Identifying mortality risks in patients with opioid use disorder using brief screening assessment: Secondary mental health clinical records analysis

BACKGROUND: Risk assessments are widely used, but their ability to predict outcomes in opioid use disorder (OUD) treatment remains unclear. Therefore, the aim was to investigate if addiction-specific brief risk screening is effective in identifying high mortality risk groups and if subsequent clinical actions following risk assessment impacts on mortality levels. METHODS: Opioid use disorder (OUD) patients were identified in the South London and Maudsley Case Register. Deaths were identified through database linkage to the national mortality dataset. Cox and competing-risk regression were used to model associations between brief risk assessment domains and all-cause and overdose mortality in 4488 OUD patients, with up-to 6-year follow-up time where 227 deaths were registered. Data were stratified by admission to general mental health services. RESULTS: All-cause mortality was significantly associated with unsafe injecting (HR 1.53, 95% CI 1.10-2.11) and clinically appraised likelihood of accidental overdose (HR 1.48, 95% CI 1.00-2.19). Overdose-mortality was significantly associated with unsafe injecting (SHR 2.52, 95% CI 1.11-5.70) and clinically appraised suicidality (SHR 2.89, 95% CI 1.38-6.03). Suicidality was associated with a twofold increase in mortality risk among OUD patients who were not admitted to mental health services within 2 months of their risk assessment (HR 2.03, 95% CI 1.67-3.24). CONCLUSIONS: Diagnosis-specific brief risk screening can identify OUD patient subgroups at increased risk of all-cause and overdose mortality. OUD patients, where suicidality is evident, who are not admitted into services are particularly vulnerable.