All-cause and cause-specific mortality due to physical illness in severe mental disorders in Europe

Abstract Background Severe mental illness (SMI), including schizophrenia-spectrum disorders (SSDs), bipolar disorder (BD) and major depressive disorder (MDD), represent the leading source of years lost to disability from all medical illnesses. Although much is known about excess mortality, there is a need to identify which domains of physical health problems are contributing to the mortality gap and whether these differ between different countries and health systems. We aimed to quantify excess mortality in severe mental illness across five European countries (Denmark, Finland, France, Poland and Sweden). Methods We used nationwide registers to identify persons diagnosed with SMIs, including SSDs, BD and MDD at ages 15-65 years. These persons were followed up until death, censoring diagnosis, age 75 years, or end of study period, whichever came first. The general population of each country was used as a reference to establish relative excess mortality in SMIs. The mortality data were aggregated over the study periods (varying between countries, 2004-2023). Random-effects meta-analysis of sex- and age-standardised mortality ratios (SMRs) to synthesize SMI-specific excess mortality across the four countries. Results In total, there were 210699 deaths attributable to SSDs, 69988 deaths to BD, and 273145 deaths to MDD. Pooled SMRs were 3.44 (95% CI: 2.81-4.23) in SSDs, 2.17 (95% CI: 1.79-2.62) in BD and 2.48 (95% CI: 2.01-3.06) in MDD. There were relatively small absolute differences in SMRs across countries, except for Poland where estimates for SSD and BD were markedly lower. The results of most common natural causes will be presented. Conclusions All three disorders showed marked excess mortality, with the highest excess mortality observed in SSDs. The relatively small absolute differences in SMRs across countries suggest consistent excess mortality patterns across these settings. Once data is available, we will add mortality data from Italy in the meta-analysis.