Studies must determine the evidence

EDITOR—As active members of the Australasian Menopause Society, we are disappointed at the conclusions that Dixon drew in his editorial on hormone replacement therapy and the breast.1 Although it may be true that hormone replacement therapy makes mammograms harder to interpret, it is far from clear that it causes breast cancer.

A recent overview by Bush et al emphasises the weakness of Dixon's argument, based, as it is, almost entirely on level three observational studies.2 Unlike Dixon's selection of studies with the highest odds ratio, Bush et al's review was of 45 studies assessing the association between use of hormone replacement therapy and risk of breast cancer. It found that risk was reduced (relative risk<0.9) in 20% of the studies, did not change in 47% (0.9-1.1), and increased in 33% (1.1-2.0). In no study did relative risk increase above 2.0, and in the 20 studies where the relation between risk of breast cancer and combined oestrogen and progestin therapy was studied only four reported a significant difference in relative risk, with two showing an increased and two a decreased risk.

The heterogeneity of these data is in stark contrast to the homogeneity of the data on mortality from breast cancer in users of hormone replacement therapy that were reviewed: all 11 of the studies reported a reduction in risk. Unlike Dixon, the authors concluded that the likelihood of an adverse effect of hormone replacement therapy on breast cancer must be small.

The Australasian Menopause Society is a …