What does the facilitation implementation strategy look like in the - - PowerPoint PPT Presentation

What does the facilitation implementation strategy look like in the real world? Evidence from the ADEPT study

Shawna N. Smith, Ph.D. University of Michigan

9th Annual Conference on the Science of Dissemination & Implementation in Health 14 December 2016

Overview & orientation

• 45 community-based sites working with an

external facilitator (EF) to implement a mental health collaborative care model (CCM) for patients with mood disorders

• Using unique data on EF tasks and

interactions with sites, we describe what facilitation looks like in the “real world”

Barriers to CCM adoption in community-based practices

• Consumers/Famil

Consumers/Families ies

– Lack of car Lack of care coor e coordination dination – Most privately insur Most privately insured seen in small practices ed seen in small practices

• Pr

Providers

• viders

– <30% of MSWs have access to EBPs <30% of MSWs have access to EBPs (Insel, 2009) – Inconsistent EMR use Inconsistent EMR use

• Systems

Systems

– Primarily FFS, few bundled payment models Primarily FFS, few bundled payment models – ACOs--mental health car ACOs--mental health care? e? (O’Donnell, 2013)

Facilitation as a tool for addressing barriers

Multi-faceted implementation strategy that focuses on “helping rather than telling”; PARiHS framework Facilitators aid sites by:

• Assessing barriers to uptake
• Guiding problem-solving in context of need

for improvement

• Ensuring communication and priority

alignment with site leadership

Rycroft-Malone et al., 2002, 2004; Kirchner et al., 2014

Flavors of Facilitation

External facilitator (EF):

• Location: Off-site, research team member
• Topical focus: Benchmarking, coaching
• Role: Confidante, outside observer

Internal facilitator (IF):

• Location: On-site, direct report to leadership
• Topical focus: Leveraging, rapport-building,

internal recognition, sustainability

• Role:

Role: Inside expert, champion

Kilbourne AM et al. 2013; 2014; Goodrich et al. 2012

Research questions

Using data from ADEPT:

• [How] do sites make use of EF?
• What does content of EF look like?
• Does EF content differ by presence of IF?

The ADEPT Study

Non- Responders

(<50% of pa+ents receiving >=3 LG sessions) k=45 sites Add External Facilita+on

REP+EF

k=23 sites Add Internal & External Facilita+on

REP+EF/IF

k=22 sites R Con+nue REP

k=5 sites

Cont REP+EF k=7 sites Cont REP+EF/IF

k=15 sites

Responders Non- responders Non- responders R Add IF

k=11 sites

Con+nue REP (A) Con+nue REP+EF/IF (C) Con+nue REP+EF (B) Con+nue REP+EF/IF (E) Responders

Month 18 and 24

Assessments

Con+nue REP k=7 sites Con+nue REP (D) Con+nue REP Con+nue REP (F) Con+nue REP

Month 6

Assessment

Month 12

Assessment

Phase 2 Follow Up Study Start

REP

k=79 sites

Responders

k=20 sites

Run-In Phase

All sites offered REP to implement EBP; Pa+ents start EBP by

Month 3

Kilbourne et al, 2014; Funding: NIMH R01 MH99899

Life Goals CCM

Decision Decision Support Support Self-management Self-management Access/Continuity Access/Continuity

Pract Practice Guidel ice Guidelines: ines:

Best practice txt, Best practice txt, metabolic syndr metabolic syndrome

• me

for mental health for mental health conditions conditions

Car Care management: e management:

Registry tracking (Symptoms, QOL, functioning Registry tracking (Symptoms, QOL, functioning)

)

General Medical Pr General Medical Provider Liaison

• vider Liaison

Life Goals Sessions: Life Goals Sessions:

CVD Risk, Symptoms CVD Risk, Symptoms Healthy Behaviors Healthy Behaviors Pr Provider Engagement

• vider Engagement

Kilbour Kilbourne et al. Psych Serv 2008; 2013 ne et al. Psych Serv 2008; 2013

The ADEPT Study

45 ADEPT sites 45 ADEPT sites received EF for at least 6 months:

• 5 sites received EF only for 6 months
• 7 sites received EF only for 12 months
• 11 sites received EF only

EF only for 6 months followed by EF+IF for 6 months

• 7 sites received EF+IF for 6 months
• 15 sites received EF+IF for 12 month

12

EF only sites

22

EF+IF only sites

11

delayed EF+IF sites

Data & Methods

Data: Study EF logged all tasks, categorizing:

• Mode (email, phone, in person)
• Personnel interacted with
• Duration (5 minute intervals)
• Primary content focus

Methods:

• Describe mode, duration and content of EF tasks
• Examine differences in EF tasks across sites

receiving EF and EF+IF sites, with focus on strategic content

EF Tasks

Initiation & benchmarking Assess context, identify needs & barriers Coaching Develop rapport, work with providers to address barriers Leveraging Identify local leadership priorities, identify additional site champions to promote LG Implementation plan development Create measurable goals, specify tasks & timeline Link to outside resources Connect with similar sites, REP technical assistance Education Requirements & expectations of personnel Ongoing marketing Develop business plan, focus on sustainability Reinforcement Positive recognition and support Consultation Discuss issues with facilitation experts Background Site research, often non-interactive Kirchner et al, 2013

EF Tasks: Strategic

Initiation & benchmarking Assess context, identify needs & barriers Coaching Develop rapport, work with providers to address barriers Leveraging Identify local leadership priorities, identify additional site champions to promote LG Implementation plan development Create measurable goals, specify tasks & timeline Link to outside resources Connect with similar sites, REP technical assistance Education Requirements & expectations of personnel Ongoing marketing Develop business plan, focus on sustainability Reinforcement Positive recognition and support Consultation Discuss issues with facilitation experts Background Site research, often non-interactive

Results

1,037 tasks were logged by EF for 45 sites between January 2015 and July 2016 Median EF time logged per site: 36 minutes/month [IQR: 25-43] Mode: 664 (64%) email (mean duration: 6 minutes) 353 (34%) phone (mean duration: 23 minutes)

Results: EF tasks, by month

20 40 60 80 100 120 50 100 150 200 250 1 2 3 4 5 6 7 8 9 10 11 12 13

Mean Time Per Site Mean Time Per Site Number of T Number of Tasks asks Mont Month

Number of Tasks Mean Time Per Site

Results: EF vs. EF+IF sites

Who did the EF interact with?

10 20 30 40 50 60 70 80

Admin Supervisors Providers IFs Other Per Percent of interactive tasks cent of interactive tasks

EF (N=564) EF+IF (N=473)

Results: EF vs. EF+IF sites

How long did the EF spend per site?

• Average EF sites:

– Overall mean: 260 minutes – Controlling for time: 12.6 minutes/day

• Average EF+IF sites:

– Overall mean: 429 minutes – Controlling for time: 21 minutes/day

*Difference not significant at p<0.05 after controlling for time

Results: EF vs. EF+IF

EF (N=564) EF+IF (N=473) Initiation & benchmarking 42%

• 17%
• Coaching

0.6% 1.1% 0.7% 0.9% Leveraging 0.2% 0.4% 3% 4%* Implementation plan development 3% 6% 6% 8% Link to outside resources 0.4% 0.8% 1% 1.5% Education 25% 51%* 35% 44% Ongoing marketing 0.6% 1% 3% 4%* Reinforcement 11% 23% 28% 35%* Consultation 8% 17%* 2% 2% Background 10%

• 4%
• *Difference significant at p<0.05 or better

Results: EF vs. EF+IF

EF (N=564) EF+IF (N=473) Initiation & benchmarking 65 minutes 51 minutes Coaching 1 4 Leveraging 2 13* Implementation plan development 13 43* Link to outside resources 1 2 Education 70 166* Ongoing marketing <1 9* Reinforcement 28 59 Consultation 26* 7 Background 73 37 *Difference significant at p<0.05 or better

EF sites spent an average of 16 minutes on strategic initiatives, EF+IF sites an average of 64 minutes.

Conclusions

• Use of EF and EF+IF strategies in

community-based implementation studies is feasible, scalable feasible, scalable

– EF logged 25-43 minutes/month with each site

• EF tasks at sites with & without IF were

largely similar, not significantly different in time commitment

• Sites with EF+IF did spend more time on

strategic tasks than EF sites, which may lead to longer-term sustainability of LG adoption

Next Steps

• Differences in study outcomes by EF & EF+IF

– ADEPT primary aim: EF vs. EF+IF as best ‘step-up’

• ption for sites non-responsive to REP?

– Delayed effect of adding IF in Phase 3 rather than starting with EF+IF in Phase 2?

• Mechanisms of effectiveness

– Relationship between EF interaction content/time and effectiveness? – Combine EF and IF activity logs to data to better elucidate mechanisms

• Cost-effectiveness of EF vs. EF+IF

– Is IF worth the additional cost compared to EF alone?

Acknowledgements

University of Michigan: University of Michigan: Amy Kilbourne, PhD, MPH Myra Kim, ScD Daniel Eisenberg, PhD Brian Perron, PhD Kristina Nord, LMSW David Goodrich, EdD Celeste Liebrecht, LMSW Julia Kyle, LMSW Michelle Barbaresso, MPH Peggy Bramlet, MEd Karen Schumacher, RN Zongshan Lai, MPH Nicholas Bowersox, PhD Kristen Abraham, PhD Danny Almirall, PhD Joanna Bratton, BSW University of Colorado-Denver: University of Colorado-Denver: Marshall Thomas, MD Robert Bremer, PhD Jeanette Waxmonsky, PhD Jenny Rementer, MA University of Pit University of Pittsbur tsburgh: gh: Harvar Harvard/V d/VA Boston: A Boston: Mario Cruz, MD Mark Bauer, MD Ronald Stall, PhD Christopher J. Miller, PhD Laura Bajor, DO University of Arkansas: University of Arkansas: CDC: CDC: JoAnn Kirchner, MD Mary Neumann, PhD John Fortney, PhD Fund Funding: ing: NIMH R01 MH99899 NIMH R01 MH99899

Extra Slides

Implementing CCM requires aligning organizational and financial incentives

Or Organizat ganizational ional Incent Incentives ives Al Alignment: ignment:

Pract Practice r ice redesign edesign Cl Clinical tools inical tools Service agr Service agreements eements Reimbursement models Reimbursement models

Financial Financial Incent Incentives ives

Depr Depression in Primary Car ession in Primary Care: Linking Clinical and System Strategies Pr e: Linking Clinical and System Strategies Program, Robert W

• gram, Robert Wood
• od

Johnson Foundation (PI: HA Johnson Foundation (PI: HA Pincus Pincus); ); Grazier Grazier KL, Klinkman MS., APMH 2006; Kilbour KL, Klinkman MS., APMH 2006; Kilbourne, ne, Thomas, et al 2006 Thomas, et al 2006

Implementation strategies can improve uptake

Highly-specified, systematic processes used to implement treatments/practices, often at the clinic or provider level, into usual care settings

Alignment requires more than guideline dissemination

– Buy-in from providers, healthcare leaders – Understanding barriers, facilitators to adoption

Life Goals CCM

An evidence-based, brief intervention based on principles of CCM, Motivational Interviewing, and combining best practices in self-management education and care management.

Six RCTs of Life Goals showed improved physical and mental health outcomes (e.g., cardiometabolic risk, depressive symptoms, overall functioning)