Thoughts on individualizing hormone replacement therapy based on the postmenopausal health disparities study data.

Abstract

This study presents an approach that can be used to address the controversy about the long-term risks and benefits of hormone replacement therapy (HRT). Categories of estradiol (E(2)) levels based on the mean and standard deviation (SD) in postmenopausal women not treated with HRT were created. E(2) levels achieved in women treated with oral or patch replacement therapy were examined. In the application of this method, data from the Postmenopausal Health Disparities Study were used because identical E(2) assay methods were used for all postmenopausal women in the study population. The findings demonstrated that approximately 42% of HRT-treated women had E(2) levels less than the mean E(2) level plus 1 SD in the controls (37 pg/ml). In contrast, approximately 11% had E(2) levels greater than the mean E(2) level plus 6 SD in the controls (126 pg/ml). Menopausal vasomotor and vaginal symptoms decreased ( p < 0.06) in increasing control-based E(2) categories. Levels of both follicle-stimulating hormone (FSH) and luteinizing hormone (LH) were significantly correlated with the E(2) control-based categories (both p = 0.000). When hormone variables and other variables (e.g., moderate alcoholic beverage consumption and bilateral ovariectomy) known to be determinants of E(2) levels in postmenopausal women not treated with HRT were included in the model for oral or patch HRT-treated women, 51.1% of the variability in the control-based E(2) categories was explained when menopause duration, use of oral as opposed to patch therapy, body mass index (BMI), scores for medical compliance, and being white entered the equation as statistically significant predictors. First, there is substantial variability in E(2) concentrations achieved with conventional oral and patch HRT. Second, various factors that are statistically significant predictors of response categories have been identified. Third, race/ethnic group plays a role in the response to HRT. Finally, the demonstration of variable responses to HRT and the application of the categorization paradigm presented will be useful in resolving the controversy about the risks and benefits of conventional HRT.