Investigating the differential impact of psychosocial factors by patient characteristics and demographics on Veteran suicide risk through machine learning extraction of cross-modal interactions
Accurate prediction of suicide risk is crucial for identifying patients with elevated risk burden, helping ensure these patients receive targeted care. The US Department of Veteran Affairs' suicide prediction model primarily leverages structured electronic health records (EHR) data. This approach largely overlooks unstructured EHR, a data format that could be utilized to enhance predictive accuracy. This study aims to enhance suicide risk models' predictive accuracy by developing a model that incorporates both structured EHR predictors and semantic NLP-derived variables from unstructured EHR. XGBoost models were fit to predict suicide risk- the interactions identified by the model were extracted using SHAP, validated using logistic regression models, added to a ridge regression model, which was subsequently compared to a ridge regression approach without the use of interactions. By introducing a selection parameter, α, to balance the influence of structured (α=1) and unstructured (α=0) data, we found that intermediate α values achieved optimal performance across various risk strata, improved model performance of the ridge regression approach and uncovered significant cross-modal interactions between psychosocial constructs and patient characteristics. These interactions highlight how psychosocial risk factors are influenced by individual patient contexts, potentially informing improved risk prediction methods and personalized interventions. Our findings underscore the importance of incorporating nuanced narrative data into predictive models and set the stage for future research that will expand the use of advanced machine learning techniques, including deep learning, to further refine suicide risk prediction methods.
Abstract: Objectives: Increasing numbers of older adults are reentering community following incarceration (i.e., reentry), yet risk of incident neurodegenerative disorders associated with reentry is unknown. Our objective was to determine association between reentry status (reentry vs never-incarcerated) and mild cognitive impairment (MCI) and/or dementia. Methods: This nationwide, longitudinal cohort study used linked Centers for Medicare & Medicaid Services and Veterans Health Administration data. Participants were aged 65 years or older who experienced reentry between October 1, 2012, and December 31, 2018, with no preincarceration MCI/dementia, compared with age-matched/sex-matched never-incarcerated veterans. MCI/dementia was defined by diagnostic codes. Fine-Gray proportional hazards models were used to examine association. Results: This study included 35,520 veterans, mean age of 70 years, and approximately 1% women. The reentry group (N = 5,920) had higher incidence of MCI/dementia compared with the never-incarcerated group (N = 29,600; 10.2% vs 7.2%; fully adjusted hazard ratio [aHR] 1.12; 95% CI 1.00-1.25). On further investigation, reentry was associated with increased risk of dementia with or without prior MCI diagnosis (aHR 1.21; 95% CI 1.06-1.39) but not MCI only. Discussion: Transition from incarceration to community increased risk of neurocognitive diagnosis. Findings indicate health/social services to identify and address significant cognitive deficits on late-life reentry. Limitations include generalizability to nonveterans.