|| Checking for direct PDF access through Ovid
Clinical trials have demonstrated that medical therapy can reduce mortality in patients with acute myocardial infarction (MI), but variations in drug adherence mean that these survival benefits do not always translate to real-world patient populations.To ascertain whether adherence to medication reduces long-term mortality in patients who have experienced an acute MI.This was a population-based, observational cohort study of patients aged 66 years and older who had been hospitalized for an acute MI in Ontario, Canada, and who survived at least 15 months following infarction. All patients received prescriptions for at least one of the following drugs: statins, β-blockers, or calcium channel blockers (CCBs). Statins and β-blockers were chosen because they have previously been demonstrated to reduce mortality in this patient population. There is no known survival advantage associated with CCBs, and this drug class was selected as a control. The number of days on which patients had medication available to them (proportion of days covered [PDC]) in the year following collection of the first prescription (0-3 months after hospitalization) was calculated, and patients were classified as having high adherence (PDC ≥80%), medium adherence (PDC 40-79%), or low adherence (PDC <40%). Kaplan-Meier analysis was used to determine relationships between adherence and mortality, and confounding variables such as age, sex, severity of MI, concomitant illness, and socioeconomic status were controlled for. Multiple sensitivity analyses were also performed.The primary outcome measure was mortality up to 1 August 2005.A total of 31,455 patients were admitted with acute MI between 1 April 1999 and 1 May 2003, of whom 57% collected a prescription for a statin, 77% for a β-blocker, and 30% for a CCB. After a median follow-up of 2.4 years, the mean adherence rates for patients receiving statins, β-blockers, and CCBs were 87.5%, 83.9%, and 78.9%, respectively. Older age, recurrent hospital admission, and psychiatric illness were identified as independent predictors of discontinuation for both statins and β-blockers (P <0.001). Among patients prescribed statins, mortality was 25% higher (95% CI 1.09-1.42, P = 0.001) and 12% higher (95% CI 1.01-1.25, P = 0.03) in those with low and intermediate adherence, respectively, than in those who had high adherence. For patients prescribed β-blockers, those with low adherence had 13% higher mortality than those with high adherence (95% CI 1.03-1.25, P = 0.008), but the difference in mortality between patients with high and intermediate adherence was not statistically significant. Adherence to medication did not have an effect on survival among patients receiving CCBs.Strict adherence to statin and β-blocker therapy after MI results in improved long-term survival, a benefit that is not conferred by adherence to CCBs.