Heart failure and suicide mortality and suicide attempts: a systematic review and meta-analysis of cohort studies

Objective This systematic review and meta-analysis aimed to synthesize longitudinal observational evidence on the associations of heart failure (HF) with suicide mortality and suicide attempts. Methods PubMed, Embase, and the Cochrane Library were systematically searched from inception to April 11, 2026, to identify cohort studies evaluating suicide mortality or suicide attempts among individuals with HF. Study quality was assessed using the Newcastle-Ottawa Scale. Because suicide mortality and suicide attempts are clinically distinct outcomes, they were analyzed separately. The primary quantitative synthesis was restricted to studies evaluating suicide death or completed suicide and reporting adjusted hazard ratios (HRs). Adjusted HRs were log-transformed and pooled using a generic inverse-variance random-effects model. Studies evaluating suicide attempts were summarized narratively because only two studies were available and they differed substantially in population, clinical setting, and effect measure. Between-study heterogeneity was assessed using Cochran's Q test and the I² statistic. Results Seven cohort studies published between 2018 and 2026, involving 17,696,608 participants, were included in the systematic review. Five studies evaluating suicide death or completed suicide contributed to the primary meta-analysis. HF was associated with suicide mortality, with a pooled adjusted HR of [HR] (95% CI: [lower–upper]; I 2 = [I 2 ]%, P for heterogeneity = [ P value]). Two studies evaluated suicide attempts but were not pooled because one reported an HR in a large community-based cohort, whereas the other reported an RR in a small, highly selected population of patients with end-stage HF undergoing left ventricular assist device implantation. No included study reported suicidal ideation as an independently extractable longitudinal outcome. Conclusion Observational cohort evidence suggests that HF is associated with an increased hazard of suicide mortality. Evidence regarding suicide attempts remains too limited and heterogeneous for a reliable pooled estimate, and no independently extractable longitudinal evidence was available for suicidal ideation. Residual confounding, between-study heterogeneity, and the small number of eligible studies limit causal interpretation and the certainty of the findings.