Examining the Effectiveness of an Adaptive Implementation Intervention to Improve Uptake of the VA Suicide Risk Identification Strategy

点击翻译按钮获取中文摘要

Reducing Veteran suicide has been one of the Department of Veterans Affairs’ (VA) top priorities for the past decade. Efforts to address Veteran suicide have typically included downstream approaches focused on suicide risk management for those who have previously been identified to be at elevated risk. The Joint Commission, however, released a Sentinel Event Alert1 highlighting findings that a substantial number of those who died by suicide had not presented to mental health clinics or psychiatric units. Research suggests that many individuals, including Veterans who die by suicide had been seen by their primary care provider or in other medical settings within the year of their death2-4. Based on this, the Office of Mental Health and Suicide Prevention (OMHSP) tasked an interdisciplinary workgroup to develop an evidence-based, population-level approach to identifying suicide risk in Veterans who present in a wide range of healthcare settings. From this work, the largest implementation of a system-wide suicide risk screening and evaluation initiative within a U.S. healthcare system was developed - the VA Suicide Risk Identification Strategy (VA Risk ID). VA Risk ID is comprised of three stages: The first stage utilizes the Patient Health Questionnaire-95, item 9 as a primary screen to maximize sensitivity; the second stage uses the Columbia Suicide Severity Rating Scale Screener6 as a secondary screen to maximize specificity, and the third stage uses the VA Comprehensive Suicide Risk Evaluation (CSRE) to gather information related to clinical impressions of acute and chronic suicide risk, which are used to develop a risk mitigation plan. The goal of the proposed project is to employ a Sequential Multiple Assignment Randomized Trial (SMART) design to improve the implementation of VA Risk ID’s three-stage screening and evaluation process to fidelity (in correct sequence, by the appropriate provider, and within the designated time frame) for Veterans receiving annual screenings for depression and PTSD. Within the SMART, two evidence-based implementation interventions will be compared to Implementation as Usual (IAU): Audit and Feedback (A/F) and Audit and Feedback plus External Facilitation (A/F+EF). IAU is available to all sites and includes proactive technical assistance, tools, training and a quality assurance dashboard. A/F will provide sites with frequent, non-punitive reports that include information on: individual site performance and comparisons with national average (based on VA Risk ID clinical performance measures), specific areas of improvement and suggested actions. A/F+EF groups will receive A/F reports and external facilitation, which includes strategies such as, stakeholder engagement, identification of barriers and facilitators and problem-solving. This project will occur over three phases: run-in phase, intervention phase I and intervention phase II. The unit of intervention is the site and randomization will occur at the site level. A sequence of decisions regarding which interventions sites will receive during intervention phases I and II are built into the study design. Decisions will be based on site performance. Specifically, only sites who do not meet the program office’s benchmark for adequate performance (i.e., completion of secondary screening and CSRE for 80% or more of eligible patients) will be randomized to receive the implementation interventions, beginning with the less intensive intervention (i.e., A/F) for 10 months. Sites that still do not meet the performance benchmark after receiving A/F, will be re- randomized at the end of Phase I to continue receiving either A/F or a more intensive intervention consisting of A/F + external facilitation (A/F + EF) for another 10 months. The primary aim seeks to understand whether the addition of A/F during the initial phase will significantly improve performance measure scores compared to IAU alone; a secondary aim will examine whether A/F+EF significantly improves performance scores compared to A/F alone. Exploratory aims will consider contextual factors which may influence response to implementation interventions as well as clinical outcomes related to VA Risk ID.