Background: Despite the effect of lowering low-density lipoprotein cholesterol (LDL-C) levels and raising high-density lipoprotein cholesterol (HDL-C) levels, combination hormone therapy did not reduce the incidence of coronary heart disease (CHD) events in the Heart and Estrogen/progestin Replacement Study (HERS). To explore possible mechanisms, we examined the association between ipid changes and CHD outcomes among women assigned to hormone therapy. Methods: HERS participants were postmenopausal women with previously diagnosed CHD who were randomly assigned to receive conjugated estrogens and medroxyprogesterone or identical placebo and then followed-up for an average of 4.1 years. Among women assigned to hormone therapy, associations between baseline-to-year-1 lipid level changes and CHD events were compared with the associations observed for baseline lipids using multivariate proportional hazards models. Results: Among women assigned to hormone therapy, CHD events were independently predicted by baseline LDL-C levels (relative hazard [RH] 0.94 per 15.6 mg/dL decrease, 95% CI 0.88-1.01) and HDL-C levels (RH 0.89 per 5.4 mg/dL increase, 95% CI 0.81-0.99), but not by triglyceride levels (RH 1.01 per 13.2mg/dL increase, 95% CI 0.97-1.06). CHD events were marginally associated with first-year reductions in LDL-C levels (RH 0.95 per 15.6mg/dL decrease, 95% CI 0.86-1.04), and were not associated with increases in HDL-C levels (RH 1.03 per 5.4 mg/dL increase, 95% CI 0.91-1.16) or triglyceride levels (RH 1.01 per 13.2 mg/dL increase, 95% CI 0.98-1.05). Conclusion: Changes in lipid levels with hormone therapy are not predictive of CHD outcomes in women with heart disease in the HERS trial.