The Million Women Study [1] is in the number of participants until now the largest observational study on the relationship between hormone treatment in the postmenopause and breast cancer. There is, however, only a very short observation time of 2.6 years for incident cases and 4.1 years for fatal cases. The elevated risk of breast cancer among women who are recent hormone users is in agreement with the results of other recent or older studies as well as the more elevated risk in combined treatment with estrogens and progestins compared with estrogen treatment alone. Surprising, however, is the risk elevation already within the first year of combined treatment. This is in disparity to other studies which found an increased risk only after longer treatment, it is also biologically not plausible or could be explained only by a promotion of pre-existing cancers. Even more surprising is that the risk could not be seen anymore already 1 year after stopping treatment. The relative risk was in the prospective randomised WHI-Study [2] in the continuous combined HRT-Group significantly increased (RR /1.24) during observation time of 5.6 years. During estrogen only treatment however, an equivalent increase was not observed until now. In the Million women study the relative risk in the estrogen /progestin group was 2.0 and in the estrogen only group 1.3. It is remarkable that also during treatment with tibolone the risk was elevated 45%. The increased mortality in the Million women study, which, however, is not significant (P /0.05) is in disagreement with the most previous studies. In a review regarding ten studies, a significantly reduced mortality was found in two studies, in five a decreased mortality which was not significant and only in three studies an increased mortality, which, however, was also not significant [3]. In the large prospective Cohort Study of the American Cancer Society (regarding 422 000 postmenopausal women), among the registered 1469 fatal cases a significantly decreased mortality was found of 16% among the hormone users [4]. Which influence has this study on our prescription attitudes? We will discuss the data of this study in comparison to the data of previous studies. Our recommendations since years have been that hormone treatment in the post menopause has to be based on an indication. But there is not another treatment possibility as effective for the estrogen deprived climacteric complaints than the treatment with estrogens or estrogens and progestins and on this indication the risk benefit analysis must be based and our patients have to be informed on both. * Tel.: /49-40-42803-3540; fax: /49-40-42803-2523. E-mail address: braendle@uke.uni-hamburg.de (W. Braendle). Maturitas 46 (2003) 101 /102