Main Article Content
Both rates of self-reported mental ill-health and the number of individuals in receipt of psychotropic medications are increasing. Research suggests that poor mental health is a major predictor of death by suicide, but measuring individual level mental ill-health can be difficult.
Objectives and Approach
This study assessed if self-reported mental health, psychotropic medication record, or both in combination better predict risk of death by suicide. Data from the 2011 Northern Ireland Census on all individuals living in the community aged 18-74 years (n=1,098,967) were linked to the Enhanced Prescribing Database (EPD) and death registrations until end of 2015. Mental health status was ascertained through a single-item self-reported mental health question from the Census and record of psychotropic medication in the EPD. Deaths by suicide were identified using ICD-10 codes. Logistic regression models examined the association between measures of mental health and death by suicide, with adjustment for individual attributes known to be associated with mental ill-health.
Of the 1,098,967 individuals, 857 died by suicide during the study period; 0.05% of those with no mental ill-health, 0.21% who self-reported poor mental health, 0.16% of those in receipt of psychotropic medication and 0.38% of those with both. In unadjusted regression models, having both self-report and psychotropic medication record was the strongest predictor of suicide (OR=8.23, 95%CI: 6.97–9.72). Upon adjustment, having both remained the strongest suicide predictor (OR=6.13, 95%CI: 4.94–7.61), followed by psychotropic medication record only (OR=4.00, 95%CI: 3.28–4.88) and self-report only (OR=2.88, 95%CI: 2.16–3.84).
Individuals with both self-reported poor mental health and receipt of psychotropic medication had the highest likelihood of death by suicide. Of the two measures individually, receipt of psychotropic medication was associated with a higher likelihood of suicide mortality. Understanding suicide risk is vitally important for the effective targeting of interventions.
This work is licensed under a Creative Commons Attribution 4.0 International License.