AimsAtrial fibrillation (AF) is common in patients with heart failure (HF). CHA2DS2‐VASc score was originally employed as a risk assessment tool for stroke in patients with AF; however, it has recently been used to predict not only stroke but also various cardiovascular diseases beyond the original AF field. We aimed to verify the CHA2DS2‐VASc score as a risk assessment tool to predict mortality in patients with HF.Methods and ResultsConsecutive 1011 patients admitted for treatment of HF were divided into three groups based on their CHA2DS2‐VASc scores: score 1–3 group (n = 317), score 4–6 group (n = 549) and score 7–9 group (n = 145). Of the 1011 HF patients, 387 (38.3%) had AF. We compared patient characteristics among the three groups and prospectively followed for all‐cause mortality. Although left ventricular ejection fraction was similar among all three groups, all‐cause mortality was higher in the score 4–6 group and score 7–9 group than in the score 1–3 group (37.9 and 29.3% vs. 15.1%, log‐rank P < 0.001). In the multivariable Cox proportional hazard analysis, the CHA2DS2‐VASc score 7–9 was an independent predictor of all‐cause mortality (all HF patients: hazard ratio (HR) 1.822, P = 0.011; HF patients with AF: HR 1.951, P = 0.031; HF patients without AF: HR 2.215, P = 0.033).ConclusionsThe CHA2DS2‐VASc score was an independent predictor of all‐cause mortality in HF patients with or without AF. This comprehensive risk assessment score may help identify HF patients who are at high risk for mortality in HF patient.