Abstract
This study explores the possible association between the risk of ischemic stroke and conjugated equine estrogen (CEE) use in women who are over 55 years old and have diabetes. Data from the National Health Insurance system of Taiwan were used to identify 428 women over 55 years old with diabetes who used CEE (0.625 mg daily) from 2003 to 2009. For comparison, 21026 women with diabetes who were from the same cohort and did not use estrogen were used as a control group, excluding patients with previous ischemic stroke at the baseline. The propensity score method was used to identify a 1:3 ratio for the matched cohort (n = 1284). Covariates used for propensity score-matching included age and comorbidities. Cox's proportional hazard model was applied to estimate the relationship between CEE use and ischemic stroke. The overall incidence of ischemic stroke was significantly lower in patients using CEE than in the control group (0.9% compared with 3.0%, p = 0.016). Further analyses using Cox's proportional hazard model revealed that after adjusting for age, comorbidities, socioeconomic status, urbanization, and other medications associated with ischemic stroke, a lower risk was present in patients with CEE use (hazard ratio: 0.34; 95% confidence interval: 0.12-0.97). Time of menopause could not be identified because of the nature of the database. CEE might decrease the risk of ischemic stroke in women with diabetes aged over 55 years, according to this population-based study.