Suicide risk in people with post-traumatic stress disorder: A cohort study of 3.1 million people in Sweden

BACKGROUND: It is unclear whether post-traumatic stress disorder [PTSD] is associated with suicide risk in the general population, whether this differs by sex, or what the population impact of PTSD is for suicide. METHODS: We constructed a nationwide cohort of all people living in Sweden, born 1973–1997, followed from their 14th birthday (or immigration, if later) until suicide, other death, emigration or 31 December 2016. We used Cox proportional hazards regression to estimate hazard ratios [HR], and calculated the population impact of PTSD on suicide. We included sensitivity analyses to explore effects of outcome and exposure definitions, and to account for potential competing risks. RESULTS: Of 3,177,706 participants, 22,361 (0•7%) were diagnosed with PTSD, and 6,319 (0•2%) died by suicide over 49•2 million person-years. Compared with women and men without PTSD, suicide rates were 6•74 (95%CI: 5•61–8•09) and 3•96 (95%CI: 3•12–5•03) times higher in those with PTSD, respectively, after sociodemographic adjustment. Suicide rates remained elevated in women (HR: 2•61; 95%CI: 2•16–3•14) and men (HR: 1•67; 95%CI: 1•31–2•12) after adjustment for previous psychiatric conditions; attenuation was driven by previous non-fatal suicide attempts. Findings were insensitive to definitions or competing risks. If causal, 1•6% (95%CI: 1•2–2•1) of general population suicides could be attributed to PTSD, and up to 53.7% (95%CI: 46.1–60.2) in people with PTSD. LIMITATIONS: Residual confounding remains possible due to depressive and anxiety disorders diagnosed in primary care but unrecorded in these registers. CONCLUSIONS: Clinical guidelines for the management of people with PTSD should recognise increased suicide risks.