The Million Women Study set out to recruit a million women between 1996 and 2001 from the United Kingdom National Health Service breast-screening program, where mammograms are done every 3 years. Approximately 50% of the patients surveyed used postmenopausal hormone therapy; the data were obtained from these individuals by filling out a questionnaire before having their mammogram. The women were monitored subsequently for breast cancer occurrence and the relationship to the use of hormone therapy. As with previous studies, no increase in risk of breast cancer was noted with past usage of any hormone preparation when the period of discontinuation exceeded 1 year, regardless of the duration of previous use. The average follow-up was 2.6 years, and unlike previous studies, even those patients taking only estrogen had an increased risk ratio of 1.30. The risk ratio for estrogen-progestin users was 2.00 and for past users was 1.01. The type of paration used did not seem to make a difference. Similar results were obtained with all products used and all doses.

After analyzing this cohort study, the authors estimate that the past decade of use of hormone therapy resulted in an extra 20,000 cases of breast cancer in the United Kingdom. It is truly unclear how they arrived at this figure. In the United States, the incidence of breast cancer had increased 4% per year through the years 1980 to 1988; this was attributed to increasing detection because of the widespread use of mammography. Since 1988, the incidence rate has been very stable, with less than 1% increase per year. When breast cancer mortality was analyzed in the Million Women Study, there were no statistically significant increases, with a total of 517 deaths recorded from breast cancer. There is a trend in that current users had a risk ratio of 1.22. This trend suggests an increase in breast cancer mortality with the use of hormones, which is the opposite of more than 90% of previous cohort studies that show a decrease in breast cancer mortality with hormone use. One cannot help but wonder whether this is in some way related to the fact that mammography is done every 3 years in the United Kingdom as opposed to annually in the United States. The authors also conclude that the use of hormone therapy reduces the sensitivity of the mammogram and increases the diagnosis of interval cancers. This also is inconsistent with previous studies.

The results of the Million Women Study also disagree with the Health Initiative, which has reported that an increase in breast cancer is not associated with the estrogen-only arm of that clinical trial. This study has to be taken seriously because of its size, but most of the results are inconsistent with other, similar reports. In general, the underlying reason for disagreement among studies appears to be related to the minimal influence that hormone therapy has on the incidence of breast cancer. The increased risk, if it exists, lies too close to the baseline. Patients who are considering hormone therapy have to understand this, weigh the benefits as well as the risks, and make their decisions appropriately.