Aim
Although education in self-management is thought to be an important aspect of the care of patients with heart failure, little is known about whether self-rated knowledge of self-management is associated with outcomes. The aim of this study was to assess the relationship between patient-reported knowledge of self-management and clinical outcomes in patients with heart failure and reduced ejection fraction (HFrEF).

Methods and results
Using individual patient data from three recent clinical trials enrolling participants with HFrEF, we examined patient characteristics and clinical outcomes according to responses to the ‘self-efficacy’ questions of the Kansas City Cardiomyopathy Questionnaire. One question quantifies patients' understanding of how to prevent heart failure exacerbations (‘prevention’ question) and the other how to manage complications when they arise (‘response’ question). Self-reported answers from patients were pragmatically divided into: poor (do not understand at all, do not understand very well, somewhat understand), fair (mostly understand), and good (completely understand). Cox-proportional hazard models were used to evaluate time-to-first occurrence of each endpoint, and negative binomial regression analysis was performed to compare the composite of total (first and repeat) heart failure hospitalizations and cardiovascular death across the above-defined groups. Of patients (n = 17 629) completing the ‘prevention’ question, 4197 (23.8%), 6897 (39.1%), and 6535 (37.1%) patients had poor, fair, and good self-rated knowledge, respectively. Of those completing the ‘response’ question (n = 17 637), 4033 (22.9%), 5463 (31.0%), and 8141 (46.2%) patients had poor, fair, and good self-rated knowledge, respectively. For both questions, patients with ‘poor’ knowledge were older, more often female, and had a worse heart failure profile but similar treatment. The rates (95% confidence interval) per 100 person-years for the primary composite outcome for ‘poor’, ‘moderate’ and ‘good’ self-rated knowledge in answer to the ‘prevention’ question were 12.83 (12.11–13.60), 12.08 (11.53–12.65) and 11.55 (11.00–12.12), respectively, and for the ‘response’ question were 12.88 (12.13–13.67), 12.22 (11.60–12.86) and 11.56 (11.07–12.07), respectively. The lower event rates in patients with ‘good’ self-rate knowledge were accounted for by lower rates of cardiovascular (and all-cause) death and not hospitalization for worsening heart failure.

Conclusions
Poor patient-reported ‘self-efficacy’ may be associated with higher rates of mortality. Evaluation of knowledge of ‘self-efficacy’ may provide prognostic information and a guide to which patients may benefit from further education about self-management.