A simulation using data from a primary care practice database closely replicated the women's health initiative trial

Area of Science:

• Cardiovascular Disease Epidemiology
• Pharmacological Research
• Clinical Trial Methodology

Background:

• Observational studies suggested hormone replacement therapy (HRT) might protect against coronary heart disease.
• The Women's Health Initiative Randomized Controlled Trial (WHI RCT) contradicted these findings, showing no cardiovascular benefit.

Purpose of the Study:

• To validate a novel observational study design by comparing it with the WHI RCT.
• To assess HRT's effect on coronary heart disease using a simulated RCT approach.

Main Methods:

• A cohort from the UK General Practice Research Database (GPRD) was used to simulate the WHI RCT.
• The simulation replicated RCT aspects, excluding randomization, for 37,730 Unexposed and 13,658 Exposed women on estrogen and norgestrel.

Main Results:

• HRT did not significantly decrease myocardial infarction in the simulated GPRD cohort (HR 0.95 [0.78-1.16]).
• Similar to the WHI RCT, HRT increased stroke, venous thromboembolic events, and breast cancer, while decreasing colorectal cancer.
• The apparent decrease in total mortality was not observed in a subset without missing data.

Conclusions:

• The GPRD-based simulation largely agreed with WHI RCT findings, showing no cardioprotective effect of HRT.
• These results generalize WHI findings and support the utility of this simulation approach for epidemiological research.