American Journal of Epidemiology
Copyright O 1996 by The Johns Hopkins University School of Hygiene and Public Health
All lights reserved
Vol 143, No 11
Printed in U.SA.
Adult Height and Risk of Breast Cancer among White Women in a CaseControl Study
Yuqing Zhang,1 Lynn Rosenberg,2 Theodore Cotton,1 L Adrienne Cupples,1 Julie R. Palmer,2
Brian L Strom,3 Ann G. Zauber,4 M. Ellen Warshauer,5 Susan Hariap,4 and Samuel Shapiro2
Data from a hospital-based case-control study were analyzed to evaluate the relation of adult height to the
risk of breast cancer among white women. The authors compared 5,358 newly diagnosed breast cancer cases
and 4,555 controls interviewed from 1976 to 1992 in hospitals located mainly in the United States. Overall,
there was no association between stature and nsk of breast cancer. In comparison with women whose heights
were less than 62 inches (< 158 cm), the adjusted odds ratios were 1.1 (95% confidence interval (Cl) 0.9-1.2),
1.0 (95% Cl 0.9-1.2), 1.0 (95% Cl 0.9-1.1), and 1.0 (95% Cl 0.8-1.2) for women with heights of 62-63, 64-65,
66-67, and >68 inches (equivalent to 158-160,163-165,168-170, and 2173 cm), respectively. There was no
consistent evidence of modification of the effect of height by other risk factors. The results suggest that adult
stature in white women is not related to the risk of breast cancer. Am J Epidemiol 1996;143:1123-8.
body height; breast neoplasms; contraceptives, oral
A variety of body build indicators, such as weight,
height, and body mass index, have been examined in
studies of breast cancer. During the past several decades, research has focused primarily on weight or
body mass index, and the effect of these factors appears to vary according to the woman's menopausal
status: Modest positive associations have been found
among postmenopausal women, while an inverse relation has sometimes been present among premenopausal women (1).
The effect of height on the risk of breast cancer has
been less thoroughly investigated, and the epidemiologic evidence does not indicate a convincing association (2, 3). In some studies, tallness is associated with
a modest increased risk of breast cancer, especially
among postmenopausal women, but other studies have
reported no association of stature with risk. Few studies (4-8) have examined whether a relation of height
Received for publication October 17, 1994, and in final form
February 21, 1996.
1
Department of Epidemiology and Blostatlstics, School of Public
Health, Boston University, Boston, MA.
2
Stone Epidemiology Unit, School of Medicine, Boston University, Brookline, MA
3
Center for Clinical Epidemiology and Biostatistics and Division
of General Internal Medicine, School of Medicine, University of
Pennsylvania, Philadelphia, PA.
* Department of Epidemiology and Biostatistics, Memorial SloanKettering Cancer Center, New York, NY.
5
Cancer Care and Research Program, New York Hospital, and
Department of Public Health, Cornell Medical Center, New York,
NY.
to the risk of breast cancer is modified by other risk
factors, and the findings have been inconsistent.
We analyzed data from a hospital-based casecontrol study to evaluate the relation of adult height to
breast cancer risk in white women, and to explore
whether the effect of height could be modified by
other breast cancer risk factors.
MATERIALS AND METHODS
Since 1976, a case-control drug surveillance study
has been conducted in hospitals located in several
geographic regions of the United States (the cities of
Boston, New York, Philadelphia, Baltimore, San Francisco, and Tucson and the state of Kansas) and in
Canada and Israel (9). Over 90 percent of patients
were interviewed in the first four US cities. The participating hospitals, which change from time to time,
include teaching, cancer, and community hospitals.
Specially trained nurse-interviewers identify and interview women admitted for breast cancer and for a wide
range of other conditions. A standard questionnaire is
used to obtain information on personal characteristics,
relevant medical history, and history of medication
use. After discharge, each patient's diagnoses, including the primary diagnosis that led to hospital admission, are abstracted from the hospital record. White
women interviewed through March 31, 1992, were
included in the present study. Ninety-two percent of
the women were from the United States, and 8 percent
1123
1124
Zhang et al.
were from Israel. Of the patients approached, approximately 4 percent refused to be interviewed.
Selection of cases
White women aged 20-69 years who met the following criteria were included in the case series: a
diagnosis of breast cancer recorded in the discharge
summary or pathology report (10) that had been made
within 1 year before the current admission, and no
other primary cancer or history of cancer. Virtually all
of the women were identified when they were admitted for treatment of the cancer. Of 5,383 case women,
information on adult height was available from 5,358
(99.5 percent), who comprised the final case series. Of
the latter, 5,127 (96 percent) had been diagnosed
within the preceding 6 months; the median age was 51
years, and 82 cases (1.6 percent) were under 30 years
of age. Almost all of the case women (97 percent)
were from the United States.
Selection of controls
Controls were selected from a pool of approximately 25,000 white women aged 20-69 years who
had been admitted for nonmalignant conditions and
who had no history of cancer. We excluded women
who were admitted for diseases judged to be related to
height or weight, such as endocrine, nutritional, and
metabolic diseases, and for diseases related to pregnancy, childbirth, and the puerperium. Women were
also excluded if their primary diagnosis was hypertension, ischemic heart disease, back pain, a disc disorder,
another orthopedic problem, or trauma, since some
studies (11, 12) have suggested that these diseases
may be associated with stature.
There were 9,018 controls remaining after the exclusions. The median age was 45 years, and 22.5
percent were under age 30. To attain more comparable
age and geographic distributions between cases and
controls, we selected a subset of the controls using
frequency matching. For each case stratum defined by
5-year age category and geographic region of the
hospitals (New York City, Boston, Philadelphia, Baltimore, San Francisco, Tucson, Kansas, and Israel), we
randomly selected up to twice as many controls as
cases. The final control series totaled 4,555 women;
the median age was 47 years, and 68 (1.5 percent)
were under 30 years of age.
The final control group included women with diseases of the digestive system (15.3 percent); diseases
of the genitourinary system (12.1 percent); diseases of
skin, muscle, and bone (15.9 percent); accidents (21.4
percent); and other conditions (35.3 percent).
Exposure and potentially confounding factors
All participants were interviewed in the hospital
after they had given informed consent. Height was
reported to the nearest inch or centimeter. Most participants reported their height in inches; thus, we used
these units rather than metric values in the analyses.
Data were also collected on many potential risk factors
for breast cancer, including years of education, religion, age at menarche, menopausal status, age at first
pregnancy, parity, history of benign breast disease,
history of breast cancer in a mother or sister, oral
contraceptive use, estrogen use, smoking, and alcohol
consumption.
Analysis
We divided the subjects into five height categories:
<62, 62-63, 64-65, 66-67, and >68 inches (equivalent to <158, 158-160, 163-165, 168-170, and
>173 cm). A height of less than 62 inches was the
reference category for relative risk estimation. The
relative risk (odds ratio) of breast cancer was estimated by multiple logistic regression. In addition to
variables for the height groups, the regression models
included indicator terms for age, year of interview,
geographic area, religion, education, current body
mass index (weight (kg)/height squared (m2)), age at
menarche, menopausal status, age at first pregnancy,
parity, history of benign breast disease, family history
of breast cancer, oral contraceptive use, conjugated
estrogen use, alcohol use, and cigarette smoking. To
test for a trend in risk, we included a continuous term
for height in the multivariate model.
To test whether the relation between height and
breast cancer risk was modified by other risk factors,
we stratified the subjects according to levels of each
risk factor and used multiple logistic regression to
obtain estimates of relative risk for height categories
within strata. Interaction terms for height and particular risk factors were used in the regression model to
test for statistical significance.
RESULTS
Table 1 shows the age-standardized height distribution for different disease categories among the controls. Most disease categories had similar height distributions, with the percentages for the lowest height
category ranging from 13.3 percent to 14.4 percent and
those for the tallest height category ranging from 8.9
percent to 12.3 percent.
We examined the age-standardized distribution of
heights according to breast cancer risk factors in the
controls. Short stature was associated with earlier
years of interview, living in Israel, fewer years of
Am J Epidemiol
Vol. 143, No. 11, 1996
Height and Breast Cancer
1125
TABLE 1. Height distribution (%) of selected hospital control* from th* United State* and
bra©!, by dls*as* category, 1976-1992
No.
Disease catsgoty
(ICO-8* oodes)
He4^1 t,t (Incr»•)
462
62-63
64-66
66-67
£68
697
14.1
31.5
23.3
20.3
10.8
550
14.1
25.3
28.2
23.5
9.0
•ubjects
Diseases of the digestive system
(520-679)
Diseases of the genitourinary
system (580-629)
Diseases of skin, muscle, and
bone (680-739)
Accidents (800-099)
Other (1-139, 280-519, and
78O-799)
726
974
13.3
13.8
25.3
24.3
23.8
32.6
25.3
18.2
12.3
11.1
1,608
14.4
26.9
26.2
23.6
8.9
Total
4,555
14.1
26.4
27.3
22.0
10.2
* ICD-8. International Classification of Diaeasa
a. Eahth Revision M 0).
t The disease-specific height distribution was standardized with the age distribution of total
eligible controls.
11 inch - Z 5 cm.
education, Jewish religion, earlier year of birth, greater
current body mass index, earlier age at menarche,
earlier age at first pregnancy, no family history of
breast cancer, and abstinence from alcohol drinking; it
was inversely associated with nulliparity. There was
little relation with menopausal status, benign breast
disease, oral contraceptive use, estrogen use, and cigarette smoking, and there was an inconsistent relation
with body mass index at age 20.
Table 2 gives relative risk estimates (odds ratios)
according to height. All risk estimates were close to
1.0. There was no indication of an increase in breast
cancer risk with increasing height (p for trend =
0.31). Compared with women in the shortest height
category (<62 inches), the crude odds ratio for breast
cancer among women in the tallest height category
(^68 inches) was 0.9. With adjustment for age, geographic area, and major breast cancer risk factors, the
risk estimate was 1.0 (95 percent confidence interval
0.8-1.2).
Relations between height and breast cancer risk
within categories of the other factors are shown in
table 3. There was little evidence for an association
between breast cancer risk and height within categories of geographic region, age at menarche, menopausal status, age at first pregnancy, parity, history of
benign breast disease, family history of breast cancer,
and conjugated estrogen use. Statistically significant
inverse associations between height and breast cancer
risk were found in some categories (such as women
interviewed in Boston, women with body mass indices
over 29 at interview, and women over 30 years of age
at first pregnancy), but there were no consistent patterns within other categories of those variables. There
was a statistically significant trend of increasing risk
of breast cancer with increasing height among oral
contraceptive users, and a significant trend of decreasing risk among nonusers. hi the multiple logistic regression model, the interaction term for height and
TABLE 2. Height distribution of 5,358 brmast canc*r
the United State* and Israel, 1976-1992
Hefont
(InchM)'
<62
62-63
64-66
66-67
268
No.
of
No.
of
oortrote
848
1,459
1,517
1,082
447
723
1,196
1,287
930
419
and 4,555 hospital controls from
Crude
ORt
Ad|ustad
OR*
io§
1.0§
1.1
1.0
1.0
1.0
1.0
1.0
1.0
0.9
96% a t
0.9-1.2
0.9-1.2
0.9-1.1
0.8-1.2
• 1 inch - 2.5 cm.
t OR, odds ratio; Cl, confidence interval.
i Odds ratios were adjusted by logistic regression for age, geographic area, religion, year of
interview, education, body mass index at interview, cigarette smoking, history of benign breast disease,
family history of breast cancer, age at menarche, menopausal status, age at first pregnancy, parity, use
of oral contraceptives, estrogen replacement therapy, and alcohol drinking.
Am J Epidemiol
Vol. 143, No. 1 1 , 1 9 9 6
1126
Zhang et al.
TABLE 3. Adjusted odd* ratio** for breast canoer In relationtoheight, aooordlngtoother breast oanoer risk factor*, In women
from the United State* and Israel, 1976-1902
Teittor
Ha l(fd (incfrMt)
\AiiLiJ JJI
WlHMO
Geographic area
Boston, Massachusetts
New Mark, New Mark
Philadelphia, Pennsylvania
Israel
Other
Body mass index§ at interview
£20
21-22
23-24
25-28
£29
Age (years) at menarche
<11
11-12
13-14
215
Menopausal status
Premenopauaal
Postmenopausal
Age (years) at first pregnancy
<20
20-24
25-29
230
Parity
NuQipara
1-2
3-4
26
Benign breast disease
No
MM
Family history of breast cancer
No
yes
Use of oral contraceptives
Nonuser
Ever use
Estrogen use
Nonuser
Ever use
tmrvfl
UW1U
<62*
62-63
84-65
66-67
i68
(pvahia)
1.0
1.0
1.0
1.0
1.0
0.9
0.9
12
1.1
12
0.9
1.0
1.3
0.6
0.9
0.7
1.0
1.2
12
12
0.8
0.8
1.1
12
1.8
0.018
0262
0.590
0.954
0.415
1.0
1.0
1.0
1.0
1.0
1.1
1.3
1.0
0.9
1.0
1.3
12
1.0
0.8
0.8
1.5
2
1.0
0.7
0.8
12
12
12
0.8
0.6
0.112
0.769
0.887
0.067
0.030
1.0
1.0
1.0
1.0
1.1
1.0
1.1
12
1.1
1.0
0.9
1.1
0.9
1.0
1.1
1.3
1.0
1.0
0.9
0.7
0.876
0.414
0.476
0.872
1.0
1.0
1.1
1.0
1.1
0.9
1.1
0.9
1.1
0.9
0.768
0.063
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.1
1.3
0.9
1.0
0.8
0.8
1.0
0.9
0.6
1.7
1.0
1.0
0.6
0.770
0.883
0.793
0.003
1.0
1.0
1.0
.0
0.9
1.0
12
1.0
1.1
1.0
1.1
0.9
1.0
1.0
1.0
0.7
0.7
1.1
12
1.1
0202
0.873
0.841
0.563
.0
.0
1.1
0.9
1.1
0.8
1.0
0.9
1.0
1.0
0.367
0.397
1.0
1.0
1.1
0.8
1.0
0.7
1.0
1.0
1.0
0.8
0276
0.828
1.0
1.0
0.9
1.3
0.9
1.3
0.6
1.4
0.8
1.5
0.002
0.047
1.0
1.0
1.0
1.0
1.1
1.0
0.8
1.0
0.8
1.0
0.321
0.391
* AD models contained age, geographic area, religion, year of interview, body mass index at interview, age at menarche, menopausal
status, aaa at first Dreanancv. oaritv. history of benkm breast olsease. famQv history of breast cancer, use of oral contraceDtiv«s. estrooen
replacement therapy, smoking history, and alcohol drinking, except that for each stratified variable itself, which excluded that particular
variable.
1 1 inch - 2.5 cm.
i Reference category.
§ Weight (kgVheight* (m*).
oral contraceptive use was statistically significant
(p < 0.001).
DISCUSSION
The results of the present study, conducted mainly
among US white women, suggest that height has little
if any relation to the risk of breast cancer. The adjusted
relative risk estimates for all height categories were
close to 1, and this was generally the case within most
levels of other risk factors.
There were a few statistically significant associations when we examined height within categories of
breast cancer risk factors. For example, the relation
appeared to be modified by oral contraceptive use.
Am J Epidemiol
Vol. 143, No. 11, 1996
Height and Breast Cancer
Among users of oral contraceptives, the risk of breast
cancer increased as stature increased; however, an
inverse association was observed among nonusers.
The finding could well be attributable to chance.
To date, the epidemiologic evidence on the relation
of height to the risk of breast cancer has not been
conclusive (3). A positive correlation between adult
height and breast cancer incidence has been observed
in several international correlational studies (13-15),
and results from many analytical epidemiologic studies (4-7, 16-30) have also indicated such an association. On the other hand, conflicting results have been
observed in many other studies (8, 24, 31-38). Of 10
studies of US Caucasian women in which the relation
of height to breast cancer risk was reported (17, 19,28,
29, 31-36), four (17, 19, 28, 29) have found a positive
association.
A biologic mechanism for a link between height and
breast cancer risk in women has not been well established. A proposed explanation is that high energy
intake during childhood and adolescence is related to
larger adult body size and earlier maturation of ovarian
function, and that these factors, in turn, are associated
with the risk of breast cancer (39). London et al. (32)
suggested that the inconsistent results might be due to
variability in early nutritional status in the study populations. Indeed, the positive findings were often seen
in the studies in which some women had experienced
nutritional deprivation early in life (19-22). Consistent with this possibility is the fact that, in data collected from African-American women in the same
study as the present study of white women, short
stature was related to reduced risk (20). Black women
have experienced greater poverty levels than white
women, and this may be reflected in their energy
balance in childhood. However, in most industrialized
countries, variability of height probably predominantly reflects genetic factors rather than differences
in early nutritional status. Positive associations have
also been reported in some studies of US white
women, among whom there is unlikely to have been
severe nutritional deprivation (17, 28, 29).
The conflicting results on the relation of height to
breast cancer have mostly been seen in case-control
studies. Selection bias may have contributed to the
inconsistent findings. In most case-control studies that
reported an association between height and risk of
breast cancer, the major objective was not to test the
relation of stature to risk. In those studies, especially
the hospital-based case-control studies, the controls
might have been satisfactory for testing the primary
hypothesis but not have been a valid comparison group
with which to assess the effect of height, because adult
Am J Epidemiol Vol. 143, No. 11, 1996
1127
height appears to be related to the risk of some other
diseases (11, 12).
The main objective of the present analysis was to
evaluate the relation between height and breast cancer
risk. Potential controls were carefully selected from
women whose diagnoses were unrelated to height. The
height distributions were similar across the different
geographic areas within the United States. They were
also similar across a range of diagnostic categories,
including diagnoses for which admission to a hospital
is usually obligatory and, therefore, less susceptible to
selection bias. With regard to selection bias in the case
series, referral patterns for cases and controls may
have differed. However, the results were consistent
across geographic regions, each of which had a different mix of teaching, cancer, and community hospitals.
Further evidence against material selection bias is the
fact that the relations of other major risk factors to
breast cancer risk in the present data are consistent
with other investigators' reports (9). Nevertheless, we
are unable to rule out the possibility that selection bias
might account for the findings.
Participants in the present study were not measured
by the interviewers, and we were unable to validate the
self-reported information. Self-reported height is fairly
accurate, but some studies (40, 41) have shown a
tendency for short women to overstate their height.
However, there is no reason to suspect that women
with breast cancer would understate or overstate their
height differently from controls.
Important breast cancer risk factors were controlled
simultaneously in multivariate analysis. The estimates
of effect of height derived from these analyses were
similar to crude ones, suggesting that there was little if
any confounding from the major breast cancer risk
factors.
In summary, the present study found no association
between adult stature and risk of breast cancer among
white women.
ACKNOWLEDGMENTS
The authors thank the many physicians who allowed their
patients to be interviewed; the nurse-interviewers who collected the data; Marguerite Angeloni, who coordinated data
collection; and Glenn Street, who assisted in data management.
REFERENCES
1. Hunter DJ, Willett WC. Diet, body size, and breast cancer.
Epidemiol Rev 1993;15:110-32.
2. Harris JR, Lippman ME, Veronesi U, et al. Breast cancer (1).
(First of three parts). N Engl J Med 1992;327:319-28.
1128
Zhang etaJ.
3. Kelsey JL. Breast cancer epidemiology: summary and future
directions. Epidemiol Rev 1993;15:256-63.
4. Ravnihar B, MacMahon B, Lindtner J. Epidemiologic features
of breast cancer in Slovenia, 1965-67. Eur J Cancer 1971;7.
295-306.
5. Adami HO, Rimsten A, Stenkvist R, et al. Influence of height,
weight and obesity on risk of breast cancer in an unselected
Swedish population. Br J Cancer 1977;36787-92.
6. Hsieh CC, Trichopoulos D, Katsouyanm K, et al. Age at
menarche, age at menopause, height and obesity as risk factors
for breast cancer associations and interactions in an international case-control study Int J Cancer 1990;46:796-800.
7. Howe GR, Hirohata T, Hislop TG, et al. Dietary factors and
risk of breast cancer: combined analysis of 12 case-control
studies. J Natl Cancer Inst 1990;82561-9.
8. Sellers TA, Kushi LH, Potter JD, et al. Effect of family
history, body-fat distribution, and reproductive factors on the
nsk of postmenopausal breast cancer. N Engl J Med 1992;
326:1323-9.
9. Helmnch SP, Shapiro S, Rosenberg L, et al. Risk factors for
breast cancer. Am J Epidemiol 1983; 117.35-45.
10. World Health Organization. International classification of
diseases: manual of the international statistical classification
of diseases, injuries, and causes of death. Eighth Revision
Geneva, Switzerland: World Health Organization, 1967.
11. Kelsey JL. Epidemiology of musculoskeletal disorders. New
York, NY: Oxford University Press, 1982:145-67.
12. Palmer JR, Rosenberg L. Shapiro S. Stature and the nsk of
myocardial infarction in women. Am J Epidemiol 1990;132:
27-32.
13. Micozzi MS. Nutrition, body size and breast cancer. Yearbook
Phys Anthropol 1985;28:175-206.
14 Gray GE, Pike MC, Henderson BE. Breast-cancer incidence
and mortality rates in different countries in relation to known
nsk factors and dietary practices. Br J Cancer 1979;39:l-7
15. Albanes D, Taylor PR. International differences in body
height and weight and their relationship to cancer incidence.
Nutr Cancer 1990;14:69-77.
16. Tretli S Height and weight in relation to breast cancer morbidity and mortality: a prospective study of 570,000 women in
Norway. Int J Cancer 1989;44:23-30.
17. Swanson CA, Brinton LA, Taylor PR, et al. Body size and
breast cancer nsk assessed in women participating in the
Breast Cancer Detection Demonstration Project Am J Epidemiol 1989;130:1133-41.
18. Tornberg SA, Holm LE, Carstensen JM. Breast cancer risk in
relation to serum cholesterol, serum beta-lipoprotein, height,
weight, and blood pressure. Acta Oncol 1988,27:31-7.
19. Swanson C, Jones DY, Schatzkin A, et al Breast cancer risk
assessed by anthropometry in the NHANES I Epidemiological
Follow-up Study. Cancer Res 1988,48:5363-7.
20. Palmer JR, Rosenberg L, Harlap S, et al. Adult height and risk
of breast cancer among US black women. Am J Epidemiol
1995:141:845-9.
21 Valaoras VG, MacMahon B, Trichopoulos D, et al Lactation
and reproductive histones of breast cancer patients in Greater
Athens, 1965-67 Int J Cancer 1969,4.350-63.
22. Staszewski J Breast cancer and body build. Prev Med 1977;
6.410-15.
23 Mirra AP, Cole P, MacMahon B. Breast cancer in an area of
high parity: Sao Paulo, Brazil. Cancer Res 1971;31:77-83.
24. Bouchardy C, Le MG, Hill C. Risk factors for breast cancer
according to age at diagnosis in a French case-control study. J
Clin Epidemiol 1990;43:267-75.
25. de Waard F, Baanders-van Halewijn EA. A prospective study
in general practice on breast-cancer risk in postmenopausal
women. Int J Cancer 1974;14:153-60.
26. Vatten LJ, Kvinnsland S. Prospective study of height, body
mass index and nsk of breast cancer. Acta Oncol 1992;31.
195-200.
27. Bnnton LA, Swanson CA. Height and weight at various ages
and nsk of breast cancer. Ann Epidemiol 1992;2:597-609
28. Kahsh LA. Relationships of body size with breast cancer. J
Cbn Oncol 1984;2.287-93.
29. Whitehead J, Carhle T, Kopecky KJ, et al. The relationship
between Wolfe's classification of mammograms, accepted
breast cancer risk factors, and die incidence of breast cancer.
Am J Epidemiol 1985;122:994-1006.
30. DeStavola BL, Wang DY, Allen DS, et al. The association of
height, weight, menstrual and reproductive events with breast
cancer results from two prospective studies on the island of
Guernsey (United Kingdom). Cancer Causes Control 1993;4:
331-40.
31. Willett WC, Browne ML, Bain C, et al. Relative weight and
risk of breast cancer among premenopausal women. Am J
Epidemiol 1985;122:731-40.
32. London SJ, Colditz GA, Stampfer MJ, et al. Prospective study
of relative weight, height, and risk of breast cancer. JAMA
1989;262:2853-8.
33. Wynder EL, MacCornack FA, Stellman SD. The epidemiology of breast cancer in 785 United States Caucasian women.
Cancer 1978;41:2341-54.
34. Bnsson J, Morrison AS, Kopans DB, et al. Height and weight,
mammographic features of breast tissue, and breast cancer
risk. Am J Epidemiol 1984;119:371-81.
35. Kolonel LN, Nomura AM, Lee J, et al. Anthropometnc indicators of breast cancer nsk in postmenopausal women in
Hawau. Nutr Cancer 1986;8:247-56.
36. Le Marchand L, Kolonel LN, Earle ME, et al. Body size at
different penods of life and breast cancer risk. Am J Epidemiol 1988;128:137-52.
37. Toti A, Agugiaro S, Amadon D, et al. Breast cancer risk
factors in Italian women: a multicentric case-control study.
Tumon 1986;72:241-9.
38. Ewertz M. Influence of non-contraceptive exogenous and endogenous sex hormones on breast cancer risk in Denmark. Int
J Cancer 1988;42:832-8
39. de Waard A, Tnchopoulos D. A unifying concept of the
aeuology of breast cancer. Int J Cancer 1988;41:666-9.
40. Stewart AL. The reliability and validity of self-reported
weight and height. J Chronic Dis 1982;35:295-309.
41. Palta M, Prineas RJ, Berman R, et al. Comparison of selfreported and measured height and weight. Am J Epidemiol
1982:115:223-30.
Am J Epidemiol
Vol. 143, No. 11, 1996