Evaluation of a Model to Target High-risk Psychiatric Inpatients for an Intensive Postdischarge Suicide Prevention Intervention.

Importance: The months after psychiatric hospital discharge are a time of high risk for suicide. Intensive postdischarge case management, although potentially effective in suicide prevention, is likely to be cost-effective only if targeted at high-risk patients. A previously developed machine learning (ML) model showed that postdischarge suicides can be predicted from electronic health records and geospatial data, but it is unknown if prediction could be improved by adding additional information. Objective: To determine whether model prediction could be improved by adding information extracted from clinical notes and public records. Design, Setting, and Participants: Models were trained to predict suicides in the 12 months after Veterans Health Administration (VHA) short-term (less than 365 days) psychiatric hospitalizations between the beginning of 2010 and September 1, 2012 (299 050 hospitalizations, with 916 hospitalizations followed within 12 months by suicides) and tested in the hospitalizations from September 2, 2012, to December 31, 2013 (149 738 hospitalizations, with 393 hospitalizations followed within 12 months by suicides). Validation focused on net benefit across a range of plausible decision thresholds. Predictor importance was assessed with Shapley additive explanations (SHAP) values. Data were analyzed from January to August 2022. Main Outcomes and Measures: Suicides were defined by the National Death Index. Base model predictors included VHA electronic health records and patient residential data. The expanded predictors came from natural language processing (NLP) of clinical notes and a social determinants of health (SDOH) public records database. Results: The model included 448 788 unique hospitalizations. Net benefit over risk horizons between 3 and 12 months was generally highest for the model that included both NLP and SDOH predictors (area under the receiver operating characteristic curve range, 0.747-0.780; area under the precision recall curve relative to the suicide rate range, 3.87-5.75). NLP and SDOH predictors also had the highest predictor class-level SHAP values (proportional SHAP = 64.0% and 49.3%, respectively), although the single highest positive variable-level SHAP value was for a count of medications classified by the US Food and Drug Administration as increasing suicide risk prescribed the year before hospitalization (proportional SHAP = 15.0%). Conclusions and Relevance: In this study, clinical notes and public records were found to improve ML model prediction of suicide after psychiatric hospitalization. The model had positive net benefit over 3-month to 12-month risk horizons for plausible decision thresholds. Although caution is needed in inferring causality based on predictor importance, several key predictors have potential intervention implications that should be investigated in future studies.

Changes in secondary outcomes associated with brief interventions for problem gambling in methadone patients.

BACKGROUND: Patients in methadone maintenance treatment (MMT) with problem gambling (PG) experience worse psychosocial outcomes than their non-PG counterparts. Interventions targeting PG in MMT may enhance psychosocial functioning beyond gambling reduction and abstinence. The present study was a secondary data analysis that examined the trajectories of non-gambling outcomes of three brief PG interventions (i.e., brief psychoeducation, brief advice, motivational enhancement therapy plus cognitive-behavioral therapy [MET + CBT]) among MMT patients. METHODS: Participants (N = 109) were engaged in substance use disorder treatment, met criteria for PG, and had a current or lifetime history of MMT. Latent growth curve models examined outcome trajectories of psychiatric, medical, legal, employment, and social problems, as well as psychological distress and quality of life. Follow-up analyses examined clinically significant change. RESULTS: MET + CBT patients reported lower medical problems at baseline and over time than the brief interventions. There was no evidence of differences between interventions on the other outcomes. Psychiatric problems and psychological distress decreased over time for the entire sample, regardless of the PG intervention. About 24% and 13% of the sample demonstrated clinically significant improvements in psychological distress from baseline to 5 months, and 5 months to 12 months, respectively. Nearly 21% of the sample showed clinically significant improvements in psychiatric problems from 5 months to 12 months. Among all patients, men and those with more severe opioid dependence symptoms demonstrated the greatest psychological improvements. CONCLUSIONS: Many patients in MMT with PG experience improvements in psychological problems, including long-term improvement, regardless of the PG intervention offered.

Changes in Neuropsychological Status during the Initial Phase of Abstinence in Alcohol Use Disorder: Neurocognitive Impairment and Implications for Clinical Care.

BACKGROUND: Neuropsychological deficits are common in individuals with alcohol use disorder (AUD) and impact daily functioning and AUD treatment outcomes. Longitudinal studies demonstrate that extended abstinence improves neuropsychological functioning. The effects of short-term abstinence are less clear. OBJECTIVES: This study examined changes in neuropsychological functioning after acute detoxification over a 10-day period at the beginning of residential AUD treatment. Notably, this study evaluated cognitive functioning according to diagnostic classifications for neurocognitive disorder according to DSM-5. METHODS: Using a within-subjects design, neuropsychological functioning of participants (N = 28) undergoing a 14-day residential AUD treatment program was assessed at two time points over 10 days (i.e., treatment entry, prior to treatment discharge). Tests of immediate memory, visuospatial abilities, attention, language abilities, delayed memory, and executive functioning were administered. RESULTS: After completing acute detoxification, almost all participants (93%) were clinically impaired in at least one of the five cognitive domains at residential treatment entry, with one third of the sample impaired on ≥3 domains. Ten days later, 71% remained clinically impaired in at least one of five cognitive domains, with 29% of the sample impaired on ≥3 domains. Significant improvement over the 10-day period was observed for immediate memory, visuospatial abilities, and overall cognitive functioning. Clinical significance of these changes is also reported. Conclusions/Importance: The results from this study help to characterize cognitive functioning in terms of neurocognitive impairment. A brief period of abstinence begins to ameliorate neuropsychological deficits, but many individuals remain cognitively impaired throughout treatment. Implications for treatment are discussed.