Background: The association between loneliness and suicide is poorly understood. We investigated how living alone, loneliness and emotional support were related to suicide and self-harm in a longitudinal design.
Methods: Between 2006 and 2010 UK Biobank recruited and assessed in detail over 0.5 million people in middle age. Data were linked to prospective hospital admission and mortality records. Adjusted Cox regression models were used to investigate relationships between living arrangements, loneliness and emotional support, and both suicide and self-harm as outcomes.
Results: For men, both living alone (Hazard Ratio (HR) 2.16, 95%CI 1.51-3.09) and living with non-partners (HR 1.80, 95%CI 1.08-3.00) were associated with death by suicide, independently of loneliness, which had a modest relationship with suicide (HR 1.43, 95%CI 0.1.01-2.03). For women, there was no evidence that living arrangements, loneliness or emotional support were associated with death by suicide. Associations between living alone and self-harm were explained by health for women, and by health, loneliness and emotional support for men. In fully adjusted models, loneliness was associated with hospital admissions for self-harm in both women (HR 1.89, 95%CI 1.57-2.28) and men (HR 1.74, 95%CI 1.40-2.16).
Limitations: Loneliness and emotional support were operationalized using single item measures.
Conclusions: For men - but not for women - living alone or living with a non-partner increased the risk of suicide, a finding not explained by subjective loneliness. Overall, loneliness may be more important as a risk factor for self-harm than for suicide. Loneliness also appears to lessen the protective effects of cohabitation.