PIP: Circulatory disease and endometrial cancer are the 2 main health risks of estrogen therapy. For circulatory diseases, oral contraceptives (OC) had been associated with nonfatal acute mycardial infarction, subarachnoid hemorrhage, and non-embolic cerebral infarction. However, various studies have produced conflicting results, with some studies failing to demonstrate positive association (Boston Collaborative Drug Surveillance Program, 1974; Rosenberg et al, 1976; Pfeffer et al, 1978) and others confirming the association (Jick et al, 1978b; Petiti et al, 1978, 1979) between OC use and circulatory disorders. Confounding by other variables, selection bias or assessment bias may partly explain the disparity in findings. Belsey et al (1979) suggested the results of OC studies carried out in the Western hemisphere are not applicable to other parts of the world due to varying life styles. The differences can also be explained by the fact that OC is used mainly by younger women and since the risk of circulatory diseases increases with increasing age, other risk factors may outweigh any effect low doses of estrogen may have. There is thus no strong evidence that noncontraceptive estrogen use in postmenopausal women is associated with increased risk of circulatory disease at normal dose levels. The same is true for endometrial cancer. Although there are studies reporting strong association between estrogen use and endometrial cancer, other studies have pointed out the various detection biases in these studies. Basic research into the biological mechanisms involved may contribute to the resolution of such problems.