Participatory System Dynamics vs Usual Quality Improvement: Is Staff Use of Simulation an Effective, Scalable and Affordable Way to Improve Timely Veteran Access to High-quality Mental Health Care?
Participatory System Dynamics vs Usual Quality Improvement: Is Staff Use of Simulation an Effective, Scalable and Affordable Way to Improve Timely Veteran Access to High-quality Mental Health Care?
Evidence-based VA care is best for meeting Veterans' mental health needs, such as depression, PTSD and opioid use disorder, to prevent suicide or overdose. But some key evidence-based practices only reach 3-28% of patients. Participatory system dynamics (PSD) helps improve quality with existing resources, critical in mental health and all VA health care. PSD uses learning simulations to improve staff decisions, showing how goals for quality can best be achieved given local resources and constraints. This study aims to significantly increase the proportion of patients who start and complete evidence-based care, and determine the costs of using PSD for improvement. Empowering frontline staff with PSD simulation encourages safe 'virtual' prototyping of complex changes to scheduling, referrals and staffing, before translating changes to the 'real world.' This study determines if PSD increases Veteran access to the highest quality care, and if PSD better maximizes VA resources when compared against usual trial-and-error approaches to improving quality.
Background: Evidence-based practices (EBPs) are the most high value treatments to meet Veterans' addiction and mental health needs, reduce chronic impairment, and prevent suicide or overdose. Over 10 years, VA invested in dissemination of evidence-based psychotherapies and pharmacotherapies based on substantial evidence of effectiveness as compared to usual care. Quality metrics also track progress. Despite these investments, patients with prevalent needs, such as depression, PTSD and opioid use disorder often don't receive EBPs. Systems theory explains limited EBP reach as a system behavior emerging dynamically from local components (e.g., patient demand/health service supply). Participatory research and engagement principles guide participatory system dynamics (PSD), a mixed-methods approach used in business and engineering, shown to be effective for improving quality with existing resources.
Significance/Impact: This study is proposed in the high priority area of VA addiction and mental health care to improve Veteran access to VA's highest quality care. The PSD program, Modeling to Learn (MTL), improves frontline management of dynamic complexity through simulations of staffing, scheduling and service referrals common in healthcare, across generalist and specialty programs, patient populations, and provider disciplines/treatments.
Innovation: Recent synthesis of VA data in the enterprise-wide SQL Corporate Data Warehouse (CDW) makes it feasible to scale participatory simulation learning activities with VA frontline addiction and mental health staff. MTL is an advanced quality improvement (QI) infrastructure that helps VA take a major step toward becoming a learning health care system, by empowering local multidisciplinary staff to develop change strategies that fit to local capacities and constraints. Model parameters are from one VA source and generic across health services. If findings show that MTL is superior to usual VA quality improvement activities of data review with facilitators from VA program offices, this paradigm could prove useful across VA services. The PSD approach also advances implementation science. Systems theory explains how dynamic system behaviors (EBP reach) are defined by general scientific laws, yet arise from idiographic local conditions. Empowering staff with systems science simulation encourages the safe prototyping of ideas necessary for learning, increasing ongoing quality improvement capacities, and saving time and money as compared to trial-and-error approaches.
Specific Aims:
Methodology: This study proposes a two-arm, 24-clinic (12 per arm) cluster randomized trial to test for superiority of MTL over usual QI for increasing EBP reach. Clinics will be from 24 regional health care systems (HCS) below the SAIL mental health median, and low on 3 of 8 SAIL measures associated with EBPs. Computer-assisted stratified block randomization will balance MTL and usual QI arms at baseline using Corporate Data Warehouse (CDW) data. Participants will be the multidisciplinary frontline teams of addiction and mental health providers.
Next Steps/Implementation: MTL was developed in partnership with the VA Office of Mental Health and Suicide Prevention (OMHSP) and if shown to be effective, scalable, and affordable for improving timely Veteran access to EBPs, MTL will be scaled nationally to more clinics by expanding MTL online resources, and training more VA staff to facilitate MTL activities instead of usual QI.
Inclusion Criteria:
24 health care systems currently functioning below the median VA mental health recommendations for Strategic Analytics for Improvement & Learning (SAIL) and below the median for 3 of 8 SAIL evidence-based treatment approaches.
Exclusion Criteria:
Health care systems functioning above median VA mental health recommendations for Strategic Analytics for Improvement & Learning (SAIL) and below the median for 3 of 8 SAIL evidence-based treatment approaches. Only one health care system can be included per arm - MTL vs QI.