Top Banner
Lessons Learned about Embedding Complex Pragmatic Trials in Delivery Systems: Collaborative Care for Chronic Pain Lynn DeBar, PhD, MPH Kaiser Permanente Washington Health Research Institute Seattle, Washington Supported by NIH Common Fund and by NINDS through cooperative agreement (with NIDA scientific advisory support) (UH3NW0088731)
15

Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

Jul 07, 2020

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

Lessons Learned about Embedding Complex

Pragmatic Trials in Delivery Systems:

Collaborative Care for Chronic Pain

Lynn DeBar, PhD, MPH

Kaiser Permanente Washington Health Research Institute

Seattle, Washington

Supported by NIH Common Fund and by NINDS through cooperative

agreement (with NIDA scientific advisory support) (UH3NW0088731)

Page 2: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

PPACT Study Design & Rationale

Page 3: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

The “ask” from clinical and health plan leadership…

What do we do with the patients with

complex pain who “belong to

everyone and no one?” How do we keep our primary care

providers from burning out and

leaving the health care system?

Page 4: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

adapted witti permission from Rollin Gallagher

Page 5: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

700%

600%

500%

400%

300%

200%

100%

0%

-100%

+660%0 Spine fusion surgeries

+307o/o8 / Lumbar spine MRI imaging'

+249%6 Epidural + Facet injections

+65%0 Back pain-related medical expenditures

-19o/o 0 Self-re po rte functioning

Multidisciplinary pain treatment

--....---....---....-----------------.-----.----....---....-- centers 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015

Are We Using an Acute Care Model for A Chronic Condition?

Page 6: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

Addiction Medicine

Behavioral Health 11 - c:::::

Social ._W_o-rk---~ -1-P___.rimary C•e ~1-,,-1.,,...-_-_P-a=in=C=l-:::::in-ic----. ,.

PT /OT

Case Management

Sleep Clinic

Patient

~ P_h_ys_iat_7~I L ___ _ . Pharmacy

\\

Neurology I ~ - -Neurosurgery

Hospital

Membership Services

.,

Rheumatology I

Occupational Medicine

Emergency Department

Chiropractic Services

I Acupuncture

Primary Care

Interdisciplinary Pain Management

Pain Management in Usual Care Embedded in Primary Care

Behavioral

Health Coach: Nurse: Goal setting &

Lifestyle Changes Care

Coordination

Pharmacist:

Medication Physical Review

Therapist: Improved

Movement

Page 7: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

DeBar et al, Contemporary Clinical Trials, 2018; DeBar et al, Translational Behavioral Medicine, 2012

PPACT Overview

AIM: Integrate interdisciplinary services into primary care to help patients adopt

self-management skills to:

• Manage chronic pain (decrease pain severity / improve functioning)

• Limit use of opioid medication

• Identify exacerbating factors amenable to treatment

Focus on feasibility and sustainability

DESIGN: Cluster (PCP)-randomized PCT (106 clusters, 273 PCPs, 851 patients)

ELIGIBILITY: Chronic pain, long-term opioid tx (prioritizing ≥ 120 MED,

benzodiazepine co-use, high utilizers [≥ 12 visits in 3 months])

INTERVENTION: Behavioral specialist, nurse case manager, PT, and pharmacist

team; 12 week core CBT + adapted movement groups

OUTCOMES: Pain (3-item PEG), opioid MED, pain-related health services, and cost

Page 8: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

RES 'EARCH Open Access

Use of PRECIS ratings in the National Institutes of Health (NIH) Health Care Trials Systems Research Collaboratory

rm E. J §001 • Gil.-i Dem r • Glona D Coro o • mbers' . n Rund 15, Dav d 1-f 'Tl , Liu', Se n hn2, Cl henne Farr '

and Russell E. Gasgow

3.0 Flexibility (Del ivery)

ABATE Infection Eligibility s

Arialy>i<v tcruilment : 3

Outcome t 2 Setting ' l'

Followl.J p Organization

Adherence Dellwry

STOPCRC El igibility s

Selling

Organiution

Adherence Delivery

- - - Planning phase - Implementation phase

PPACT Elfgibility s

Setting

Organltalion

Adherence Delivery

TiME

Adherence Delivery

LIRE

Outco~ Setting

Folowup organization

Figure 3. PPACT PRECIS-2 Scoring

4.3 Eligibility

4.9 4.7Primary SettingOutcome

3.4

Fig. 1 PRECIS wheels as assessed by raters for ecJCh of the five trials at two time points. Ratings on a 1 - 5 seal 3.6 Flexibilitymore pragmati c ratings. The dashed line indicates the planning phase. The solid line Indicates the implementa (Adherence)

Johnson el al. Trials (2016) 17:32 00110 1186/s13063-016-1158-y

Page 9: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

Barriers Scorecard

Barrier Level of Difficulty

1 2 3 4 5

Enrollment and engagement of patients/subjects

X

Engagement of clinicians and Health Systems

X

Data collection and merging datasets X

Regulatory issues (IRBs and consent) X

Stability of control intervention X

Implementing/Delivering Intervention Across Healthcare Organizations

X

1 = little difficulty 5 = extreme difficulty

Page 10: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

“Opioids are what I would consider an adjunctive treatment. Everything we know about pain is that this is a complex biopsychosocial phenomenon,

and that we need to address the psychosocial contributors to pain.”

- Roger Chou (interviewed for Medscape, 5/1/2017)

Page 11: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

[p)~Y((e[H] (Q)~(Q)(e~~[L

BIO MEDICAL

#1 Barrier…. Tackling the Mind-Body Divide

Page 12: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

KPNW Cllusters

KPGA Clusters

KPHII Cllusters

IMI A M 20 14 - Year 2 2015 -Year 3 2 01 6 -Year4

JI J A s 0 N D J F IMI A M JI J A s 0 N D J F 1M A M J J A s 0 NI D J F 1M A

,::,, 1alb 4a.lb 7aJb 10alb I 13aJb -~ 2aJb 15alb Salb 11alb 14alb ',nj t= jaJb 6a/b I :fa/ ti 12a.lb l :JaJb I -

g, l alb 14alb l a.lb 10a.lb I l .lalb ·2 2alb Salb Sa.lb 11alb 14alb I .!iii t'!:e 3alb 6alb 9'alb 12alb I 15aJb I

>----

- -0, 1aJb 4alb I la.lb 10aib 13-alb I

.!a ,:= 2alb 5a.lb Ba.lb 111a1'b 14a.lb I ~ Ila.lb I l1i b I

I 19alb

1

111 I 15a.lb I

~UH3 lntlrJntiJn )imJline I I I I I I I I I I I I I I PPA

2014 - Year 2 2015 - Year J 2016 - Year 4

Mar ft(f)< May Jun Jul Aug Sep Oct Nov Dec Jan Fell Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar I ft(f)< I May I Jun I Jul I All!l Sep Oct Nov Dec Jan Feb

I I I I I

= 1ai'o -RKW 4 ab-RKW 7_,, -RKW 10ab 13a,'b I I 16"" 19a,'b

- >---- I 2a.'b -MTS Sab -MrS Bft -MTS 11 a'b 14al> 111a1i 20 .,, 3a.b-OOC ti.Jl~-CRC 9 .. . QRC 12ab 1Sa.b I I 1B•b - >---- I I I I I I I

f 1 •b 4ab I 11., 10aib 2al> s., I a.,,

3.,, s., I 9ab I I I I I

f 1a.'b -CRES 2aib - CLMB 5 a'b - cascade 8 MI- GwilnettA 11 ft -Glerllake A I \4ab -SolAIMoodB - >----3aib-CRES 6ab- T- A 9a'b-GwirnettB 12 aill -GleriakeB I I >----4aib -Pma 7a.l> -T-B 10 a'b- GwimeaC I I 13ab - Sodhwood A

I I I I I

f 1a.b-Mai.iiLani I I I I 2o1> -Maulari

I I l lft -\.ahana

I I I I I

f 1a'b-Qn, I I

I I 2•b -<Jahu I I l l ft -Oalu

I I I I I

f 1a.'o-~u I I c:::::J - Intervention Periods I I I 2•b -<Jahu c:::::J - (Intake visits1 Group sessions, Post-program visits} I I I 13 ""-0alu

I I I I I

The plan, the reality, & generalizable lesson learned

Page 13: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

If we knew then what we know now…the one predictable constant is change

• A sense of clinical urgency can lead to quick and sometimes unstable program shifts to which you may need to adapt

• Difference between “good” and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health integration)

• Stakeholder engagement is a continuous and intensive activity, requires two-way communication, and needs to be both top down and bottom up

Page 14: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

"-tftfd ltSt!llg the PPAC1111ltMD!JOl'I Ill Ont kPNW lifti<. T~WllhPCl\l,ithtMt xoltdNlk.-ldtfililNlpwtftbwho wouldbtllthcll'lllll!Mt.p,ogtlffl.~~.,.,llllftl .. lt tO!lducttdbJt~l\l,dbak,lnuM~ll,ph)'\ialtilfflpnt, tndpll,mlicbt

Th t MN"I of r,ffiU:htm rmnklltri, ■ .an llldiridulllld ~ plan UWC w1lgu6dttlWp,\•~tillCIPf'AC'TtMfflltwoughovttht),mc,nlhprog11111 P,1.,mti. s.y 1My.,..o,1tare plin1 !Mt JptM lo tt,,,, ....... •I UOIINidMlft.lt.yllltlhfpnicnibtailW,tidtntd>t\lhttr

llf strtGg!hl. wfid,ta thew p,r,,cw tlOtb to IGIIUgf p.itl illCI ~hlMgftltor ....... llfflCt,oqllUtrtl'ltctltlli1p,iol n

----.stoatii.p,ll, t pt 1IIIU\ ,yhkllll.

PP!t{,T'lun1 ~v nvrrv.,,e.,_fe_

/JS.A

Advice?... know what you are stepping into

• Local champions / surveillance invaluable

• Challenging the status quo requires persistent and vertical health care system partnership

• Rethink your process evaluation toolkit

Page 15: Lessons Learned about Embedding Complex Pragmatic Trials ... · Difference between “good”and “bad” contextual features can be a matter of timing (e.g., PCMH, behavioral health

Unique Benefits of the Collaboratory

• Very supportive group of investigators, CoC, and NIH personnel candid about challenges

• Great sounding board for helping one to construct most rigorous and interpretable trial possible

• Unique learnings from building partnerships with those in very different science domains