Adherence-adjusted efficacy with intensive versus standard statin therapy in patients with acute myocardial infarction in the IDEAL study

BACKGROUND: The Incremental Decrease in End Points through Aggressive Lipid Lowering trial showed that the primary endpoint major coronary event was reduced by 11% (0.78-1.01) using atorvastatin 80 mg versus simvastatin 20-40 mg in patients with coronary heart disease (P=0.07). Adherence was high in both treatment groups but significantly higher in patients treated with simvastatin. DESIGN: The Incremental Decrease in End Points through Aggressive Lipid Lowering was a prescription trial with a prospective randomized open label endpoint evaluation. METHODS AND RESULTS: Adherence was calculated as exposure time on prescribed drugs divided by total follow-up time until death or end of follow-up and was a potential confounder. Adjusting for categorical adherence below or above 80% by two methods revealed that the relative risk reduction of the primary endpoint was more in the region of 15% (P=0.02) than 11% as found unadjusted. Censoring at the first occurrence of a cardiovascular event rather than at death increased this estimate to 17% (P=0.02). Noncardiovascular mortality was reduced on atorvastatin treatment by 21% (1-37%) after adjustment for adherence, whereas such reduction was not observed for cardiovascular mortality. CONCLUSION: This study found that the difference in adherence between treatment groups may have underestimated the true effect of the treatment differential. Usage of prospective randomized open label endpoint evaluation design should be carefully considered when well-known treatments are compared with rather new ones and especially in segments where patients could be more vulnerable, as in the elderly. Nonadherers in a clinical trial may be at especially high risk of fatal and nonfatal endpoints from various diseases and should be carefully monitored.