Is the population level of anxiety and depression associated with mortality? Data from the ESSE-RF study

Abstract

Aim. To study the associations of subclinical and clinical anxiety and depression, assesed by the Hospital Anxiety and Depression Scale (HADS), ≥8 points and ≥11 points, respectively, with all-cause mortality and cardiovascular mortality, as well as with the total number of nonfatal cardiovascular events (CVEs) in Russia.Material and methods. The study included male and female population aged 25-64 years from the Epidemiology of Cardiovascular Diseases and their Risk Factors in Regions of Russian Federation (ESSE-RF) study who signed an informed consent. The study included 16941 people (men, 6811 (40,2%)). To assess the anxiety and depression, HADS scale was used. The statistical analysis included individuals with subclinical/clinical (HADS ≥8) and clinical (HADS ≥11) anxiety and depression (HADS-A and HADS-D). The median of prospective follow-up was 5,5 years (from 2012 to 2019). The composite endpoint (CE) (cardiovascular death and nonfatal CVE occurred in 268 (4,2%) men and 203 (2,1%) women, while all-cause — in 220 (3,2%) men and 152 (1,5%) women.Results. The results obtained showed that the survival rate of men and women with varying degrees of anxiety (HADS-A ≥8 and ≥11) was associated with all-cause mortality and CE in women. However, this relationship has not been confirmed in multivariate models. In the multivariate Cox proportional hazards model 1 (M1), a significant association of depression with all-cause mortality in women was revealed — HADS-D ≥8: relative risk (RR), 2,22; 95% confidence interval (CI): 1,56-3,15 and ≥11: RR, 2,43; 95% CI: 1,65-3,59 (pConclusion. Depression (HADS-D ≥8 and ≥11) in Russian women was significantly associated with all-cause mortality and CE (HADS-D ≥8). In men, depression (HADS-D ≥8) was significantly associated with allcause mortality when only conventional risk factors were included in the model, without taking into account prior cardiovascular disease. Anxiety in multivariate models was not associated with all-cause mortality and CE in both sex groups.