MENTAL HEALTH AND SUBSTANCE USE

Research Objective National health care efforts have been focused on ensuring patients have access to safe and effective chronic pain treatment while reducing the risk of opioid use disorder, overdose, and death. Within Veterans Health Administration (VHA), recent policy has sought to improve targeting of clinical interventions for patients with high‐dose opioid prescriptions. The Stratification Tool for Opioid Risk Management (STORM) is a safety initiative implemented by VHA to help clinicians evaluate and mitigate opioid‐related risks among their patients. STORM’s predictive model and web‐based dashboard help clinicians identify patients with an opioid prescription who are at high risk of serious adverse events (SAEs) (such as overdose, overdose death, or suicide) and may benefit from clinical interventions to mitigate that risk. We examine the extent to which VHA clinicians effectively target patients for risk mitigation prior to and following implementation of the STORM dashboard. Additionally, we analyze the extent to which implementation of STORM is associated with increased rates of mitigation within each patient risk strata. Study Design This study leverages a national observational study of VHA prescription, inpatient, outpatient, and mortality data to categorize patients based on risk profiling. We identify clinical treatment activity across thirteen distinct risk mitigation strategies (such as psychosocial assessments) for a high‐risk patient population prior to and after implementation of STORM. Population Studied Since early 2018, the STORM algorithm has calculated a daily risk score for SAEs for each Veteran receiving an outpatient opioid prescription, based upon demographic information, previous SAEs, substance use, and mental health history from electronic medical records. This study focuses on high‐risk patients, defined as those at or above the 95th percentile of risk score, and further stratifies within this group of patients to examine comparative rates of risk mitigation. Principal Findings Prior to implementation of the STORM dashboard and mandated use of the associated scoring algorithm to target clinical interventions, VHA providers were already providing risk mitigation interventions at higher rates for higher risk patients than for lower risk patients. Specifically, patients in the top 1% of risk were observed to receive statistically significantly higher rates of risk mitigation than patients at lower risk levels including timely drug screening, psychological assessment, and safety planning. Following the nationwide implementation of STORM, higher rates of risk mitigation strategies were observed within each strata of high‐risk patients. For example, the rate of psychosocial assessments and use of safety plans were observed to statistically significantly increase by 6‐34% across each risk strata following implementation. Conclusions Evidence from the pre‐STORM period suggests that VHA clinicians judge inherent risk when managing patients with opioid prescriptions and making clinical treatment decisions. VHA opioid patients at the highest risk for SAEs are more likely to receive risk mitigation than patients with lower risk. Inclusion of high‐risk patients in STORM is associated with even higher rates of mitigation strategies across all risk strata. Implications for Policy or Practice Policy initiatives that help clinicians identify high‐risk patients may enhance targeting and use of risk mitigation strategies among vulnerable patient populations. Primary Funding Source Department of Veterans Affairs.