Helping Care Teams Improve Implementation of Helping Care Teams Improve Implementation of Medication Assisted Therapies for Alcohol and Medication Assisted Therapies for Alcohol and Opioid Use Disorders Opioid Use Disorders @LZPhD @LZPhD Lindsey Zimmerman, PhD Lindsey Zimmerman, PhD Office of Mental Health and Suicide Prevention National Center for PTSD, Dissemination & Training Division @DLounsburyNYC DLounsburyNYC David Lounsbury, PhD David Lounsbury, PhD Division of Community Collaboration & Implementation Sci Created by Epidemiology & Population Health Albert Einstein College of Medicine mtl.info@va.gov mtl.info@va.gov 1
Workshop Learning Objectives Time Time Workshop Workshop Focus Focus Modeling to Learn (MTL) Modeling to Learn (MTL) Helping Care Teams Improve Helping Care Teams Improve 1:00 PM – 2:30 PM 1:00 PM – 2:30 PM Implementation of Medication Implementation of Medication Assisted Therapies for Alcohol Assisted Therapies for Alcohol and Opioid Use Disorders and Opioid Use Disorders 1. Introduce principles of systems science systems science that can be applied to improve implementation of evidence-based pharmacotherapy (EBPharm or MAT). implementation 2. Describe why participatory learning participatory learning from simulation improves team's decision-making related to MAT. 3. Demonstrate the Medication Management (MM) Medication Management (MM) module of MTL. 4. Illustrate how simulation learning, using hyper-local team data, helps to identify the best way to optimize local MAT implementation resources optimize local MAT implementation resources. 2
mtl.how/team 3
The problem of EBP reach in teams: How can we reach more patients with our highest quality care? Other services Evidence-based practices 28% 72% Source: VA Strategic Analytics for Improvement and Learning, FY 2017
Veterans Health Administration Model of a US National Health Care System American J. Public Health 97, 2007 1. VA innovates with national dissemination efforts to train providers in evidence-based mental health practices 2. Enterprise-wide quality measures 3. Clinical practice guidelines and mandates for evidence-based care 4. National electronic health information system 5. Mental health care coordinated in multidisciplinary teams
What works to improve EBP reach, why, and under what conditions? xkcd.com Understanding causes of EBP reach, in local context, is critical to our stakeholders .
Our aims. • develop a systems understanding of VA mental health services and the limited reach of evidence-based mental health care. • empower mental health stakeholders to make locally optimized quality improvement decisions.
Target State: Target State: Lean Lean SMART Goal SMART Goal By April 2015, 40% of patients newly seen in outpatient mental health at Menlo Park for depression, PTSD, or anxiety disorders will have two psychotherapy visits completed within 28 days from time of intake assessment. Specific. Specific. Measurable. Measurable. Actionable Actionable : : if never achieved morale may suffer. Realistic Realistic : : with the available resources. Time frame Time frame : : A due date.
Local clinic strategies are needed to address local differences. Clinic 1 Clinic 1 Clinic 2 Clinic 2 3548 unique patients/year 3548 unique patients/year 2043 unique patients/year 2043 unique patients/year Lower caseload per provider Higher caseload per provider Rare wait for initial appointment Occasional waitlist to get into clinic 5.2 psychiatrists per 9 EBPsy providers 5.2 psychiatrists per 9 EBPsy providers 3.0 psychiatrists per 4 EBPsy providers 3.0 psychiatrists per 4 EBPsy providers Higher EBPsy providers/MD ratio Lower EBPsy provider/MD ratio Higher EBPsy base rate Higher EBPharm base rate Providers often self refer for EBPs Providers often self refer for EBPs Referrals to other providers by necessity Referrals to other providers by necessity Multiple on-site specialty programs Multiple on-site specialty programs Only telehealth specialty care Only telehealth specialty care Training program site multiple disciplines Training program site multiple disciplines No trainees providing care No trainees providing care Most groups "open" (ongoing Most groups "open" (ongoing Most groups "closed" (infrequent Most groups "closed" (infrequent enrollment) enrollment) opening) opening) Shorter time to next available Longer time to next available appointment appointment 9
MTL MTL focuses on learning among focuses on learning among frontline frontline teams teams making EBP-related care decisions. making EBP-related care decisions. Drawn from Hovmand 2014 & Scaccia et al., 2015 Drawn from Hovmand 2014 & Scaccia et al., 2015 Scientific Scientific Problem Problem Why problems persist Why problems persist Model Model Stakeholders cannot or do not Learning learn and adapt to their situation. General General Conflict or lack of stakeholder Capacity Capacity Coordination consensus. Policies are inconsistent with Analysis the real system constraints. EBP Specific EBP Specific The underlying structure of the Capacity Capacity Restructuring system prevents workable 10 solutions.
We define limited EBP reach among our patient population as a system behavior. Other services Evidence-based practices 28% 72% Source: VA Strategic Analytics for Improvement and Learning, FY 2017
Saturation achieved during structural behavioral validity testing . Barlas, 1996
National Center for PTSD Office of Mental Health & Suicide VA Employee Education Services Prevention OUR STAKEHOLDERS OUR STAKEHOLDERS VA policy-makers, patients, and providers from psychiatry, psychology, social work, nursing & certified peer support specialists Office of Healthcare Veteran Patients (VAPOR) Transformation Directors of Outpatient Mental Core Modeling Group of Frontline Health & VISN MH Leads Staff Frontline Teams 13
VAPOR introduces MTL https://mtl.how/intro 14
Our PSD approach – Participatory Research: A partnership approach to research that equitably involves stakeholders in all aspects of the research process and in which all partners contribute expertise and share decision-making and ownership. Hovmand, 2014; Oetzel et al., 2018 15
Participatory Research is an epistemology. • Scientific inquiry that that actively considers the scope of current knowledge, its limits and validity. • Participatory research asks, what knowledge is privileged or absent? 16
Virtual Facilitation Transparent Local Real-time Data Simulation 1. Equitable access to resources. 2. Mutual learning. 3. Shared decision-making.
Theory of Change Theory of Change 19
MTL resources help teams look back two years and look ahead two years. 20
Why is PSD effective? Participatory learning to develop ‘ Systems Thinking .’
“Staff” and “Time” costs as dynamics. mtl.how/sim
Causal mechanisms (dynamics) are made transparent for local learning. Red = - Read in from existing team data - Standardized mtl.how/demo 23
MTL tools helps frontline staff find the best local changes faster. mtl.how
MTL shows whether things may get better before worse or worse before better. mtl.how 25
Is PSD/MTL effective for improving EBP reach? Strong signal in R21 pilot clinics. 36 mos. 36 mos. sustained *HCS = Regional sig. improvement health care system + 3 SD ( α = .003) 20 mos . sustained 20 mos sig. improvement + 3 SD ( α = .003) Key: Green = Upper control limit (UCL) Red = 12-month pre-PSD EBP proportion Purple = Lower control limit (LCL) SD = standard deviations
Measure twice cut once. Measure twice cut once. Look before you leap . Look before you leap
Principles of the open science movement: • collaborative • free and open • transparent and reproducible science. 28
mtl.how 29
You can review Modeling to Learn session guides at mtl.how Session guides, links, and cheatsheets. 30
Participatory Learning to develop Systems Thinking. Partner Build Apply 31
We developed a secure website for reviewing team trends over time. 32
MTL resources help teams look back two years and look ahead two years. 33
34
Why is Modeling to Learn effective? 35
Start a new Medication Management Session. 1. Review the team data and “i” information. 2. Zoom in/out to review system stories and complexity reveals for each care setting. 3. Run, examine the output, and save a base case of no new decisions. 36
Select an MM Learning Mode. 37
Modeling to Learn helps teams manage tradeoffs within existing staff resources. 38
Can we increase the number of patients with OUD that receive EBPharm without increasing wait-times for other patient needs? 39
Not all medication management staff resources are the same. 40
Team Question: Can we increase the number of patients with OUD that receive EBPharm without increasing wait-times for other patient needs? 41
Recommend
More recommend
