Suicide Risk Assessment in Early Psychosis: a Waste of Time?
Poster B15, Friday, October 21, 11:30 am - 1:00 pm, Le Baron
Javier-David Lopez-Morinigo1, Rina Dutta1, Andrea C Fernandes1, Hitesh Shetty1, Matthew Broadbent1, Robert Stewart1, Anthony S David1; 1Institute of Psychiatry, Psychology and Neuroscience. King's College London
PURPOSE: To investigate the sensitivity, specificity, positive and negative predictive values of routine clinical suicide risk assessment in non-affective psychosis. METHODS: Participants: Those patients with early non-affective psychosis (ICD-10 codes: F20-29) from the South London and Maudsley NHS Foundation Trust (London, UK), which is a large pseudonymised research database linked with national mortality, who took their lives over 2007-2013 (cases) were compared with unmatched controls, i.e. non-suicide completers with non-affective psychosis. Measures: Suicide risk assessment consists of 15 present/absent tick-boxes enquiring about suicide risk factors. Positive responses (yes=1, no=0) can be summed to create total scores, i.e. the higher the score the greater the suicide risk. Suicide was ascertained from the UK Office for National Statistics certificate of death. Statistics: ROC curves were plotted to assess the sensitivity and specificity of suicide risk assessment total scores to predict suicide. The prevalence of suicide in non-affective psychosis (5%) was used to estimate the positive and negative predictive values. RESULTS: 31 cases (22 males, age at first contact: 30.2 ± 9.5, age at death: 34.7 ± 11.2) and 214 controls had a documented risk assessment and were therefore compared. The best cut-off point was 3-4, with a sensitivity of 0.58 and a specificity of 0.57. The area under the curve was 0.63 (95%CI 0.51-0.74). The positive predictive value was 0.06, while the negative predictive value was 0.96. CONCLUSION: Routine clinical suicide risk assessment appears to be of little relevance for patients with early non-affective psychosis. More targeted and timely assessments are needed.
Topic Area: First Episode Psychosis