- © 2007 Canadian Medical Association or its licensors
Nadia Khan and Finlay McAlister1 conclude that β-blockers are efficacious for hypertension in younger but not in older patients, but their conclusions are based on questionable statistical methods.
In their Methods section, Khan and McAlister state that “Meta-analyses for all outcomes were performed using random-effects models.” When I tried to reproduce their results by entering data into Review Manager (the Cochrane Collaboration's software for conducting reviews; version 4.2 for Windows), I also observed a significant reduction of cardiovascular events in younger patients (relative risk [RR] 0.86, 95% confidence interval [CI] 0.74-0.99), but this result was based on a fixed-effects model. With the “true” random-effects model, the CI was wider and included the value 1 (RR 0.86, 95% CI 0.75-1.00). Relative to other antihypertensive drugs, β-blockers seemed to increase the risk of cardiovascular events in older patients (RR 1.06, 95% CI 1.01-1.10), but again in a fixed-effects model. With a random-effects model, the CI includes 1 (RR 1.07, 95% CI 1.00-1.14).
REFERENCE
- 1.↵