In heart failure, a holistic approach incorporating the patient’s perspective is vital for prognosis and treatment. Self-rated health has strong associations with adverse events and short-term mortality risk, but long-term data are limited. We investigated the predictive value of two consecutive self-rated health assessments with regard to long-term mortality in a large, well characterised sample of elderly patients with stable chronic heart failure.

We measured self-rated health by asking ‘In general, would you say your health is: 1, excellent; 2, very good; 3, good; 4, fair; 5, poor?’ twice: at baseline and the end of a 12-week beta-blocker up-titration period in the CIBIS-ELD trial. Mortality was assessed in an observational follow-up after 2–4 years. A total of 720 patients (mean left ventricular ejection fraction 45±12%, mean age 73±5 years, 36% women) rated their health at both time points. During long-term follow-up, 144 patients died (all-cause mortality 20%). Fair/poor self-rated health in at least one of the two reports was associated with increased mortality (hazard ratio 1.42 per level; 95% confidence interval 1.16–1.75; P<0.001). It remained independently significant in multiple Cox regression analysis, adjusted for N-terminal pro B-type natriuretic peptide (NTproBNP), heart rate and other risk prediction covariates. Self-rated health by one level worse was as predictive for mortality as a 1.9-fold increase in NTproBNP.

Poor self-rated health predicts mortality in our long-term follow-up of patients with stable chronic heart failure, even after adjustment for established risk predictors. We encourage clinicians to capture patient-reported self-rated health routinely as an easy to assess, clinically meaningful measure and pay extra attention when self-rated health is poor.