Attributable risk of psychiatric and socio-economic factors for suicide from individual-level, population-based studies: a systematic review

The overall importance of a risk factor for suicide in a population is determined not only by the relative risk (RR) of suicide but also the prevalence of the risk factor in the population, which can be combined with the RR to calculate the population attributable risk (PAR). This study compares risk factors from two well studied domains of suicide research - socio-economic deprivation (relatively low RR, but high population prevalence) and mental disorders (relatively high RR risk, but low population prevalence). RR and PAR associated with suicide was estimated for high prevalence ICD-10/DSM-IV psychiatric disorders and measures of socio-economic status (SES) from individual-level, population-based studies. A systematic review and meta-analysis was conducted of population-based case-control and cohort studies of suicide where relative risk estimates for males and females could be extracted. RR for any mental disorder was 7.5 (6.2-9.0) for males and 11.7 (9.7-14.1) for females, compared to RR for the lowest SES groups of 2.1 (1.5-2.8) for males and 1.5 (1.2-1.9) for females. PAR in males for low educational achievement (41%, range 19-47%) and low occupational status (33%, range 21-42%) was of a similar magnitude to affective disorders (26%, range 7-45%) and substance use disorders (9%, range 5-24%). Similarly in females the PAR for low educational achievement (20%, range 19-22%) was of a similar magnitude to affective disorders (32%, range 19-67%), substance use disorder (25%, range 5-32%) and anxiety disorder (12%, range 6-22%). The findings of the present study suggest that prevention strategies which focus on lower socio-economic strata (more distal risk factors) have the potential to have similar population-level effects as strategies which target more proximal psychiatric risk factors in the prevention and control of suicide.