Benefits and Limitations of Suicide Risk Assessment Processes in VA

Dobscha SK. Benefits and Limitations of Suicide Risk Assessment Processes in VA. Paper presented at: VA / Department of Defense Suicide Prevention Annual Conference; 2015 Jan 27; Dallas, TX.

Related HSR&D Project(s)

• DHI 08-096 – Outcomes and Correlates of Suicidal Ideation in OEF/OIF Veterans

Objectives: The main objective of this study was to identify process outcomes among Operation Enduring Freedom and Operation Iraqi Freedom (OEF/OIF) Veterans with depression who received brief suicidal ideation risk assessment (BSRA) in VA facilities. In this presentation our goals were to 1) Identify the extent to which processes of care are affected by positive BSRAs and 2) Explore Veterans' experiences of VA's SI assessment processes. Methods: Mixed-methods, case-control study. We collected quantitative data from administrative databases of three large, geographically diverse 1a VA facilities in order to identify OEF/OIF Veterans screened for depression from April 2008 - September 2009. We determined who received BRSA within 30 days, and then examined processes of care over 12 months. Follow-up analysis examined same-day primary care clinician actions in response to positive BSRAs. We also conducted individual interviews of 34 OEF/OIF Veterans with SI within 2-6 months after BSRAs were conducted in primary care. Qualitative methods examined Veterans' perceptions of brief SI assessments and the healthcare system's response to screening. Results: Among new to care OEF/OIF Veterans (n = 465) who had been administered BSRAs, 32% had positive BSRAs. Eighty percent were diagnosed with mental health condition(s) on the same day. Over the 12-month follow-up, 81% had at least one mental health (MH) visit, 53% filled at least one antidepressant prescription, and 5% had inpatient psychiatric admission. Multivariate modeling suggested that married Veterans, those assessed for SI by a MH clinician, and those with depression or anxiety disorder were more likely to have MH visits and to receive antidepressants. Veterans whose BSRA was positive were more likely to have MH visits, but not antidepressants. The second model added PHQ-2 score, which was associated with MH visits, but positive BSRAs were no longer significantly associated with MH visits or antidepressant prescriptions. This suggests that the BSRA result may not add predictive information about subsequent utilization above and beyond depression severity. In follow-up analysis of same-day primary care clinician actions, most discussed mental health disorders and follow-up treatment, and risk factors including pain, partner/relationship difficulties, and occupational difficulties. Male Veterans and those with known SUDs were more likely to be counseled to reduce use of alcohol/drugs, while Veterans with recent SI and other suicide behaviors were more likely to have discussions about firearms (but not about reducing use of alcohol/drugs). In qualitative findings, Veterans expressed appreciation that BSRA questions were straightforward, but also frustration with their inability to provide context/nuance, discomfort disclosing SI outside of patient-centered conversation, and repeated BSRAs which led to perception of communication gaps within the team and futility about getting needs addressed. Implications: The location and type of clinician who administers BSRAs, and mental health diagnosis or severity, may be as good or better predictors of subsequent care than BSRA results (+ or -). Primary care clinicians need more training and support for addressing what to do after SI is detected. Suicide risk assessment will be more effective if performed by a provider who knows the patient best. Clinicians may need more training and support for embedding SI risk assessment into patient-centered conversations; repetitive standardized (routinized) assessment should be avoided.