Table 5 presents risk factors for breast cancer generally regarded as established, together with their approximate relative risks. With the exception of age, country of birth, and a history of breast cancer in both a mother and a sister, all of the relative risks reported to date are of a relatively modest magnitude. Thus, new risk factors need to be identified and knowledge of existing risk factors refined. Factors for which the evidence of an etiologic role has mounted over the past several years, but which are not yet considered to be established, include the protective effects of parity and lactation in certain age groups and the increased risks associated with alcohol consumption and with DES exposure during pregnancy. In addition, physical activity has emerged as a factor worthy of further study. Some evidence suggests that use of oral contraceptives for several years at an early age modestly increases the risk for breast cancer diagnosed before age 35 and perhaps age 45. Use of estrogen-replacement therapy for 20 years or more has been found by a few studies to increase the risk for breast cancer in the postmenopausal years; further studies of very long-term users are needed. Also, other risks and benefits of these hormones need to be taken into account when women decide whether to use them. Surprisingly elusive has been the etiologic role of endogenous hormones, especially in view of the large number of studies that have been concerned with them. A better understanding of the role of endogenous hormones should help explain the mechanisms of action of known and suspected risk factors. Areas of high priority for further research thus include establishing with more certainty whether the risk for breast cancer is increased in any subgroups of women who use oral contraceptives and estrogen-replacement therapy and determining the etiologic roles of specific endogenous hormones. The possible risks associated with alcohol consumption and lack of physical activity need to be studied more thoroughly, and ideas about new potential risk factors are needed. Although epidemiologic studies will continue to be concerned with diet, enthusiasm for its etiologic role in women has been considerably dampened by the lack of association in many of the studies reported to date. The studies in women exposed to radiation, DES, and oral contraceptives suggest that the timing of some exposures may be critical, since the effects of these agents may mostly be limited to specific time periods of rapid breast development.(ABSTRACT TRUNCATED AT 400 WORDS)