LAUSR

The excess mortality observed in people with severe mental disorders (SMD) has not improved over time and is not captured in estimates of disease burden. This study aimed to improve upon previous analytic approaches to account for potential sources of heterogeneity in pooled mortality estimates. A systematic review of studies examining excess mortality in people with psychotic disorders and bipolar disorder was conducted. PubMed, EMBASE and PsycINFO were searched from January 1, 1980 to 31/12/20. Studies were eligible if they were longitudinal; the study population was diagnosed according to established criteria, not restricted to subgroups and the disorder was primary and not acute or transient; and mortality was reported in comparison to the general population or a control group without SMD. Meta-regression models were used to calculate pooled relative risks (RRs) of all-cause and cause-specific mortality adjusted for study- and population-level covariates. Risk of bias was assessed using an adaptation of the Newcastle-Ottawa scale. A total of 76 studies were included in the analyses. Covariates incorporated into the final models included age, sex, population type and mid-year. The adjusted RR for all-cause mortality in schizophrenia was 2.89 (95% CI 2.50 to 3.34) and 2.51 (95%CI 2.10 to 3.00) for bipolar disorder. There were larger RRs for broader categories of psychotic disorders. Mortality was elevated in each cause of death examined. Most of the heterogeneity between studies could not be accounted for. Future research should investigate underlying causal pathways and find ways to incorporate the excess mortality associated with SMD into global health estimates.