Evaluation of treatment in 35 cases of bipolar suicide

The aim of the present study was to evaluate clinical factors relevant to suicide prevention (including treatment) in cases of bipolar suicide with available therapeutic histories.

Victorian Coroner's Office data enabled identification of suicides that occurred between March 1993 and December 2001. Cases involving sufficient clinical notes to enable diagnosis of DSM-IV bipolar disorder and review of treatment were de-identified and assessed by an expert clinical panel.

From 3752 suicides, 35 eligible bipolar subjects (22 men, 13 women) aged 40.3 +/- 1.8 years were identified. Duration of illness was 11.9 +/- 1.1 years. A total of 86% had made at least one previous suicide attempt, and 83% were in the depressed phase of illness. A total of 63% manifested psychosis at some time during lifetime illness. Fourteen per cent were inpatients, and 26% suicided within 6 weeks of hospital discharge. The panel's retrospective risk assessment concluded that only 48% of cases could have been assessed as high risk. In the 4 weeks prior to suicide, treatment was rated as not reaching benchmark standards in 60% of cases. Electroconvulsive therapy had been given to 11%, lithium to 43% (but definitely therapeutic in only 11%), 31% had never been treated with lithium, and psychosocial interventions did not reach adequate standards in 57% during the previous year.

In the majority of bipolar suicide cases in the present case series the subjects did not receive treatment at or above a benchmark standard, often due to illness and situational factors, but also possibly due to inadequate clinical interventions. Strategies to improve treatment may reduce suicide in bipolar disorder.