Using the 1958 birth cohort, based on British births in March 1958, individuals were followed up to adulthood. We used data collected at birth and at age 7 years from various informants. Suicides occurring up to 31 May 2009 were identified from linked national death certificates. Multivariable Cox proportional hazard models were used to investigate risk factors.
Altogether 12399 participants (n = 44 suicides) had complete data. The strongest prenatal risk factors for suicide were: birth order, with risk increasing in later-born children [p
trend = 0.063, adjusted hazard ratio (HR)], e.g. for fourth- or later-born children [HR = 2.27, 95% confidence interval (CI) 0.90–5.75]; young maternal age (HR = 1.18, 95% CI 0.34–4.13 for ⩽19 years and HR = 0.41, 95% CI 0.19–0.91 for >29 years, p
trend = 0.034); and low (<2.5 kg) birth weight (HR = 2.48, 95% CI 1.03–5.95). The strongest risk factors at 7 years were externalizing problems in males (HR = 2.96, 95% CI 1.03–8.47, p
trend = 0.050) and number of emotional adversities (i.e. parental death, neglected appearance, domestic tension, institutional care, contact with social services, parental divorce/separation and bullying) for which there was a graded association with risk of suicide (p
trend = 0.033); the highest (HR = 3.12, 95% CI 1.01–9.62) was for persons with three or more adversities.
Risk factors recorded at birth and at 7 years may influence an individual's long-term risk of suicide, suggesting that trajectories leading to suicide have roots in early life. Some factors are amenable to intervention, but for others a better understanding of causal mechanisms may provide new insights for intervention to reduce suicide risk.