Characterizing Veteran suicide decedents that were not classified as high-suicide-risk

Evaluating Evidence-Based Psychotherapy Utilization Patterns Among Suicide-Risk-Stratified Veterans Diagnosed With Posttraumatic Stress Disorder

Posttraumatic stress disorder (PTSD) is a prevalent psychiatric condition, particularly among US Veterans. PTSD-diagnosed patients are more likely to experience suicidal ideation, suicide attempts and death by suicide when compared to non-PTSD-diagnosed patients. The US Department of Veterans Affairs (VA) emphasizes evidence-based psychotherapy (EBP) for PTSD, including prolonged exposure and cognitive processing therapy. This study focuses on how suicide risk impacts PTSD by evaluating utilization of nondifferentiated psychotherapy and EBP in a national sample of VA patients diagnosed with PTSD who died by suicide in 2017-2018. The study used a dataset of VA patients diagnosed with PTSD who died by suicide and received psychotherapy in the year before death (cases) and patients who had comparable diagnoses, demographics and received psychotherapy during the same interval and remained alive (controls). Cases and controls were matched on suicide risk (high, moderate and low). The study tracked nondifferentiated psychotherapy and EBP and analyzed cases and control utilization rates across risk-tiers. The final sample included high-risk (cases = 171; controls = 2052), moderate-risk (cases = 428; controls = 4280) and low-risk (cases = 53; controls = 529) patients. EBP utilization was markedly low, especially among cases. Higher proportions of moderate- and low-risk controls received EBP and received more sessions than matched cases. Even with VA efforts to promote EBPs, usage remains limited, particularly among patients who die by suicide. Further research is needed to understand utilization barriers and improve EBP delivery to better support PTSD-diagnosed patients and reduce their suicide risk.

Using Natural Language Processing to develop risk-tier specific suicide prediction models for Veterans Affairs patients

Suicide is a leading cause of death. Suicide rates are particularly elevated among Department of Veterans Affairs (VA) patients. While VA has made impactful suicide prevention advances, efforts primarily target high-risk patients with documented suicide risk. This high-risk population accounts for less than 10% of VA patient suicide deaths. We previously evaluated epidemiological patterns among VA patients that had lower classified suicide risk and derived moderate- and low-risk groupings. Expanding upon VA's leading suicide prediction model, this study uses national VA data to refine high-, moderate-, and low-risk specific suicide prediction methods. We selected all VA patients who died by suicide in 2017 or 2018 (n = 4584), matching each case with five controls who remained alive during treatment year and shared suicide risk percentiles. We extracted all sample unstructured electronic health record notes, analyzed them using natural language processing, and applied machine-learning classification algorithms to develop risk-tier-specific predictive models. We calculated area under the curve (AUC) and suicide risk concentration to evaluate predictive accuracy and analyzed derived words. RESULTS: Our high-risk model (AUC = 0.621 (95% CI: 0.55-0.68)), moderate-risk (AUC = 0.669 (95% CI: 0.64-0.71)), and low-risk (AUC = 0.673 (95% CI: 0.63-0.72)) models offered significant predictive accuracy over VA's leading suicide prediction algorithm. Derived words varied considerably, the high-risk model including chronic condition service words, moderate-risk model including outpatient care, and low-risk model including acute condition care. Study suggests benefit of leveraging unstructured electronic health records and expands prediction resources for non-high-risk suicide decedents, an historically underserved population.