Women’s Health Initiative Clinical Trials: The Effects of Calcium and Vitamin D Supplementation and Hormone Therapy on LDL-Cholesterol and Other Cardiovascular Disease Risk Factors
Abstract
Objective: To analyze the treatment effect of calcium and vitamin D (CaD) supplementation, hormone therapy (HT), both CaD and HT, and neither on cardiovascular disease (CVD) risk factors. Design: A prospective, randomized, double-blind, placebo controlled trial among Women’s Health Initiative postmenopausal participants. In the HT trial a total of 27,347 women were randomized to HT (0.625 mg/d of conjugated equine estrogens [CEE] alone or 0.625 mg of CEE plus 2.5 mg of medroxyprogesterone acetate [MPA] daily) versus placebo. In the CaD trial, 36,282 women were randomized to 1,000 mg of elemental calcium carbonate plus 400 IU of vitamin D3 daily, compared with placebo. A total of 1,521 women participated in both trials and were in the 6% subsample of trial participants with repeated blood sample collections at baseline and years 1, 3, and 6. The predefined primary outcome of this analysis was low-density lipoprotein cholesterol (LDL-C) and other CVD risk factors (such as high-density lipoprotein cholesterol, triglycerides, glucose, blood pressure, weight, and waist circumference) as secondary outcomes reporting the average effect over both visits that occurred after CaD randomization. Results: The average treatment effect with 95% confidence intervals during follow-up, in LDL-C, compared to placebo, was −1.6 (−5.5, 2.2) mg/dL for CaDalone, −9.0 (−13.0, −5.1) mg/dL for HT-alone, and −13.8 (−17.8, −9.8) mg/dL for CaD plus HT. The p-value for interaction was 0.26, indicating no evidence of a synergistic effect of CaD x HT on LDL-C. However, there was evidence that CaD x HT had a synergistic effect on decreasing LDL at low total intakes (dietary and supplements) of vitamin D (p-int = 0.03) and calcium (p-int = 0.06). The effect of HT plus CaD, on all of the remaining CVD risk factors, tended to be larger in magnitude than the effects of either HT-alone or CaD-alone, but none of the HT x CaD interactions were statistically significant. Conclusion: The average reductions in LDL-C during follow-up were greater among women randomized to both CaD and HT than for those randomized to either intervention alone or to placebo. However, statistical tests for interaction between CaD and HT were non-significant. Average Group Means (95% CI) during follow-up of Treatment Group Effects