LAUSR

I read with great interest the article by Sokoreli and colleagues1 published in the European Journal of Heart Failure. The .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. authors investigated the prognostic value of psychosocial factors for predicting events after hospitalization in patients with heart failure (HF). Psychosocial variables, living alone and the presence of cognitive impairment were significantly associated with recurrent events, and I have two concerns about their study with special reference to mortality. First, Matsushita et al.2 conducted a prospective study to evaluate the effect of social factors on the prognosis of patients with HF. By multivariate Cox regression analysis, hazard ratio of no partner in the elderly, aged ≥75, for mortality was 1.500 (95% confidence interval 1.043–2.157). Patients with no partner can be classified into social vulnerability, and the lack of communication with partner would become a risk predictor for prognosis in patients with HF. Second, Shen et al.3 investigated the effect of marital status, social support, depression, and C-reactive protein on the mortality of patients with chronic HF. By multivariate logistic regression analysis, adjusted odds ratios (95% confidence intervals) of unmarried status, serum C-reactive protein level, and New York Heart Association class for mortality were 2.20 (1.01–4.79), 1.91 (1.05–3.50), and 2.41 (1.41–4.13), respectively. Combination of biological and social factors as independent variables would be more appropriate for comprehensive analysis.