LAUSR

Objective: To examine associations of estrogen preparations with an index of health risks versus benefits. Methods: Using data from 45,112 participants of the Women's Health Initiative Observational Study (average follow-up 5.5 years), we examined associations of estrogen type and oral conjugated equine estrogen (CEE) dose with time to first global index event (GIE), defined as coronary heart disease, breast cancer, stroke, pulmonary embolism, hip fracture, colorectal cancer, endometrial cancer, or death. Results: Oral CEE less than 0.625 mg/d + progestogen (P) users had a lower risk of a GIE (adjusted hazard ratio 0.74, 95% confidence interval 0.56-0.97) than oral CEE 0.625 mg/d + P users. GIE risk in oral CEE 0.625 mg/d + P users was greater with at least 5-year use (adjusted hazard ratio 1.22, 95% confidence interval 1.06-1.41) than with less than 5-year use. In women with prior hysterectomy, compared with women taking oral CEE 0.625 mg/d for less than 5 years, GIE risk was similar with oral CEE below 0.625 mg/d, oral estradiol (E2), and transdermal E2, whether used for less than 5 years or for at least 5 years. There was no difference in GIE risk between users of the following: oral CEE + P versus oral E2 + P; oral CEE + P versus transdermal E2 + P; oral E2 + P versus transdermal E2 + P. Findings were similar among women with hysterectomy taking estrogen alone. Conclusions: The summary index of risks versus benefits was similar for oral CEE versus oral or transdermal E2-containing regimens. CEE + P containing less than 0.625 mg/d of CEE (vs 0.625 mg/d) for less than 5 years appeared safer.