BACKGROUND: Many clinicians believe that hormone replacement therapy (HRT) protects against coronary heart disease (CHD) in women. However, recent reports have cast some doubt on this because of lack of dose-response or duration-response effects. Since CHD is common in women--about half of all postmenopausal women will get it and about a third of these will die from it--the effect of HRT on CHD is of great public health importance. AIM: To determine the degree of cardioprotection conferred by HRT, including the effect of duration, time since last issue, the addition of progestogens, route of administration, and dose. DESIGN: Population-based case-control study. SETTING: Nine general practices recruited from the Trent Focus Collaborative Research Network. METHOD: A total of 417 female cases with CHD matched by age and practice to 2435 controls with a case-control ratio of 1:5.8 were studied. The main outcome measure was the odds ratio for CHD calculated by conditional logistic regression adjusted for diabetes, hypertension, body mass index, and smoking. RESULTS: No evidence was found, either from univariate analysis or multivariate analysis, that use of HRT was associated with reduced risk of CHD (odds ratio = 1.32; 95% confidence interval = 0.93 to 1.87). Indeed, the trend was in the opposite direction. There was no association for different types of HRT (opposed or unopposed) or routes of administration. Similarly, there was no association for current or past use and no effect for dose or duration. CONCLUSION: This study adds to growing evidence that HRT does not confer cardioprotection. Until there is robust evidence to the contrary, general physicians need to assess risks and benefits of HRT independently of any possible reduction in risk of CHD.