Alcohol, tobacco and breast cancer--collaborative reanalysis of individual data from 53 epidemiological studies, including 58,515 women with breast cancer and 95,067 women without the disease.

Alcohol and tobacco consumption are closely correlated and published results on their association with breast cancer have not always allowed adequately for confounding between these exposures. Over 80% of the relevant information worldwide on alcohol and tobacco consumption and breast cancer were collated, checked and analysed centrally. Analyses included 58,515 women with invasive breast cancer and 95,067 controls from 53 studies. Relative risks of breast cancer were estimated, after stratifying by study, age, parity and, where appropriate, women's age when their first child was born and consumption of alcohol and tobacco. The average consumption of alcohol reported by controls from developed countries was 6.0 g per day, i.e. about half a unit/drink of alcohol per day, and was greater in ever-smokers than never-smokers, (8.4 g per day and 5.0 g per day, respectively). Compared with women who reported drinking no alcohol, the relative risk of breast cancer was 1.32 (1.19-1.45, P<0.00001) for an intake of 35-44 g per day alcohol, and 1.46 (1.33-1.61, P<0.00001) for >/=45 g per day alcohol. The relative risk of breast cancer increased by 7.1% (95% CI 5.5-8.7%; P<0.00001) for each additional 10 g per day intake of alcohol, i.e. for each extra unit or drink of alcohol consumed on a daily basis. This increase was the same in ever-smokers and never-smokers (7.1% per 10 g per day, P<0.00001, in each group). By contrast, the relationship between smoking and breast cancer was substantially confounded by the effect of alcohol. When analyses were restricted to 22 255 women with breast cancer and 40 832 controls who reported drinking no alcohol, smoking was not associated with breast cancer (compared to never-smokers, relative risk for ever-smokers=1.03, 95% CI 0.98-1.07, and for current smokers=0.99, 0.92-1.05). The results for alcohol and for tobacco did not vary substantially across studies, study designs, or according to 15 personal characteristics of the women; nor were the findings materially confounded by any of these factors. If the observed relationship for alcohol is causal, these results suggest that about 4% of the breast cancers in developed countries are attributable to alcohol. In developing countries, where alcohol consumption among controls averaged only 0.4 g per day, alcohol would have a negligible effect on the incidence of breast cancer. In conclusion, smoking has little or no independent effect on the risk of developing breast cancer; the effect of alcohol on breast cancer needs to be interpreted in the context of its beneficial effects, in moderation, on cardiovascular disease and its harmful effects on cirrhosis and cancers of the mouth, larynx, oesophagus and liver.

Estimates of the risk of cardiovascular death attributable to low-dose oral contraceptives in the United States.

OBJECTIVE: Our purpose was to estimate the annual risk of death in the United States from cardiovascular disease attributable to low-dose combination oral contraceptives. STUDY DESIGN: Estimates of the risk of death from cardiovascular disease attributable to low-dose oral contraceptives were modeled on data from studies published through 1997 and from age-specific mortality rates in the United States for 1993 and 1994. RESULTS: Attributable risk of death from cardiovascular disease resulting from oral contraceptive use is 0.06 and 3.0 per 100,000 nonsmokers 15 to 34 years of age and 35 to 44 years of age, respectively. In smokers this risk increases, respectively, to 1.73 and 19.4 per 100,000 users in these 2 age groups; however, 97% and 85% of this risk is due to the combined effects of smoking and using oral contraceptives. The attributable risk of death from cardiovascular disease in nonsmoking oral contraceptive users is lower than the risk of death from pregnancy in nonusers of oral contraceptives at all ages; however, among smoking oral contraceptive users more than 35 years of age, the excess risk of death from oral contraceptives is higher than the risk of death from pregnancy. CONCLUSION: There is virtually no excess attributable risk of death from cardiovascular disease related to oral contraceptive use in young women. However, smokers more than 35 years of age should use a nonestrogen contraceptive.

PIP: The annual risk of death in the US from cardiovascular disease attributable to low-dose combination oral contraceptives (OCs) was estimated through use of data from studies published in 1980-1997 and from age-specific mortality rates for 1993 and 1994. Four cardiovascular disease categories were included: myocardial infarction, venous thromboembolism and pulmonary embolism, ischemic stroke, and hemorrhagic stroke. The overall risk of death from cardiovascular disease among nonsmoking users of low-dose OCs is 0.06/100,000 women in the 15-34 year age group and 3.03/100,000 women in the 35-44 year age group. For young nonsmokers, the excess mortality risk associated with OC use is smaller than the risk of death from pregnancy, whether terminated by abortion or carried to term. Among OC users who smoke, the risk of cardiovascular mortality is 1.73/100,000 in 15-34 year olds and 19.4/100,000 in women 35-44 years old; however, 97% and 85% of this risk, respectively, is composed of the combined OC-smoking risk. Among smoking OC users over 35 years of age, the excess risk of death from OCs exceeds the risk of death from pregnancy. Young nonsmokers raise their risk of death from cardiovascular disease by less than 10% (0.60-0.65/100,000) by using OCs, while young women who do not use OCs increase their risk of death by 260% (0.60-1.57/100,000) by smoking cigarettes. For older women, the corresponding increases are 95% among nonsmoking OC users and 315% among smoking nonusers. These estimates indicate that women over 35 years of age who smoke should not be permitted to use either low- or high-dose OCs because of the excess attributable risk of death from cardiovascular disease.

Cardiovascular safety of oral contraceptives. What has changed in the last decade?

PIP: As a result of careful patient selection, the cardiovascular safety of oral contraceptives (OCs) has improved dramatically in the past decade. The incidence of stroke and myocardial infarction is exceedingly low among women who use OCs containing 35 mcg or less of ethinyl estradiol, and formulations containing under 50 mcg of estrogen account for almost all current use in the US. This article reviews the epidemiologic data on use of OCs of varying steroid dosages on the risks of myocardial infarction, hemorrhagic and ischemic stroke, and venous thromboembolism. Although four studies published since 1995 have suggested that OCs containing desogestrel or gestodene increase the risk of venous thromboembolism above that associated with levonorgestrel, these findings are likely due to prescribing bias and differences in the duration of use. The greatest risk of an arterial cardiovascular event comes from smoking while taking OCs.^ieng

Epidemiology of venous thromboembolic disease and OC use.

PIP: A review of recent epidemiologic studies that have detected an association between the use of oral contraceptives (OCs) containing the progestins gestodene and desogestrel and venous thromboembolism (VTE) risk suggests evidence of bias. Reviewed are five major case-control and cohort studies: World Health Organization Collaborative Study, Boston Collaborative Drug Surveillance Program Study, European Transnational Study, and the Leiden Study. Three major sources of bias could account for the increased VTE risk among users of third-generation compared to second-generation OCs: 1) selective prescription of newer formulations to higher-risk women; 2) the increased tendency for women with suspected VTE to be more likely to be referred for diagnostic testing and hospitalization if they are taking the newer OCs rather than older formulations; and 3) attrition of susceptibles. The lack of any proposed biological basis for the observed association between VTE and the new progestins, compared with previous knowledge about the responsibility of estrogen for increased VTE risk, raises additional doubts about the findings. Any evaluation should balance the effects of OCs on overall risk of cardiovascular disease against protection from pregnancy and noncontraceptive health benefits such as a reduced risk of certain cancers.^ieng

CDC WONDER: a comprehensive on-line public health information system of the Centers for Disease Control and Prevention.

OBJECTIVES: CDC WONDER, a comprehensive on-line public health information system of the Centers for Disease Control and Prevention (CDC), was developed to place timely, action-oriented information in the hands of public health professionals. METHODS: A unified system was developed de novo to be used for and to evolve along with public health. All data are stored and updated on the CDC mainframe. RESULTS: CDC WONDER provides menu-driven access to 24 databases with information on mortality, hospital discharges, cancer incidence, notifiable diseases, acquired immunodeficiency syndrome, the Morbidity and Mortality Weekly Report, etc.; each database has on-line documentation. Results can be tabulated and graphed, and there is full-text searching of textual databases. Non-CDC staff have access via telephone connection. From August 1991 through June 1992, system databases were accessed 10,698 times, and there were 842 users (mean of 97 new users per month). CONCLUSIONS: CDC WONDER has shown that it is possible to build a large, on-line database of scientific data for public health professionals. CDC WONDER provides a common foundation from which to build information-based public health plans and policy and could help strengthen the public health system.

Age-specific differences in the relationship between oral contraceptive use and breast cancer.

BACKGROUND: Nearly all studies have suggested that the use of oral contraceptives (OC) is not associated with the aggregate risk of breast cancer diagnosed in women aged 20-54 years. Because of age-specific differences in the breast cancer-parity relationship and because of age-specific differences in other breast cancer risk factors, the Centers for Disease Control reexamined data from the Cancer and Steroid Hormone Study (CASH) to assess whether OC use has different effects on the risk of breast cancer at different ages of diagnosis. METHODS: This population-based case-control study was designed to examine the relationship between the use of OC and the risk of breast, ovarian, and endometrial cancer. CASH was conducted in eight geographic areas in the United States during 1980-1982. All participants were interviewed at home with a pretested standardized questionnaire including a calendar of life events and a photograph book of all pills marketed in the United States. RESULTS: We found that the relationship between the risk of breast cancer and OC use appeared to vary by the age at diagnosis. Among women aged 20-34 years at diagnosis or interview, those who had ever used OC had a slightly increased risk of breast cancer (odds ratio [OR], 1.4; 95% confidence interval [CI], 1.0-2.1) compared with women of the same ages who had never used OC. Among these women, there were no trends of increasing or decreasing risk with any measure of OC use. Among women aged 35-44 years, there was no association between OC use and breast cancer. Among women aged 45-54 years, those who used OC had a slightly decreased risk of breast cancer (OR, 0.9; 95% CI, 0.8-1.0). Among these women, risk estimates decreased significantly with increasing time since first and last use. CONCLUSIONS: Although the slightly increased risk estimates for the youngest women were compatible with findings by other investigators, the decreased risk estimates for the oldest women have not been described in as many studies. Available data provide no reasons to change prescribing practices or the use of OC that are related to the breast cancer risk.

Age-specific differences in the relationship between oral contraceptive use and breast cancer.

Nearly all studies have suggested that the use of oral contraceptives (OC) is not associated with the aggregate risk of breast cancer diagnosed in women aged 20-54. Because of age-specific differences in the breast cancer-parity relationship and because of age-specific differences in other breast cancer risk factors, the Centers for Disease Control reexamined data from the Cancer and Steroid Hormone Study to assess whether OC use has different effects on the risk of breast cancer at different ages of diagnosis. This was a population-based case-control study conducted in eight geographic areas in the United States during 1980-1982. In these data, the relationship between the risk of breast cancer and OC use appeared to vary by age at diagnosis. Among women aged 20-34 years at diagnosis or interview, those who had ever used OC had a slightly increased risk of breast cancer (odds ratio 1.4, 95% confidence interval 1.0-2.1) when compared with women of the same ages who had never used OC. Among these women, there were no trends of increasing or decreasing risk with any measure of OC use. Among women aged 35-44 years, there was no association between OC use and breast cancer. Among women aged 45-54 years, those who used OC had a slightly decreased risk of breast cancer (odds ratio 0.9, 95% confidence interval 0.8-1.0). Among these women, the risk estimates decreased significantly with increasing time since first and last use. Although the slightly increased risk estimates for the youngest women are compatible with findings by other investigators, the decreased risk estimates for the oldest women have not been described in as many studies. Available data provide no reasons for changes in prescribing practices or in the use of OC as related to breast cancer risk.

The independent associations of parity, age at first full term pregnancy, and duration of breastfeeding with the risk of breast cancer. Cancer and Steroid Hormone Study Group.

Although the important influence of a woman's reproductive history on her risk of breast cancer is widely recognized, it is not clear whether this is wholly accounted for by the age at her first full-term pregnancy, or whether there are additional, independent influences of breastfeeding or number of children. To examine the respective contributions to the risk of breast cancer of these reproductive factors, we used logistic regression methods to analyze data from a multicenter case-control study, the Cancer and Steroid Hormone Study. Included in the analysis were 4599 women, 20-55 years of age, identified as having an initial diagnosis of breast cancer by one of eight collaborating population-based cancer registries. The 4536 controls were women of similar ages selected by random dialing of households with telephones in the same eight areas. As expected, age at first full-term pregnancy exerted a strong influence on the risk of breast cancer. However, after it and other potentially confounding factors had been controlled for, parity and duration of breastfeeding also had a strong influence on the risk of breast cancer. Compared with women of parity one, women of parity seven or greater had an adjusted relative risk of breast cancer of 0.59 (95% CL, 0.44-0.79). Compared with parous women who never breastfed, women who had breastfed for 25 months or more had an adjusted relative risk of 0.67 (0.52-0.85). These results do not support the supposed preeminent importance of age at first full-term pregnancy among the reproductive determinants of breast carcinogenesis. Resolution of this issue may have important implications for elucidating hormonal influences on breast cancer and for projecting future trends in the disease.

The evaluation of the data collection process for a multicenter, population-based, case-control design.

This report details the methods the authors used to conduct the Cancer and Steroid Hormone Study, a multicenter, population-based, case-control study of oral contraceptive use in relation to breast, endometrial, and ovarian cancer diagnosed during 1980-1982. The authors have documented their methods and rationale, and the results of their data collection efforts as a practical guide for the planning and conduct of large case-control studies. They observed the following: 1) the Surveillance, Epidemiology, and End Results program is a useful epidemiologic resource for identifying cases from which to evaluate risk factors for cancer in the United States; 2) random digit dialing is an effective and efficient method for screening for eligible controls for a population-based study; 3) with the cooperation of community pathologists, histologic specimen slides can be retrieved and reviewed for diagnostic confirmation and histologic subclassification of cancer for greater than 95% of the cases interviewed; and 4) data reported during personal interviews of study participants can be validated by reviewing medical records for more than 75% of study participants who reported medical events that occurred during the 10 years before the beginning of the study.

PIP: This study, a detailed report of the methods used to conduct the Cancer and Steroid Hormone study, is presented in order to aid other researchers in designing and carrying out multicenter case-control studies. Data collection for the study was carried out by the Division of Reproductive Health of the Centers for Disease Control in 8 locations: Atlanta, Detroit, San Francisco, Seattle, Connecticut, Iowa, New Mexico, and Utah. Case ascertainment, interviewing, and retrieval of pathology slides and medical records were carried out by the Surveillance, Epidemiology, and End Results centers of the National Cancer Institute. Cases were women, aged 20-54, newly diagnosed for breast, ovarian or endometrial cancer between December 1, 1980, and December 31, 1982 -- 20 years after the 1st sales of oral contraceptives in the US. Controls were women aged 20-54, selected by random digit dialing of households with telephones. 1 control was selected for each breast cancer case of matching age and geographic location. Interviewers were trained and administered standardized questionnaires to determine medical, reproductive, and oral contraceptive use histories. To aid in accuracy of recall, respondents recorded dates of major life events in a women's health study calendar. They were also given color photographs of all oral contraceptives marketed in the US. Field personnel collected slides of ovarian and endometrial cancers. Slides of benign breast lesions reported by both cases and controls were collected, as were medical records of women who reported diagnoses of infertility or loss of one or both ovaries. 80% of the breast cancer cases, 70% of the ovarian, and 74% of the endometrial cancer cases were interviewed, as were 83% of the controls. Only 2%, 7%, and 6% of the interviews for the breast, ovarian, and endometrial cancers respectively failed to match the Surveillance, Epidemiology, and End Results data. Retrieval for expert review of the pathology slides of recently diagnosed ovarian or endometrial cancer was 97% and 95%. Retrieval of pathologic proof of benign breast disease, infertility diagnoses, and ovary removal was high for the period after 1970 but low for conditions reported to have occurred before 1950. The study concludes that Surveillance, Epidemiology, and End Results tumor registry records are an excellent source for cancer epidemiology data; that random digit dialing is an effective source of controls; that histologic specimen slides can be retrieved to confirm over 95% of cancer cases interviewed; and that data reported during interviews can be substantiated by medical records for over 75% of study participants.

The risk of breast cancer in postmenopausal women who have used estrogen replacement therapy.

We studied the association between estrogen replacement therapy (ERT) and the risk of breast cancer as part of the Cancer and Steroid Hormone Study. All subjects in the analysis were postmenopausal women enrolled from eight geographic areas. Women 25 to 54 years old with newly diagnosed breast cancer were identified through population-based tumor registries and diagnosed between Dec 1, 1980, and Dec 31, 1982. Controls were selected from the same eight geographic areas by the random digit dialing of residential telephone numbers. Analyses included 1369 cases and 1645 controls. Among women with bilateral oophorectomy, the relative risk of breast cancer for women who had ever used ERT was 1.3, compared with women who had never used ERT. Among women who had undergone hysterectomy but who still had at least one ovary, the relative risk was 1.1; among women who reported a natural menopause, the relative risk was 0.8. Overall, the risk of breast cancer did not appear to increase appreciably with increasing ERT duration or latency, even for durations and latencies of 20 years or longer.

Assessment of screening as a preventive technology: the example of HTLV-III/LAV antibody testing.

Screening for disease control can be defined as a preventive technology that is used to examine asymptomatic people in order to classify them as likely or unlikely to have the disease that is the object of screening. Screening may consist of routine physical examinations, radiologic procedures, semi-invasive procedures such as endoscopy, or serologic tests. In this paper, a new serologic test is considered. Acquired immunodeficiency syndrome (AIDS) is a devastating disease with high mortality, recently shown to be caused by a retrovirus named human T-lymphotropic virus Type III (HTLV-III) or lymphadenopathy-associated virus HTLV-III antibody in serum specimens. Performance characteristics are excellent with high sensitivity and specificity when reactive serum specimens are checked for consistency of response by enzyme immunoassay (EIA). This test is now part of the screening protocol for all blood donation centers to decrease the risk of HTLV-III transmission via blood or blood products. About 0.2% (1 in 400) of blood donors have repeatedly reactive EIA tests to HTLV-III antibody. Approximately one-third of these donors have other laboratory evidence of infection. Screening for HTLV-III is a new technology that illustrates virtually all of the factors that need consideration in an assessment of disease screening. This paper explores these technical, epidemiologic, economic, legal, social, and ethical factors.

The mortality risk associated with hysterectomy.

To study the risks of mortality associated with hysterectomy that are specific to age, race, surgical approach, and associated conditions, we used data collected by the Commission on Professional and Hospital Activities during 1979 and 1980. Four hundred seventy-seven deaths were recorded among 317,389 women having abdominal hysterectomies and 46 deaths among 119,972 women having vaginal hysterectomies. The mortality rates for hysterectomy, standardized for age and race, were higher for procedures associated with pregnancy or cancer than for procedures not associated with these conditions (29.2, 37.8, and 6.0 per 10,000 procedures, respectively). Hysterectomies associated with pregnancy or cancer constituted 8% of all hysterectomies performed. However, 61% of all deaths occurred in women with pregnancy- or cancer-related conditions. The mortality rate associated with hysterectomy increased with age and was twice as high among black women.

Family history and the risk of breast cancer.

To investigate whether a family history of breast cancer increases a woman's risk of developing breast cancer, we analyzed data from the Centers for Disease Control's Cancer and Steroid Hormone Study. The 4,735 cases were women 20 to 54 years old with a first diagnosis of breast cancer ascertained from eight population-based cancer registries; the 4,688 controls were women selected at random from the general population of these eight areas. Compared with women without a family history of breast cancer, women who had an affected first-degree relative had a relative risk of 2.3; women with an affected second-degree relative had a relative risk of 1.5; and women with both an affected mother and sister had a relative risk of 14. The risk of breast cancer for a woman was higher if her first-degree relative had unilateral rather than bilateral breast cancer or had breast cancer detected at a younger rather than older age. For women aged 20 to 39, 40 to 44, and 45 to 54 years, the estimated annual incidence of breast cancer per 100,000 women attributable to a first-degree family history of breast cancer was 51.9, 115.1, and 138.6, respectively, and that attributable to a second-degree family history of breast cancer was 12.1, 19.2, and 92.4, respectively.

Endometrial cancer: how does cigarette smoking influence the risk of women under age 55 years having this tumor?

This analysis of the Cancer and Steroid Hormone Study, a multicenter, population-based case control investigation of hormone use by women of reproductive age and endometrial, breast, and ovarian cancer shows that cigarette smoking is not associated with either an increased or a decreased risk of endometrial cancer. This study included 437 women with endometrial cancer and 3200 control subjects, all of whom were between the ages of 20 and 54 years at the time of interview. The absence of any alteration of the risk of endometrial cancer and smoking was found consistently no matter which variable was used as a measure of smoking--ever or never smoked cigarettes, former or current smoking, light or heavy smoking, or age smoking began.

Confirmation of the preoperative diagnoses for hysterectomy.

Few previous studies have examined the relationship between the preoperative and pathologic diagnoses for hysterectomy. To determine the percentage of preoperative diagnoses that were confirmed by pathologic examination, we analyzed data from the Collaborative Review of Sterilization, a multicenter study of hysterectomies and tubal sterilizations in women aged 15 to 44 years. Data were collected from patient interviews and chart reviews. Of the 1851 women included in this study, 1283 (69%) had abdominal hysterectomies and 568 (31%) had vaginal hysterectomies. Overall, 52% of the hysterectomies were performed for a preoperative diagnosis that could potentially be confirmed by pathologic examination. Pathologic examination actually confirmed the preoperative diagnosis of endometrial hyperplasia in 95% of the cases, cervical intraepithelial neoplasia in 89%, leiomyomas in 84%, pelvic inflammatory disease in 75%, adenomyosis in 48%, and endometriosis in 47%. Among all of the potentially confirmable diagnoses, 80% were confirmed. The remaining 48% of the women who had hysterectomies had preoperative diagnoses that were not amenable to confirmation by pathology. Most of these were for one of three diagnoses: menstrual bleeding disorders, pelvic pain, or pelvic relaxation. In 47% of these cases, pathologic examination showed leiomyoma or adenomyosis; no abnormalities were found in 38% of these cases.

Tubal sterilizations performed in freestanding, ambulatory-care surgical facilities in the United States in 1980.

In 1981 the American Association of Gynecologic Laparoscopists and the Division of Reproductive Health, Centers for Disease Control, jointly conducted a study of tubal sterilizations performed in 141 freestanding, ambulatory-care surgical facilities in 1980 in the United States. Information was collected through mailed questionnaires and telephone interviews. Of 330 potential responding facilities, 141 we identified as freestanding, ambulatory-care surgical facilities. About 16,500 tubal sterilizations were performed in these facilities in 1980. The mean number of tubal sterilizations per freestanding, ambulatory-care surgical facility was 212. Sixty-seven percent of tubal sterilizations were performed in the south and west. General anesthesia was the anesthetic method used in 97% of the procedures. Nearly 91% of tubal sterilizations were done via laparoscopy, with bipolar electrocoagulation the tubal-occlusion method used most frequently. After tubal sterilization the patients were observed for an average of 2.4 hours before discharge. The average cost of laparoscopic tubal sterilization was $801; for nonlaparoscopic tubal sterilization it was $850.

Mortality among young black women using contraceptives.

Data on the risk of death associated with various contraceptive methods are incomplete. Therefore, we analyzed the mortality rates for young, black inner-city women who used one of four methods of contraception--oral contraceptives, depomedroxyprogesterone acetate, intrauterine (contraceptive) devices, and barrier methods. The subjects were 30,580 15- to 44-year-old women who enrolled at a family planning clinic between 1967 and 1972 and who were observed by monitoring death certificates through the end of 1977. Forty percent of the 218 deaths observed were from accidents and violence. Use of this family planning clinic greatly reduced the risk of death from childbearing; only two deaths were associated with pregnancy and childbirth, compared with the 24 deaths expected. Overall, users of the four methods died at similar, low rates. Given that this study involves considerable loss to follow-up, possible acute effects of contraceptives (eg, infections or thrombosis) are more accurately estimated than possible long-term effects (eg, cancer).

Menstrual changes after tubal sterilization.

The authors used data from a multicenter prospective study of female sterilization surgery to study changes in menstrual function following tubal sterilization. Duration of menstrual bleeding, menstrual cycle length, cycle regularity, amount of menstrual bleeding, menstrual pain, and intermenstrual bleeding were examined. The authors followed 2456 women for two years after tubal sterilization surgery. Each woman served as her own control; her menstrual function at the two-year follow-up interview was compared with her menstrual function at the preoperative interview. Except for menstrual pain among women who underwent unipolar electrocoagulation procedures, there was no increase in the prevalence of adverse menstrual function after tubal sterilization. For all menstrual variables, 50% or more of women with adverse function preoperatively had an improvement by two years after tubal sterilization.

Alcohol consumption and risk of breast cancer.

To investigate whether women who consume alcoholic beverages have a greater risk of breast cancer than women who never drink data from a population-based, case-control study, the Centers for Disease Control's Cancer and Steroid Hormone Study, were examined. During the first 15 months of data collection, 1226 women aged 20--54 with newly diagnosed breast cancer and 1279 women of the same age randomly selected from the general population were interviewed. Women who consumed alcoholic beverages had no greater risk of breast cancer than non-drinkers (relative risk 1.0, 95% confidence interval 0.8--1.2). Breast-cancer risk was not associated with the average amount of alcohol consumed per week nor with the type of alcoholic beverage consumed. Compared with non-drinkers, the relative risks of breast cancer for women who ever drank beer, wine, or spirits were 1.0, 0.8, and 0.9, respectively.