Factors associated with suicidal thoughts in a large community study of older adults
Faculty of Computing, Health and Science
School of Medical Sciences / Systems and Intervention Research Centre for Health
Background: Thoughts about death and self-harm in old age have been commonly associated with the presence of depression, but other risk factors may also be important. Aims: To determine the independent association between suicidal ideation in later life and demographic, lifestyle, socioeconomic, psychiatric and medical factors. Method: A cross-sectional study was conducted of a communityderived sample of 21 290 adults aged 60-101 years enrolled from Australian primary care practices. We considered that participants endorsing any of the four items of the Depressive Symptom Inventory - Suicidality Subscale were experiencing suicidal thoughts. We used standard procedures to collect demographic, lifestyle, psychosocial and clinical data. Anxiety and depressive symptoms were assessed with the Hospital Anxiety and Depression Scale. Results: The 2-week prevalence of suicidal ideation was 4.8%. Male gender, higher education, current smoking, living alone, poor social support, no religious practice, financial strain, childhood physical abuse, history of suicide in the family, past depression, current anxiety, depression or comorbid anxiety and depression, past suicide attempt, pain, poor self-perceived health and current use of antidepressants were independently associated with suicidal ideation. Poor social support was associated with a population attributable fraction of 38.0%, followed by history of depression (23.6%), concurrent anxiety and depression (19.7%), prevalent anxiety (15.1%), pain (13.7%) and no religious practice (11.4%). Conclusions: Prevalent and past mood disorders seem to be valid targets for indicated interventions designed to reduce suicidal thoughts and behaviour. However, our data indicate that social disconnectedness and stress account for a larger proportion of cases than mood disorders. Should these associations prove to be causal, then interventions that succeeded in addressing these issues would contribute the most to reducing suicidal ideation and, possibly, suicidal behaviour in later life.