Granger BB, Swedberg K, Ekman I, Granger CB, Olofsson B, McMurray JJ, Yusuf S, Michelson EL, Pfeffer MA; CHARM investigators

Adherence to candesartan and placebo and outcomes in chronic heart failure in the CHARM programme: double-blind, randomised, controlled clinical trial

BACKGROUND: Chronic heart failure (CHF) is an important cause of hospital admission and death. Poor adherence to medication is common in some chronic illnesses and might reduce the population effectiveness of proven treatments. Because little is known about adherence in patients with CHF and about the consequences of non-adherence, we assessed the association between adherence and clinical outcome in the CHARM (Candesartan in Heart failure: Assessment of Reduction in Mortality and morbidity) programme.
METHODS: CHARM was a double-blind, randomised, controlled clinical trial, comparing the effects of the angiotensin receptor blocker candesartan with placebo in 7599 patients with CHF. Median follow-up was 38 months. The proportion of time patients took more than 80% of their study medication was defined as good adherence and 80% or less as poor adherence. We used a Cox proportional hazards regression model, with adherence as a time-dependent covariate in the model, to examine the association between adherence and mortality in the candesartan and placebo groups.
FINDINGS: We excluded 187 patients because of missing information on adherence. In the time-dependent Cox regression model, after adjustment for predictive factors (demographics, physiological and severity-of-illness variables, smoking history, and number of concomitant medications), good adherence was associated with lower all-cause mortality in all patients (hazard ratio [HR] 0.65, 95% CI 0.57-0.75, p<0.0001). The adjusted HR for good adherence was similar in the candesartan (0.66, 0.55-0.81, p<0.0001) and placebo (0.64, 0.53-0.78, p<0.0001) groups.
INTERPRETATION: Good adherence to medication is associated with a lower risk of death than poor adherence in patients with CHF, irrespective of assigned treatment. This finding suggests that adherence is a marker for adherence to effective treatments other than study medications, or to other adherence behaviours that affect outcome. Understanding these factors could provide an opportunity for new interventions, including those aimed at improving adherence.