Part artic icip ipato atory ry S Syst ystem em D Dyna ynami - - PowerPoint PPT Presentation

Part artic icip ipato atory ry S Syst ystem em D Dyna ynami mics cs: Tri riang ngul ulat ating ng Ele lectr tron

• nic

ic He Heal alth th Reco ecord rds, s, St Stak akeh ehold

• lder

er E Expe xpert rtis ise a e and nd Sim imula lati tion

• n Mo

Mode deli ling g to to E Expa pand nd E Evid iden ence ce- Bas ased d Pr Prac actic ices es

Veterans Pending Tx Screen positive rate F rac of veterans who screen positive for PTSD depression or anxiety Screening appts per day Veterans first therapy appt completion rate F irst appt tx capacity F irst appt tx capacity ratio Effect of tx capacity on time to first therapy appt Average time to first therapy appt Standard therapy appt wait time Initial tx wait list Veterans Pending Second Appt Veterans second therapy appt completion rate Average time to second therapy appt <Standard therapy appt wait time> Effect of tx capacity on time to second therapy appt Second appt tx capacity Second appt tx capacity ratio Total appt capacity Proportion allocated to first tx appt Average time to complete EBT

@LZPhD

December 15, 2016

Lindsey Zimmerman, PhD

Clinical & Community Psychologist, Implementation Science National Center for PTSD, Dissemination & Training Division Affiliate Instructor, University of Washington School of Medicine

Lindsey.zimmerman@va.gov

1

David Lounsbury, PhD, Craig Rosen, PhD Rachel Kimerling, PhD, Jodie Trafton, PhD & Steve Lindley MD, PhD

These views are my own.

The views and opinions expressed herein do not necessarily state or reflect those

• f the United States Government.

2

The aim of our research:

• …is to develop a systems understanding
• f mental health service delays and how

they contribute to limited reach of evidence-based mental health care.

• …and empower mental health

stakeholders to make optimized quality improvement decisions with ex ante assessments of their proposed redesign solutions.

3

Future of Implementation Science

Brownson, Colditz & Proctor, 2012

“The next generation of studies…will address the sustainable integration of interventions within dynamic health care delivery systems and the implementation

• f evidence-based systems of care rather

than the individual intervention.”

• David Chambers (2012)

Setting - time, place, and circumstances in which something

• ccurs or develops

Organization - act or process of

• rganizing or of being organized

Context - interrelated conditions in which something exists or

• ccurs

Merriam Webster, 2016

5

Adoption as a situated individual and collective behavior: The interrelated time, place, and circumstances in which providers commit to and are organized to initiate an EBP.

6

System – A set of elements interconnected in such a way that they produce their own internal dynamics. The system, to a large extent, causes its

• wn behavior.

Meadows, 2008

7

Structure determines behavior. Structure determines behavior.

8

We define EBP reach as the proportion of patients with a PTSD, depression or SUD diagnosis who a) initiate b) timely EBP session c) complete an adequate, therapeutic EBP dose

9

Reach as a system behavior: the whole set of mechanisms by which the needs of the patient population are or are not addressed by their health care system.

10

There is strong evidence for EBPsy and EBPharm as the best approaches to meet patient demand in VA.

• Evidence-based Psychotherapy (EBPsy)

– Depression, PTSD, Substance Use Disorder (SUD)

• Evidence-based Pharmacotherapy

(EBPharm)

– Depression, Alcohol Use Disorder (AUD) and Opioid Use Disorder (OUD)

• EBPsy and EBPharm reduce PTSD and

depression symptoms, reduce alcohol or

• piate use and thereby, reduce risk of

chronic impairment, relapse, suicide and

• verdose.

11

Palo Alto VA Health Care System

• Main facility

and nine

• utpatient

clinics

• +85K patients

served

• >17,000 patients

receive MH care each year

12

10 20 30 40 50 60 70 80

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

Days from intake to first appointment New OEF/OIF PTSD Intakes

Wait Time 1 Goal

Exi xisti ting ng S Stat ate: e: I Inad adeq equa uate e pr prop

• port

rtio ion n of f pat atien ents ts g gett ttin ing g tim imel ely, y, hi high gh-qu qual alit ity y car are

13

Tar arget et S Sta tate: e: Lea ean SMA MART RT Go Goal al

By April 2015, 40% of patients newly seen in

• utpatient 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.

14

0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 100.00% 29-Sep 29-Oct 29-Nov 29-Dec 29-Jan 29-Feb 31-Mar 30-Apr 31-May 30-Jun 31-Jul 31-Aug % of new patients with 2nd appointment within 2 weeks

Implemented changes

Local improvement, but wide variability & goal not achieved.

15

0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0%

% of New Intake patients Two Psychotherapy Visits within one month of Intake

Gap between scheduling and completing psychotherapy remained.

16

Tar arget et S Sta tate: e: Lea ean SMA MART RT Go Goal al

By April 2015, 40% of patients newly seen in

• utpatient 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. Measurable. Attainable: if never achieved morale may suffer. Realistic: with the available resources. Time frame: A due date.

System dynamics makes complexity more manageable.

We look at interdependence among clinic components. We partner with frontline staff and leadership to learn together. Hard for any single person to grasp the interrelated whole system. Everyone will have multiple opportunities for input. Everyone has relevant expertise.

18

PSD em embeds a a susta tainable le QI proces ess and d tool tool in in the e local l settin ing

Scientific Model Problem Description of why quality problems persist. Subjective Learning Stakeholders cannot or do not learn and adapt to their situation. Coordination Conflict or lack of stakeholder consensus. Objective Analysis Policies are inconsistent with the real system constraints. Restructuring The underlying structure

• f the system prevents

workable solutions.

Drawn from Hovmand (2014)

19

Modeling saves time and effort.

We use simulation to see impacts of proposed changes

• n the whole system

in real time before we actually try to implement anything. We also determine when we should achieve our goal.

Average Time to Tx

20 15 10 5 10 20 30 40 50 60 70 80 90 100 Time ( Day)

Days Average time to case manage appt : Demo calibration Average time to med manage appt : Demo calibration Average time to first therapy appt : Demo calibration Average time to complete E BT : Demo calibration

20

Systems have properties of self-

• rganization, emergence & adaptation.

Chambers, Stange, & Glasgow, 2013 – Dynamic Sustainablity Framework 21

Systems have properties of self-organization, emergence & adaptation.

“It is not yet clear how best to tailor interventions and therefore not clear what the effect of an optimally tailored intervention would be.”

Baker et al., 2015 – Cochrane Database of Systematic Reviews

22

PSD: Addresses Multifinality and Equifinality

Menlo Park Stockton 3548 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 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 Referrals to other providers by necessity Multiple on-site specialty programs Only telehealth specialty care Training program site multiple disciplines No trainees providing care Most groups "open" (ongoing enrollment) Most groups "closed" (infrequent

• pening)

Shorter time to next available appointment Longer time to next available appointment

23

Reach as a system behavior: the purposes of subunits in a system may add up to an

• verall behavior no one wants;

changing elements usually has the least effect on the system.

Meadows, 2008

24

Systems have properties of self-

• rganization, emergence & adaptation.

Plan Do Study Act Plan Study Do Act

Damschroder et al., 2009 – Process domain: Engage, Plan, Execute, Evaluate Consolidated Framework for Implementation Research

25

System Dynamics provided the

• riginal conceptualization of a

learning organization.

26

Data drawn from health records.

Individual Psychotherapy 60-minute

Patient panel Average time in service Standard wait time to service Max Appointment Supply/Day % missed appointments Frequency of appointments Duration of appointment % of intakes starting service

Team Blue 168 120 7 1.8 30.2% 12 60 15% Team Green 239 120 7 3.7 15.6% 12 60 15% Recovery 107 120 7 5.9 11.8% 12 60 15% WCC 83 120 7 5.2 24.1% 12 60 15% PCT 112 120 7 8.7 17.9% 12 60 15% Trainees 238 120 7 0.7 28.4% 12 60 15% 947

27

Participatory System Dynamics Modeling Participatory System Dynamics Modeling

• 1. Stakeholder engagement & input

Mental models made explicit

5 10 15 20 25 30 35 40 2015 2020 2025 2030 Facility A Facility B

• 3. Implementation impacts

tested via simulation

• 2. System ‘behavior’
• bserved holistically

Administrative data, stakeholder estimates & research evidence

4 & 5. Re-design selected, implemented & tracked

• 1. Participate
• 2. Calibrate
• 3. Simulate
• 4. Translate
• 5. Evaluate
• 6. Iterate

28

Modeling can help a group of people get on the same page.

How, when, where patients flow. Modeling increases accuracy.

WE EACH USE OUR WE EACH USE OUR OWN MENTAL MODELS. OWN MENTAL MODELS.

29

Community-based 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; Israel, Eng, Schulz & Parker, 2013

30

National Center for PTSD National Program Evaluation Resource Center (OMHO/PERC) Veterans Engineering Resource Center (VERC) Core Modeling Group of Frontline Staff Full Staff Director of Outpatient Mental Health, MD Veteran Patients (VAPOR) Modeler, Consultant

TEAM PSD STAKEHOLDERS

Meets monthly all together; some workgroups meet more frequently

We also partner with national stakeholder We also partner with national stakeholder groups. groups.

31

BHIP, psychologist BHIP, RN SMI, psychologist BHIP & Telehealth, MD PCT, psychologist WCC, psychologist Clinic Coordinator, LCSW Director of Outpatient Mental Health, MD Facilitator, Implementation Scientist Modeler, Consultant

CORE MODELING GROUP

Meets every two weeks for one-hour.

We put a local stakeholder engagement process We put a local stakeholder engagement process in in place place.

32

Capacity–building – activity that builds durable resources & enables the setting to continue an evidence-based intervention after external support ends.

• EBP-specific Capacity
• General Capacity

Rabin & Brownson, 2012; Scaccia et al., 2015

33

34

35

The research question in our current effort:

Using simulation tests of stakeholder hypotheses about what procedural and policies changes would increase timely access to EBPs:

– Which proposals are most likely to align existing mental health resources to maximally increase reach of EBPs?

36

Stakeh eholder Hypoth theses: Testin ing poli licy/pro rocedure re chang nges.

BHIP to PCT: What if intake was in BHIP and PCT started treatment without another intake?

37

250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 3250 3500 1 5 9 13 17 21 25 29 33 37 41 45 49

EBP Sessions Completed EBP Sessions Completed Weeks Weeks

SC MHC Base Case (3027) SC Streamline PTSD Referral (3523) BHIP MHC Base Case (909) BHIP Streamline PTSD Referral (1171)

PCT Streamline Scenario

38

Participatory system dynamics for implementation planning.

39

Please email me for even more “tech specs”

PSD is innovative for implementation science.

• Generalizability. One process & tool to improve EBP

implementation in any local healthcare setting.

• Stakeholder Engaged. PSD is practice-based

research.

• Fit’ and system ‘capacity’ formally specified.

‘Mental models’ are not sufficiently precise for improved alignment of EBP implementation.

40

PSD is innovative for implementation science.

• Empirically quantified mechanisms or

implementation ‘barriers and facilitators.’ Causal attributions are structured in the model and validated with calibrated parameters from health system data.

• Simulation versus trial-and-error. Rather than

guesswork, wasted resources or unintended consequences, system impacts are tested before

• changes. Without simulation, implementation

strategies can only be improved via trial-and-error in the real world.

41

Next St Steps: I : Impleme ementatio tion Rese esearch h Test P PSD for

• r impro

roving E EBP rea each.

• Increase reach of EBP initiation.
• Increase reach of therapeutic EBP

dose.

• Increase EBP timeliness.

NI NIDA A R2 R21 Qu Quasi si- ex exper erim imen enta tal p pre re/p /pos

• st

ev evalu luat atio ion n

• Establish virtual facilitation

procedures.

• Refine code, algorithms, quality

checks.

NC NCPTS TSD Pi Pilo lot r rem emot

• te

e PS PSD w wit ith h vi virtu tual al fa facil ilit itat atio ion

42

R21DA042198-01

THANK YOU to Team PSD

lindsey.zimmerman@va.gov @LZPhd

Questions?

Co-Investigators Project Support Partners David Lounsbury McKenzie Javorka Menlo Park Mental Health Staff Ann LeFevre, Maya Kopell, Trisha Vinatieri, Bruce Linenberg, Pompa Malakar, Rosemarie Geiser, Sarah Walls, Gigi Fernandez, Emily Hugo, Martha Losch, Jessica Cuellar Craig Rosen Cora Bernard Veterans Advisory Partnership for Operations and Research Erik Ontiveros, Trent Van Dyke, Leroy Edwards, Tammy Thompson Rachel Kimerling Dan Wang Program Evaluation Resource Center Matthew Neuman, Matthew Boden, Hugo Solares, Shalini Gupta, David Wright, Susan Martins Jodie Trafton Swap Mushiana Veterans Engineering Resource Center Tom Rust Steve Lindley Alexandra Ballinger

43

Feed eedba back ck st stru ruct cture ure i in n sys syste tems ms: : Und nders rsta tand nding ng t the he co cont ntin inuum um o

• f

f car are e in in a a sys ystem em. .

44

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

45

“Best Guess" Regarding Key Drivers

• f Delays
• f Delays

PTSD/Depression/Anxiety referred to psychotherapy at intake unless actively refuse Ability to easily refer patients for psychotherapy with the fewest number of steps Schedulers need accurate initial psychotherapy slots Sufficient number of psychotherapy slots available

Grids not accurate No Standard Protocols Delayed referrals Variations in practice Multiple Rules