To the Editor.
—The article by Dr Stanford and colleagues1 attracted strong media attention to the issue of estrogen replacement therapy (ERT) and breast cancer risk. Although their epidemiological case-control study found no overall risk increase associated with ERT, the authors did observe "[m]odest increased risk estimates... in thin women who used menopausal hormone therapy." Other investigators also have observed elevations in breast cancer risk associated with ERT in lean postmenopausal women.Our literature search for 1985 to 1995 identified six published epidemiological investigations1-6 that provided estimates of breast cancer risk associated with ERT in postmenopausal women and stratified on body mass index (BMI). A meta-analysis of the relative risks (RR) for relatively low BMI is shown in Table 1. The estimates of RR show a narrow range (1.3 to 2.2), and the combined data suggest a 70% increase in breast cancer risk associated with ERT in lean