Unemployment and mortality-a longitudinal prospective study on selection and causation in 49321 Swedish middle-aged men

Background: Unemployment is associated with increased risk of mortality. It is, however, not clear to what extent this is causal, or whether other risk factors remain uncontrolled for. The aim of this study was to investigate the association between unemployment and all-cause and cause-specific mortality, adjusting for indicators of mental disorder, behavioural risk factors and social factors over the life course. Methods: This study was based on a cohort of 49321 Swedish males, born 1949/51, tested for compulsory military conscription in 1969/70. Data on employment/unemployment 1990-4 was based on information from the Longitudinal Register of Education and Labour Market Statistics. Information on childhood circumstances was drawn from National Population and Housing Census 1960. Information on psychiatric diagnosis and behavioral risk factors was collected at conscription testing in 1969/70. Data on mortality and hospitalisation 1973-2004 were collected in national registers. Results: An increased risk of mortality 1995-2003 was found among individuals who experienced 90 days or more of unemployment during 1992-4 compared with those still employed (all-cause mortality HR 1.91, 95% CI 1.58 to 2.31. Adjustment for risk factors measured along the life course considerably lowered the relative risk (all cause mortality HR 1.30, 95% CI 1.06 to 1.58). Statistically significant increased relative risk was found during the first 4 years of follow up (all-cause mortality, adjusted HR 1.57, 95% CI 1.13 to 2.18, but not the following 4 years (all cause mortality, adjusted HR 1.17, 95% CI 0.91 to 1.50). Conclusion: The results suggest that a substantial part of the increased relative risk of mortality associated with unemployment may be attributable to confounding by individual risk factors.