Depression and risk of future suicidal behaviour in patients with deliberate self-harm

Depression is strongly associated with suicidal behaviour, but whether major depressive episode (MDE) diagnosis and depressive symptom severity provide equivalent prognostic information after self-harm remains unclear.

739 adults were interviewed at three Swedish hospitals after an episode of deliberate self-harm; 503 women and 236 men, age range 18-95, mean age 38.4, 610 suicide attempts (SA) and 129 non-suicidal self-injuries (NSSI). The Montgomery-Åsberg Depression Rating Scale (MADRS-S), the Suicide Assessment Scale (SUAS) and the Suicide Intent Scale (SIS) were administered. Diagnoses were set using the Mini International Neuropsychiatric Interview (MINI). Fatal and non-fatal suicidal behaviour within 5 years were identified by national registers. Regression models were adjusted for age, sex, emotionally unstable personality disorder, bipolar disorder, baseline suicidal ideation and previous suicidal behaviour.

MDE was diagnosed in 72% (n = 534). Fatal or non-fatal suicidal behaviour occurred in 265 participants 36%). MDE was not independently associated with subsequent suicidal behaviour, whereas MADRS-S score was weakly associated after full adjustment (OR = 1.02, 95% CI 1.00-1.05, p = .037). Discriminatory performance was limited (AUC = 0.62).

In this diagnostically heterogeneous post-self-harm cohort, dimensional depressive symptom burden captured prognostic variation not evident from categorical MDE status. However, the association was modest, and MADRS-S had limited individual-level predictive utility. Depressive symptom severity should therefore be interpreted as one component of a broader multidimensional clinical assessment, rather than as a stand-alone risk prediction measure.