 26 weeks gestation. This latter finding was not statistically
significant, however. Moreover, both an increasing number of pregnancies and an
increasing number o f full-term pregnancies significantly decreased risk for breast
8
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. O dds ratios (ORs) and 95% confidence intervals (CIs) for breast cancer
________________ associated with various reproductive variables________________
case
All Women
control OR 95% Cl case control
Whites
OR 95% Cl case
African-Americans
control OR 95% Cl
Ever pregnant
No 591 483 1.00 464 365 1.00 127 118 1.00
Yes 3983 4194 0.77 0.66-0.90 2469 2651 0.71 0.58-0.86 1494 1543 0.91 0.68-1.22
P value for effect modification test 0.31
Ever full-term pregnancy
Never pregnant 591 483 1.00 464 365 1.00 127 118 1.00
Pregnant but no full-term pregnancy 299 322 0.75 0.61-0.92 202 210 0.75 0.59-0.96 97 112 0.79 0.54-1.16
Ever full-term pregnancy 3684 3868 0.77 0.65-0.91 2287 2438 0.67 0.54-0.84 1397 1430 0.95 0.70-1.28
P value for effect modification test 0.24
Number of pregnancies
Never pregnant 591 483 1.00 464 365 1.00 127 118 1.00
1 560 539 0.82 0.68-0.99 356 351 0.75 0.60-0.94 204 188 0.98 0.69-1.39
2 1046 1003 0.82 0.68-0.98 746 710 0.75 0.60-0.94 300 293 0.94 0.67-1.30
3 929 1022 0.73 0.61-0.87 611 692 0.65 0.52-0.81 318 330 0.89 0.64-1.23
4 658 690 0.78 0.65-0.94 402 428 0.71 0.56-0.89 256 262 0.93 0.66-1.30
5 365 405 0.75 0.60-0.93 189 226 0.64 0.48-0.85 176 179 0.96 0.67-1.37
6+ 424 535 0.67 0.54-0.82 185 244 0.60 0.45-0.79 239 291 0.79 0.56-1.12
Trend P 0.0018 0.0010 0.27
P value for effect modification test 0.20
Number of full-term pregnancies
Never pregnant 591 483 1.00 464 365 1.00 127 118 1.00
1 770 717 0.88 0.71-1.08 435 434 0.72 0.56-0.94 335 283 1.16 0.83-1.64
2 1371 1355 0.84 0.70-1.01 969 933 0.75 0.59-0.95 402 422 0.93 0.67-1.29
3 842 905 0.79 0.65-0.96 548 607 0.67 0.52-0.86 294 298 1.00 0.72-1.40
4 381 443 0.74 0.59-0.92 212 268 0.61 0.46-0.81 169 175 0.96 0.67-1.39
5+ 318 447 0.58 0.46-0.73 123 196 0.46 0.33-0.64 195 251 0.73 0.51-1.05
Pregnant but no full-term pregnancy 299 322 0.76 0.62-0.93 202 210 0.76 0.59-0.97 97 112 0.79 0.54-1.16
Trend P 0.0076 0.0001 0.19
P value for effect modification test 0.13
Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 2. (continued)
case
All Women
control OR 95% Cl case
Whites
control OR 95% Cl case
African-Americans
control OR 95% Cl
Age at first full-term pregnancy
<=19 1046 1191 1.00 389 509 1.00 657 682 1.00
20-24 1368 1461 1.04 0.92-1.18 908 984 1.16 0.97-1.38 460 477 0.96 0.80-1.16
25-29 777 718 1.19 1.02-1.39 597 553 1.37 1.12-1.68 180 165 1.00 0.77-1.31
30+ 492 497 1.05 0.88-1.27 393 392 1.25 0.99-1.59 99 105 0.80 0.57-1.12
Pregnant but no full-term pregnancy 299 321 0.86 0.69-1.08 202 209 1.05 0.78-1.41 97 112 0.68 0.48-0.97
Never pregnant 591 483 1.14 0.93-1.40 464 365 1.38 1.06-1.80 127 118 0.86 0.61-1.21
Trend P 0.19 0.016 0.33
P value for effect modification test 0.096
Yrs since last full-term pregnancy
21+ 2090 2099 1.00 1278 1312 1.00 812 787 1.00
11-20 903 1033 0.94 0.82-1.08 524 622 1.03 0.85-1.23 379 411 0.85 0.69-1.05
7-10 310 295 1.13 0.92-1.40 222 204 1.34 1.03-1.74 88 91 0.88 0.61-1.27
4-6 191 208 0.99 0.77-1.28 138 133 1.22 0.89-1.67 53 75 0.66 0.43-1.01
0-3 174 218 0.83 0.64-1.08 119 160 0.86 0.62-1.18 55 58 0.83 0.53-1.32
Pregnant but no full-term pregnancy 299 321 0.79 0.64-0.97 202 209 0.89 0.68-1.17 97 112 0.67 0.47-0.94
Never pregnant 591 483 1.04 0.87-1.24 464 365 1.16 0.93-1.46 127 118 0.85 0.61-1.17
Trend P 0.59 0.77 0.16
P value for effect modification test 0.61
Ever breastfeeding
No 1951 1855 1.00 1072 1028 1.00 879 827 1.00
Yes 2031 2334 0.81 0.73-0.89 1417 1619 0.82 0.72-0.94 614 715 0.78 0.67-0.92
Never pregnant 591 483 1.32 1.08-1.62 464 365 1.32 1.03-1.68 127 118 1.26 0.86-1.85
P value for effect modification test 0.81
O
Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 2. (continued)
case
All Women
control OR 95% Cl case
Whites
control OR 95% Cl case
African-Americans
control OR 95% Cl
Ever breastfeeding at least 2wks
Never breastfeeding 1951 1855 1.00 1072 1028 1.00 879 827 1.00
Ever breastfeeding < 2wks 165 175 0.90 0.71-1.13 91 90 0.99 0.72-1.36 74 85 0.80 0.57-1.13
Ever breastfeeding 2+ wks 1B66 2159 0.80 0.72-0.89 1326 1529 0.81 0.71-0.93 540 630 0.78 0.66-0.92
Never pregnant 591 483 1.32 1.08-1.62 464 365 1.32 1.03-1.69 127 118 1.26 0.86-1.85
P value for effect modification test 0.57
Total breastfeeding (month)
0 1951 1855 1.00 1072 1028 1.00 879 827 1.00
< 1 277 311 0.86 0.72-1.03 167 170 0.95 0.75-1.21 110 135 0.75 0.56-0.99
1-6 821 902 0.82 0.72-0.93 576 629 0.82 0.70-0.96 245 273 0.81 0.66-1.00
7-23 663 776 0.79 0.69-0.91 493 569 0.80 0.67-0.95 170 207 0.77 0.61-0.98
24+ 267 349 0.73 0.60-0.89 181 248 0.73 0.57-0.93 86 101 0.76 0.54-1.06
Never pregnant 591 483 1.32 1.08-1.62 464 365 1.32 1.03-1.69 127 118 1.26 0.86-1.85
Trend P <0.0001 0.0036 0.0062
P value for effect modification test 0.70
Note 1: ORs are adjusted for categorical variables of age, ethnicity, family history of breast cancer, age at menarche, age at first full-
term pregnancy, number of full-term pregnancies, duration of breastfeeding, BMI, education, poverty index, and study
Note 2: Never pregnant and pregnant but no full-term pregnancy have been assigned the reference value for trend test and 9ffect
modification test.
cancer in white women (trend P = 0.0010 for number o f pregnancies and 0.0001 for
number o f full-term pregnancies). Among whites, women who had 5 or more full-
term pregnancies had less than half the risk o f breast cancer compared with women
who had never been pregnant. In African-American women, there was no
statistically significant protective effect o f number o f pregnancies or number of full-
term pregnancies. However, the test for effect modification by ethnicity did not
reach statistical significance.
Late age at first full-term pregnancy was associated with increasing breast cancer
risk in white women (trend P = 0.016). White women whose first full-term
pregnancy occurred at age 25-29 had a 37% increased risk in breast cancer as
compared with those who had their first full-term pregnancy at age 19 or younger
(95% Cl: 1.12-1.68). However, we did not observe this effect in African-American
women. Nevertheless, only marginal effect modification by ethnic group was noted
(P = 0.096).
There was no evidence that a woman who had had a recent full-term pregnancy was
at increased risk of breast cancer as compared with women who had their last full-
term pregnancy over 20 years ago (OR: 0.83,95% Cl: 0.64-1.08 for women who had
their last full-term pregnancy within 3 years; OR: 0.99, 95% Cl: 0.77-1.28 for
women who had their last full-term pregnancy within 4-6 years). However, we
observed an increased risk o f breast cancer among white women who were 7-10
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
years since last full-term pregnancy as compared with white women who had it over
20 years ago. There was no significant trend in either ethnic group (trend P = 0.77
and 0.16 in white and African-American women, respectively).
Breastfeeding was associated with a reduced risk of breast cancer in both ethnic
groups. Women who ever breastfed had about 20% lower breast cancer risk than
women who never breastfed. In addition, the longer a woman breastfed in her
lifetime, the lower was her breast cancer risk (trend P < 0.0001 for whites and
African- Americans combined). The statistically significant dose-response
relationship between breast cancer and duration o f lactation was observed for both
white and African-American women. White and African American women who
experienced at least 24 months o f breastfeeding were, respectively, at 27% and 24%
reduced breast cancer risk when compared with women who never breastfed.
Overall, all the magnitude of risk was lower for African-American women than for
white women for all o f the pregnancy variables that we examined, but the effect of
lactation was similar in the two ethnic groups. Ethnicity did not significantly modify
any o f the observed associations between breast cancer risk and these reproductive
variables.
Table 3 shows that average number of full-term pregnancies and duration o f lactation
varied according to the woman’s reference age group. Older women had on average
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
a more full-term pregnancies than younger women in both ethnic groups. However,
trends were different in lactation. The average duration of lactation o f white women
less than 40 years was about twice that of women older than 55 years (mean lactation
= 44.9 and 21.3 weeks in control ever pregnant white women, less than 40 years and
older than 55 years, respectively). This downward trend with age was not observed
for African-American women.
Table 4 shows that the association between breast cancer risk and number of full-
term pregnancies varied according to the woman’s reference age group. Among
white women 40 years o f age or older, those who had had a full-term pregnancy
were at approximately 30% lower risk o f breast cancer than women who had not had
a full-term pregnancy. The amount o f reduction in risk per full-term pregnancy
increased with increasing age such that, for women over 55 years, a 13% reduction
in breast cancer risk per full-term pregnancy was observed (95% Cl: 0.79-0.96).
This reduction in risk per full-term pregnancy was much smaller (ranging from 1%
to 7%) among African-American women. But there were no statistically significant
effect modification by ethnic groups for any age group (results not shown).
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. Characteristics of reproductive variables am ong ever pregnant white and African-American women
according to ag e group
Total No. full-term pregnancies
Whites African-Americans Whites
Total Breastfeeding (wks)
African-Americans
Cases Controls Cases Controls Cases Controls Cases Controls
N Mean N Mean N Mean N Mean N Mean N Mean N Mean N Mean
Age group
<40 years 381 1.8 373 2.0 195 2.1 193 2.1 381 33.4 376 44.9 195 14.5 194 22.9
40-55 years 1309 2.0 1473 2.1 882 2.3 923 2.5 1309 30.3 1473 34.6 882 14.8 923 17.5
>55 years 799 2.8 802 3.1 415 3.5 425 3.6 799 17.1 801 21.3 413 29.1 426 25.5
Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 4. O dds ratios (ORs) and 95% confidence intervals (CIs) of breast cancer associated with
num ber of full-term pregnancies according to ag e group
All Whites African-Americans
OR per full-term OR per full-term OR per full-term
Ever vs. never full-term pregnancy In full-term Ever vs. never full-term pregnancy in full-term Ever vs. never full-term pregnancy in full-term
pregnancy pregnant women pregnancy pregnant women pregnancy pregnant women
OR 95% Cl OR 95% Cl OR 95% Cl OR 95% Cl OR 95% Cl OR 95% Cl
Age group
<40 years 1.01 0.61-1.67 0.97 0.84-1.11 0.69 0.35-1.38 0.93 0.76-1.13 1.24 0.54-2.85 0.99 0.79-1.23
40-55 years 0.77 0.61-0.96 0.96 0.90-1.02 0.66 0.49-0.88 0.95 0.87-1.04 0.98 0.66-1.44 0.95 0.87-1.04
>55 years 0.73 0.52-1.04 0.90 0.84-0.97 0.70 0.45-1.09 0.87 0.79-0.96 0.81 0.43-1.55 0.93 0.83-1.04
Note: ORs are adjusted for categorical variables of ethnicity, age at menarche, age at first full-term pregnancy, breastfeeding
duration, BMI, family history of breast cancer, study sites, education and poverty index (see text).
The association between lactation and breast cancer risk also varied according to
reference age group (Table 5). Overall, we observed breastfeeding had a protective
effect in all age groups combining white and African-American women with the
greatest reduction observed among women under age 40 years (OR: 0.52,95% Cl:
0.38-0.72 for age younger than 40 years, OR: 0.89, 95% Cl: 0.78-1.03 for ages 40-55
years, and OR: 0.83, 95% Cl: 0.70-0.99 for ages over 55 years). The reduction in
risk with ever having breastfed was statistically significant in white women and in
African-American women younger than 40 years and was somewhat greater for
whites (OR: 0.43, 95% Cl: 0.28-0.67) than for African-Americans (OR: 0.58, 95%
Cl: 0.35-0.95). In addition, among white women in this age group, for each 6
months’ increase in duration of lactation, breast cancer risk decreased about 18%,
whereas we did not find this protective effect in white women aged over 40 or in
African-American women in any age group. However, these differences by ethnicity
were not statistically significant (results not shown).
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 5. O dds ratios (ORs) and 95% confidence intervals (CIs) of breast cancer associated with
breastfeeding in ever pregnant wom en according to ag e group
A ll Whites African-Americans
Ever vs. never
breastfeeding
OR per 6 month
breastfeeding
Ever vs. never
breastfeeding
OR per 6 month
breastfeeding
Ever vs. never
breastfeeding
OR per 6 month
breastfeeding
OR 95% Cl OR 95% Cl OR 95% Cl OR 95% Cl OR 95% Cl OR 95% Cl
Age group
<40 years 0.52 0.38-0.72 0.85 0.77-0.93 0.43 0.28-0.67 0.82 0.73-0.92 0.58 0.35-0.95 0.86 0.71-1.05
40-55 years 0.89 0.78-1.03 0.96 0.92-1.00 0.89 0.74-1.07 0.96 0.91-1.02 0.88 0.71-1.08 0.94 0.86-1.01
>55 years 0.83 0.70-0.99 0.96 0.90-1.02 0.86 0.69-1.07 0.93 0.86-1.01 0.74 0.54-1.00 0.97 0.88-1.07
Note: ORs are adjusted for age at menarche, age at first full-term pregnancy, number of full-term pregnancies, BMI, family history of
breast cancer, study sites, education and poverty index (see text).
00
DISCUSSION:
In this multi-centered population-based case-control study we observed that: 1) risk
o f breast cancer decreased with early age at first full-term pregnancy, parity and
multiparity in both white and African-American women; 2) lactation reduced risk for
breast cancer in both ethnic groups. There was an inverse dose-response relationship
between total duration o f lactation and the risk o f breast cancer in both ethnic
groups; 3) overall, the magnitude o f risk was lower for African-American women
than white women for all o f the pregnancy variables that we examined, but the effect
o f lactation was similar in the two ethnic groups. However, ethnicity did not
significantly modify any o f the observed associations between breast cancer risk and
these reproductive variables.
There is abundant epidemiologic evidence showing that parity, multiparity, and early
age at first full-term pregnancy decreases the risk for breast cancer (MacMahon et
al., 1982; Yuan et al., 1988; Ewertz et al., 1990; Kelsey et al., 1993; Albrektsen et
al., 1994; Michels et al., 1996; Chie et al., 2000). In the present study, compared to
women who had never had a full-term pregnancy, women who had had at least one
were at a 23% lower risk o f breast cancer, which is comparable to other studies
(Albrektsen et al., 1994; Michels et al., 1996). Michel et al. observed an OR of 0.73
(95% Cl: 0.65-0.83) in their multi-center case-control study consisting of 4,599 cases
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and 4,536 control women aged 20-55 years. Albrektsen et al. observed an incidence
rate ratio o f 1.09 (95%CI: 0.97-1.22) for nulliparous compared to parous women in a
large national cohort study o f white women under age 55 in Norway. One concern
with method o f random digit dialing selection o f controls could be that they would
be more likely to represent women with many children who stay home. Given the
comparability of our results to those o f prior studies, it is unlikely that our controls
represent a biased sample with respect to pregnancy history.
Pathak et al postulated that a crossover in incidence curves between the two races
would be expected based on the effects and distribution of age at first full-term
pregnancy and parity (Pathak et al., 2000). In the Cancer and Steroid Hormone
(CASH) study 22.5 % of parous white control women had first full-term pregnancy
prior to age 20 years, whereas 55.6% o f parous black control women had their first
child by that age (Mayberry and Stoddard-Wright, 1992). In our study, 20.9% of
white women and 47.7% of African-American women controls had their first full-
term pregnancy prior to age 20 years. Our results support different reproductive
patterns in these two ethnic groups. Although the effect o f parity was not as strong
in African-American women as in whites, there were no statistical significant
differences between these ethnic groups. Thus it is possible that reproductive
patterns can explain somewhat higher incidences in African-American women
younger than 40 years.
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Previous studies that have found that the increased risk in breast cancer for
nulliparous compared to parous women was mostly observed among women with
breast cancer diagnosed after about 40-45 years o f age (Lubin et al., 1982; Pathak et
al., 1986). In the present study, we also found that this increased risk in women aged
over 40 years at reference date, although the increased risk in women older than 55
years did not reach statistical significance (OR: 0.73,95% Cl: 0.52-1.04). In
addition, in the present study, breast cancer risk significantly decreased with
increasing number of pregnancies and increasing number o f full-term pregnancies.
These effects were mainly confined to women aged 55 years and older, which is also
consistent with several previous studies (Lubin et al., 1982; Pathak et al., 1986;
Kvale et al., 1987; Bruzzi et al., 1988; Negri et al., 1988; Yuan et al., 1988; Layde et
al., 1989; Ewertz et al., 1990; Lund, 1991; Wang et al., 1992).
Chie et al. analyzed data from multi-state (Maine, Massachusetts, New Hampshire,
and Wisconsin), population-based case-control studies which involved 9,891 breast
cancer cases and 12,271 controls and found that each 5-year increase in age at first
full-term pregnancy was associated with an OR of 1.07 (95% Cl: 1.01, 1.03) (Chie et
al., 2000). In the present study, we found similar result, breast cancer risk increased
10 percent with each 5-year increasing in age at first full-term pregnancy in white
women (95% Cl: 1.02, 1.18). Restricting analysis to women aged 45 years and
older, since most of them have completed all their pregnancies, the results were
essentially unchanged.
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A previous study by Bruzzi et al. (Bruzzi et al., 1988) observed a trend o f increasing
risk with decreasing number of years since last birth among women under 50 years
who had 2 or more children. Compared with women whose last full-term birth had
occurred 10 or more years earlier, those whose last birth had occurred 7-9 years, 3-6
years, and < 3 years had relative risks of breast cancer o f 1.37, 1.81, and 2.66,
respectively. We observed that the risk o f breast cancer was elevated for white
women whose last full-term pregnancy was 7 to 10 years earlier but not for those
recent full-term pregnancies. Since most women over 55 years would not have had a
pregnancy in the past ten years, we further restricted analysis to women aged 45 or
younger. The results showed that compared with white women whose last full-term
pregnancy was at least 21 years earlier, the increased risk at 7 to 10 years since last
full-term pregnancy in whites aged 45 or younger was no longer statistically
significant (OR: 1.25,95% Cl: 0.73-2.15). The result for white women whose last
pregnancy was 11 to 20 years was below 1.0 (OR: 0.82, 95% Cl: 0.49-1.35) relative
to women who had their last full-term pregnancy more than 20 years earlier. We
also see no elevated risk for white women whose last full-term pregnancy was within
the 3 years before their reference date (OR: 0.81, 95% Cl: 0.46-1.44). No elevated
in risk was observed for African-American women under 45 years (results not
shown). Although we have examined this risk in all women who had a full-term
pregnancy, our data do not support the hypothesis that a recent full-term pregnancy
increases breast cancer risk.
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Effects of breastfeeding on breast cancer risk have been inconsistent in previous
studies. Most o f recent case-control studies have indicated statistically significant
inverse association, with relative risks ranging from 0.4 to 0.9, either confined to
young or premenopausal women (Lubin et al., 1982; Byers et al., 1985; McTieman
and Thomas, 1986; Layde et al., 1989; Yoo et al., 1992; UKNC-CS, 1993; Newcomb
et al., 1994); or for women diagnosed at various ages (Tao et al., 1988; Yuan et al.,
1988; Wang et al., 1992; Enger et al., 1998; Gilliland et al., 1998). Increasing
duration of breastfeeding with decreasing breast cancer risk was also found in some
of these studies (Byers et al., 1985; McTieman and Thomas, 1986; Tao et al., 1988;
Yuan et al., 1988; Wang et al., 1992; Yoo et al., 1992; UKNC-CS, 1993; Newcomb
et al., 1994; Gilliland et al., 1998) and yet not in other case-control studies (Adami et
al., 1990; Thomas and Noonan, 1993; Katsouyanni et al., 1996; Freudenheim et al.,
1997; McCredie et al., 1998) and two large cohort studies (Kvale et al., 1987;
Michels et al., 1996).
The inconsistencies in previous results may come from using different types of
controls in case-control studies. Population-based case-control studies more often
find an inverse association between breast cancer risk and increasing duration of
lactation (Lubin et al., 1982; Byers et al., 1985; McTieman and Thomas, 1986; Tao
et al., 1988; Yuan et al., 1988; Layde et al., 1989; Wang et al., 1992; UKNC-CS,
1993; Newcomb et al., 1994; Enger et al., 1998; Gilliland et al., 1998), whereas
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
hospital based case-control studies more frequently have shown no association
(MacMahon et al., 1970; Negri etal., 1996; Ramon et al., 1996).
A recent Icelandic Cohort study (Tryggvadottir et al., 2001) observed that an inverse
association between total duration of breastfeeding and breast cancer risk with the
adjusted odds ratio o f 0.77 per 6 months’ increase in duration of breastfeeding,
mainly confined to women younger than 40 years at diagnose. In the present study,
we also observed an inverse dose-response relationship between lactation and breast
cancer in both ethnic groups. In consistent with their results, the reduction in risk
with ever having breastfed was significant in women younger than 40 years with
somewhat greater for whites than for African-American women. Our results were
also in consistent with other previous studies. One study in Los Angeles County o f
women aged 40 years or younger showed that women who breastfed at least 16
months had 34 percent decreased risk compared with parous women who never
breastfed (Enger et al., 1997). Newcomb et al reported that with an increasing
cumulative duration o f lactation, there was a decreasing risk of breast cancer among
premenopausal women (P for trend <0.001) in Northeast region of America
(Newcomb et al., 1994). Moreover, the two previous large cohort studies did not
suggest an overall association between lactation and breast cancer, one did not
include any women diagnosed under the age o f 40 (Kvale et al., 1987), and in the
other the majority o f cases were postmenopausal (Michels et al., 1996).
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The diversity of duration o f breastfeeding between populations in different
geographic area (Lipworth et al., 2000) may provide another explanation for the
discrepancies between results. For instance, in China, where most available
evidence suggests a protective effect o f long-term lactation (Tao et al., 1988; Yuan et
al., 1988; Wang et al., 1992), more than half o f the parous control women lactated
for at least three years (Wang et al., 1992). Adjusted odds ratios o f around 0.4 - 0.5
were found among women who breastfed for 10 years or more as compared with
women who breastfed less than 3 years (Yuan et al., 1988; Wang et al., 1992), while
lactation for 3-5 years had little effect. However, only 6 percent o f women have
breastfed for at least 24 months in the United States (London et al., 1990). The
National Survey for Family Growth conducted by National Center for Health
Statistics (NCHS) showed that the average duration o f lactation among whites during
the years 1990-1993 was 30 weeks vs. 23 weeks for African Americans (National
Institutes of Health, 2000). These data indicate limited detection o f lactation in US
populations; thus, it is more difficult to detect effects o f long-term lactation. In the
present study, the average duration of lactation is about 21 weeks and 8 percent of
women reported lactating for at least 24 months in controls. This is similar to the
average lactation status o f US populations and once again, indicates that our study
population is representative o f US women in general.
Even within same geographic area, the public opinion regarding breastfeeding is
changing. For instance, the generation of women older than 55 years were not
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
encouraged for breastfeeding, they were generally less breastfeeding during their
reproductive ages. Whereas in recent years, the public believes that breastfeeding is
best for the baby, women tend to breastfeed their baby more. In consistent with
changing o f public opinion, interestingly, recent studies more often showed the
protective effect o f breastfeeding, whereas earlier studies more frequently miss this
effect. Not surprisingly, in the present study, we observed that average total duration
of breastfeeding differed by age, in whites being highest in women younger than 40
years and lowest in women older than 55 years. Whereas in African-American
women we found on average a lower duration of breastfeeding in all ages.
The reduction in risk with ever having breastfed was significant in women younger
than 40 years with somewhat greater for whites than for African-American women
may help us understanding the crossover pattern in incidence rates which showed in
young women it is greater in the African-American population, whereas in white
women, it is higher in middle-aged and elderly women. Despite the different
distributions o f known risk factors for breast cancer development (early age at
menarche, late age at first full-term pregnancy, and a late age o f menopause) in
African-American and white populations (Gray et al., 1980; Mayberry and Stoddard-
Wright, 1992), the somewhat stronger lactation protection in white women younger
than 40 years than in African-American women might be another explanation for this
crossover. More investigation is needed to examine whether this is a coincidence or
a true causal relationship. It is also possible that the postulated protective effect of
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
breastfeeding lasted only for a certain period after cessation o f lactation, which may
cause the shift around the age of 40.
Several biological mechanisms have been postulated for the inverse association
between lactation and breast cancer risk. Lactation may reduce the breast cancer risk
by delaying ovulation, modifying pituitary, and ovarian hormone secretion
(Henderson et al., 1982; Byers et al., 1985; McTieman and Thomas, 1986; Petrakis
et al., 1987). Long-term endogenous hormonal changes, such as, reduced estrogen
levels and increased prolactin production, may decrease a woman's cumulative
exposure to estrogen, which may inhibit the initiation or growth of breast cancer
cells (Byers et al., 1985; Key and Pike, 1988). Breast fluid estrogen levels,
independent of serum estrogen levels, were found to be lower in premenopausal
parous women who were breastfeeding than in parous women who did not lactate or
in nulliparous women (Petrakis et al., 1987). This might be another reason why the
protective effect o f lactation appeared stronger in premenopausal women. Excretion
of carcinogenic agents from the breast ductal tissue during breastfeeding was also
suggested (Murrell, 1991). Lactation, like pregnancy, may also directly affect risk
by making breast tissue more resistant to carcinogenesis through increasing
differentiation in the epithelial cells of the mammary ducts (Russo and Russo, 1994).
The strengths o f our study included a population-based design with community
controls which not only provide a good representation o f population, but also provide
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the power to examine breast cancer risks among white and African-American women
respectively. Additional strengths were standardized data collection, recording and
processing procedures to minimize inter- and intra-center variation, comprehensive
set of risk factors for breast cancer that allow us fully adjust confounding in the
analyses. The response rate for cases and controls were similar.
In conclusion, our results support that the parity, multiparity, early age at first full-
term pregnancy, and lactation are all associated with decreased risk in breast cancer.
In addition, lactation showed an inverse dose-response relationship with breast
cancer. Overall, although parity factors were less associated with breast cancer risk
in African-American women than in whites, the differences between white and
African-American women did not reach statistically significance. The effect of
lactation was similar in both ethnic groups. If breastfeeding, a potential modifiable
factor for breast cancer, indeed provide a protective effect, incidences o f breast
cancer would be expected a reducing by simply lactating in women who are at
reproductive ages.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES:
Adami, H. O., R. Bergstrom, et al. (1990). Absence of association between
reproductive variables and the risk of breast cancer in young women in
Sweden and Norway. Br J Cancer 62(1): 122-6.
Albrektsen, G., I. Heuch, et al. (1994). Breast cancer incidence before age 55 in
relation to parity and age at first and last births: a prospective study o f one
million Norwegian women, [see comments]. Epidemiology 5(6): 604-11.
Breslow, N. and N. Day (1980). Statistical Methods in Cancer Research.
Bruzzi, P., E. Negri, et al. (1988). Short term increase in risk o f breast cancer after
full term pregnancy. BMJ 297(6656): 1096-8.
Byers, T., S. Graham, et al. (1985). Lactation and breast cancer. Evidence for a
negative association in premenopausal women. Am J Epidemiol 121(5): 664-
74.
Chie, W. C., C. Hsieh, et al. (2000). Age at any full-term pregnancy and breast
cancer risk. Am J Epidemiol 151(7): 715-22.
Dalker, J. and B. D. Proctor (2000). U.S. Census Bureau, Current Population
Reports. Washington DC, U.S. Government Printing Office.
Enger, S. M., R. K. Ross, et al. (1997). Breastfeeding history, pregnancy experience
and risk o f breast cancer. Br J Cancer 76(1): 118-23.
Enger, S. M., R. K. Ross, et al. (1998). Breastfeeding experience and breast cancer
risk among postmenopausal women. Cancer Epidemiol Biomarkers Prev
7(5): 365-9.
Ewertz, M., S. W. Duffy, et al. (1990). Age at first birth, parity and risk of breast
cancer: a meta-analysis o f 8 studies from the Nordic countries. Int J Cancer
46(4): 597-603.
Freudenheim, J. L., J. R. Marshall, et al. (1997). Lactation history and breast cancer
risk, [see comments]. Am J Epidemiol 146(11): 932-8.
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Gilliland, F. D., W. C. Hunt, et al. (1998). Reproductive risk factors for breast cancer
in Hispanic and non-Hispanic white women: the New Mexico Women's
Health Study. Am J Epidemiol 148(7): 683-92.
Gray, G. E., B. E. Henderson, et al. (1980). Changing ratio of breast cancer incidence
rates with age o f black females compared with white females in the United
States. J Natl Cancer Inst 64(3): 461-3.
Henderson, B. E., R. K. Ross, et al. (1982). Endogenous hormones as a major factor
in human cancer. Cancer Res 42(8): 3232-9.
Katsouyanni, K., L. Lipworth, et al. (1996). A case-control study o f lactation and
cancer o f the breast. Br J Cancer 73(6): 814-8.
Kelsey, J. L., M. D. Gammon, et al. (1993). Reproductive factors and breast cancer.
Epidemiol Rev 15(1): 36-47.
Key, T. J. and M. C. Pike (1988). The role of oestrogens and progestagens in the
epidemiology and prevention of breast cancer. Eur J Cancer Clin Oncol
24(1): 29-43.
Kvale, G., I. Heuch, et al. (1987). A prospective study o f reproductive factors and
breast cancer. I. Parity. Am J Epidemiol 126(5): 831-41.
Layde, P. M., L. A. Webster, et al. (1989). The independent associations o f parity,
age at first full term pregnancy, and duration o f breastfeeding with the risk of
breast cancer. Cancer and Steroid Hormone Study Group. J Clin Epidemiol
42(10): 963-73.
Lipworth, L., L. R. Bailey, et al. (2000). History of breast-feeding in relation to
breast cancer risk: a review of the epidemiologic literature, [see comments]. J
Natl Cancer Inst 92(4): 302-12.
London, S. J., G. A. Colditz, et al. (1990). Lactation and risk of breast cancer in a
cohort o f US women. Am J Epidemiol 132(1): 17-26.
Lubin, J. H., P. E. Bums, et al. (1982). Risk factors for breast cancer in women in
northern Alberta, Canada, as related to age at diagnosis. J Natl Cancer Inst
68(2): 211-7.
Lund, E. (1991). Breast cancer mortality and the change in fertility risk factors at
menopause: a prospective study o f 800,000 married Norwegian women.
Epidemiology 2(4): 285-8.
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MacMahon, B., T. M. Lin, et al. (1970). Lactation and cancer o f the breast. A
summary o f an international study. Bull World Health Organ 42(2): 185-94.
MacMahon, B., M. Purde, et al. (1982). Association o f breast cancer risk with age at
first and subsequent births: a study in the population o f the Estonian
Republic. J Natl Cancer Inst 69(5): 1035-8.
Marchbanks, P., J. McDonald, et al. (2001 (in press)). The NICHD women's
contraceptive and reproductive experiences study: methods and operational
results. Ann Epidemiol.
Mayberry, R. M. and C. Stoddard-Wright (1992). Breast cancer risk factors among
black women and white women: similarities and differences. Am J Epidemiol
136(12): 1445-56.
McCredie, M., C. Paul, et al. (1998). Reproductive factors and breast cancer in New
Zealand. Int J Cancer 76(2): 182-8.
McTieman, A. and D. B. Thomas (1986). Evidence for a protective effect of
lactation on risk of breast cancer in young women. Results from a case-
control study. Am J Epidemiol 124(3): 353-8.
Michels, K. B., W. C. Willett, et al. (1996). Prospective assessment o f breastfeeding
and breast cancer incidence among 89,887 women, [see comments]. Lancet
347(8999): 431-6.
Murrell, T. G. (1991). Epidemiological and biochemical support for a theory on the
cause and prevention o f breast cancer. Med Hypotheses 36(4): 389-96.
National Institutes o f Health, N. C. I., Cancer Statistics Branch (2000). Health,
United States, 2000.
Negri, E., C. Braga, et al. (1996). Lactation and the risk o f breast cancer in an Italian
population. Int J Cancer 67(2): 161-4.
Negri, E., C. La Vecchia, et al. (1988). Risk factors for breast cancer: pooled results
from three Italian case-control studies. Am J Epidemiol 128(6): 1207-15.
Newcomb, P. A., B. E. Storer, et al. (1994). Lactation and a reduced risk of
premenopausal breast cancer, [see comments]. N Engl J Med 330(2): 81-7.
Pathak, D. R., J. R. Osuch, et al. (2000). Breast carcinoma etiology
Current knowledge and new insights into the effects o f reproductive and hormonal
risk factors in black and white populations. Cancer 88(S5): 1230-1238.
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Pathak, D. R., F. E. Speizer, et al. (1986). Parity and breast cancer risk: possible
effect on age at diagnosis. Int J Cancer 37(1): 21-5.
Petrakis, N. L., M. R. Wrensch, et al. (1987). Influence o f pregnancy and lactation on
serum and breast fluid estrogen levels: implications for breast cancer risk. Int
J Cancer 40(5): 587-91.
Ramon, J. M., J. M. Escriba, et al. (1996). Age at first full-term pregnancy, lactation
and parity and risk o f breast cancer: a case-control study in Spain. Eur J
Epidemiol 12(5): 449-53.
Russo, J. and I. H. Russo (1994). Toward a physiological approach to breast cancer
prevention. Cancer Epidemiol Biomarkers Prev 3(4): 353-64.
Software, E. (1999). Epilog Windows. Pasadena, EPICENTER Software.
Tao, S. C., M. C. Yu, et al. (1988). Risk factors for breast cancer in Chinese women
o f Beijing. Int J Cancer 42(4): 495-8.
Thomas, D. B. and E. A. Noonan (1993). Breast cancer and prolonged lactation. The
WHO Collaborative Study o f Neoplasia and Steroid Contraceptives. Int J
Epidemiol 22(4): 619-26.
Track, B. J. (1996). Breast cancer in African American women: epidemiology and
tumor biology. Breast Cancer Res Treat 40(1): 11-24.
Tryggvadottir, L., H. Tulinius, et al. (2001). Breastfeeding and reduced risk o f breast
cancer in an Icelandic cohort study. Am J Epidemiol 154(1): 37-42.
UKNC-CS, G. (1993). Breast feeding and risk o f breast cancer in young women.
United Kingdom National Case-Control Study Group. BMJ 307(6895): 17-
20.
Wang, Q. S., R. K. Ross, et al. (1992). A case-control study of breast cancer in
Tianjin, China. Cancer Epidemiol Biomarkers Prev 1(6): 435-9.
Yoo, K. Y., K. Tajima, et al. (1992). Independent protective effect of lactation
against breast cancer: a case-control study in Japan. Am J Epidemiol 135(7):
726-33.
Yuan, J. M., M. C. Yu, et al. (1988). Risk factors for breast cancer in Chinese
women in Shanghai. Cancer Res 48(7): 1949-53.
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Endometrial cancer following breast cancer treatment: Tumor characteristics and predictors of survival
PDF
beta3-adrenergic receptor gene Trp64Arg polymorphism and obesity-related characteristics among African American women with breast cancer: An analysis of USC HEAL Study
PDF
Associations of weight, weight change and body mass with breast cancer risk in Hispanic and non-Hispanic white women
PDF
An exploration of nonresponse with multiple imputation in the Television, School, and Family Project
PDF
Androgens and breast cancer
PDF
A comparative study of environmental factors associated with multiple sclerosis in disease-discordant twin pairs
PDF
A joint model for Poisson and normal data for analyzing tumor response in cancer studies
PDF
Familiality and environmental risk factors of peptic ulcer: A twin study
PDF
BRCA1 mutations and polymorphisms in African American women with a family history of breast cancer identified through high throughput sequencing
PDF
Descriptive epidemiology of thyroid cancer in Los Angeles County, 1972-1995
PDF
Polymorphisms in ileal sodium-dependent bile acid transporter gene and the risk of colorectal adenomas
PDF
Analysis of gene-environment interaction in lung cancer
PDF
Construction and validation of multivariable population pharmacokinetic models: Utility of allometric forms to predict the pharmacokinetic disposition of gentamicin in pediatric patients with app...
PDF
Associations between lung function growth and air pollution in two cohorts in Southern California children
PDF
Does young adult Hodgkin's disease cluster by school, residence and age?
PDF
Effects of ambient air pollution on monthly wheeze and asthma exacerbation in southern California school children
PDF
Determinants of mammographic density in African-American, non-Hispanic white and Hispanic white women before and after the diagnosis with breast cancer
PDF
A study of pediatric oncology nurses' attitudes to and knowledge of genetic testing
PDF
Cognitive dysfunction and occupations with likely significant magnetic field exposure: A cross-sectional study of elderly Mexican Americans
PDF
Predictive factors for the progression of glucose tolerance over time in Latino women after gestational diabetes mellitus
Asset Metadata
Creator
Wang, Yaping
(author)
Core Title
Do reproductive factors have the same effect on breast cancer risk in African-American and White women?
School
Graduate School
Degree
Master of Science
Degree Program
Applied Biometry
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
Black studies,health sciences, oncology,health sciences, public health,OAI-PMH Harvest
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Ursin, Giske (
committee chair
), Bernstein, Leslie (
committee member
), Gauderman, William James (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-292710
Unique identifier
UC11341359
Identifier
1411040.pdf (filename),usctheses-c16-292710 (legacy record id)
Legacy Identifier
1411040.pdf
Dmrecord
292710
Document Type
Thesis
Rights
Wang, Yaping
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
Tags
Black studies
health sciences, oncology
health sciences, public health