Incidence and Predictors of Suicide Attempts in DSM-IV Major Depressive Disorder: A Five-Year Prospective Study

Objective: Prospective long-term studies of risk factors for suicide attempts among patients with major depressive disorder have not investigated the course of illness and state at the time of the act. Therefore, the importance of state factors, particularly time spent in risk states, for overall risk remains unknown. Method: In the Vantaa Depression Study, a longitudinal 5-year evaluation of psychiatric patients with major depressive disorder, prospective information on 249 patients (92.6%) was available. Time spent in depressive states and the timing of suicide attempts were investigated with life charts. Results: During the follow-up assessment period, there were 106 suicide attempts per 1,018 patient-years. The incidence rate per 1,000 patient-years during major depressive episodes was 21-fold (N = 332 [95% confidence interval [CI] = 258.6-419.2]), and it was fourfold during partial remission (N = 62 [95% CI = 34.6-92.4]) compared with full remission (N = 16 [95% CI = 11.2-40.2]). In the Cox proportional hazards model, suicide attempts were predicted by the months spent in a major depressive episode (hazard ratio = 7.74 [95% CI = 3.40-17.6]) or in partial remission (hazard ratio = 4.20 [95% CI = 1.71-10.3]), history of suicide attempts (hazard ratio = 4.39 [95% CI = 1.78-10.8]), age (hazard ratio = 0.94 [95% CI = 0.91-0.98]), lack of a partner (hazard ratio = 2.33 [95% CI = 0.97-5.56]), and low perceived social support (hazard ratio = 3.57 [95% CI = 1.09-11.1]). The adjusted population attributable fraction of the time spent depressed for suicide attempts was 78%. Conclusions: Among patients with major depressive disorder, incidence of suicide attempts varies markedly depending on the level of depression, being highest during major depressive episodes. Although previous attempts and poor social support also indicate risk, the time spent depressed is likely the major factor determining overall long-term risk. (Am J Psychiatry 2010; 167:801-808)