1. 1 Evaluation of Patient-Centered Medical Home (PCMH) Initiatives Meredith B. Rosenthal, PhD February 24, 2009

2. 2 Acknowledgments  My work on patient-centered medical homes is supported by The Commonwealth Fund, a national, private foundation based in New York City that supports independent research on health care issues and makes grants to improve health care practice and policy and the Colorado Trust.  Many of the ideas articulated here have been shaped by conversations with the Patient Centered Medical Home Evaluators’ Collaborative, also supported by the Commonwealth Fund.  The views presented here are those of the author and not necessarily those of The Commonwealth Fund, the Colorado Trust, their directors, officers, or staff. 2

3. 3 PCMH Initiatives Are Proliferating  PCPCC reports more than 20 private sector-initiated PCMH pilots  Medicare demonstration  Numerous existing and emerging Medicaid initiatives  Very high aspirations for impact: On access On quality On cost

4. 4 What is the Evidence Base?  Body of literature on value of primary care (see, for example, Starfield review)  Published studies of impact of Wagner’s Chronic Care Model  Reports of successful initiatives that shared some elements of current PCMH Community Care of North Carolina Geisinger Health System

5. 5 What is missing?  Previous studies lacked standardized definition of the intervention – most initiatives are now trying to adhere to the Joint Principles and use the NCQA PPC- PCMH to measure “medical homeness”  More rigorous attempt to isolate the impact of the PCMH from other factors: most of the current data are cross- sectional  Answers to a much richer set of research questions

6. 6 Some Key Research Questions Current Evaluations Will Address  Do practices that conform to PCMH criteria deliver: better quality of care? better patient experiences? lower total cost? improved physician and staff satisfaction?  What does it take to turn practices into PCMHs?  Is there a business case for the PCMH – for payers? For providers?

7. 7 Challenges for Evaluation  Selection of outcome measures is complicated by: High level of aspirations for the PCMH Differing views among stakeholders on what the PCMH should and can deliver Relatively short time frame of most pilots  Interventions will evolve as experience grows (moving target)  Valid comparison groups may be hard to identify  Some pilots are relatively small

8. 8 Selecting Meaningful Measures  Expectations have already been set for the PCMH  Need a more fully developed logic model of what the PCMH implies for: what physicians and other staff will do differently how patients access care, manage their own health  We need to identify “leading indicators” of beneficial changes in accordance with this logic model: e.g., time to first ambulatory visit after a hospitalization

9. 9 PCMH Evaluators’ Collaborative  Researchers actively engaged in a PCMH evaluation  Objectives: Reach consensus about a standard set of data collection instruments Reach consensus about a standard, core set of outcome measures Share the Collaborative's consensus on instruments, metrics and/or methodological lessons with interested researchers around the country through public venues Foster an ongoing and supportive exchange where evaluators share ideas that improve their evaluation designs

10. 10 Preliminary Work of the Collaborative  Quarterly plenary conference calls  Measurement workgroups will convene between calls to propose standards for: Patient experience Physician/staff experience Medical homeness Clinical quality Cost/efficiency Process/implementation metrics  Proposed measure sets to be vetted with larger group of stakeholders

11. 11 Standardizing measurement: patient experience  A number of validated instruments are available, including: Clinician Group CAHPS Ambulatory Care Experience Survey Primary Care Assessment Survey  Each has strengths and limitations  All tend to focus on a single doctor rather than the team concept built into the PCMH  Some elements of PCMH such as non-visit care are not well represented

12. 12 Physician/Staff Surveys  Physicians and practice staff can report on both the objective functioning of the practice (How often do you have access to medication lists?) and subjective experience  Validated instruments exist including: Physician Worklife Survey (Linzer et al) Minimizing Error, Maximizing Outcomes study (Also Linzer et al.)  Key domains are addressed: which questions will be most salient/sensitive?

13. 13 Measuring Medical Homeness: Structures and Processes  NCQA PPC-PCMH is the basis for judging “medical homeness” in most pilots – for payment, participation  Evaluations also will use these survey results  Additional data collected through open-ended interviews and structured instruments  Workgroup will convene to identify additional dimensions of the PCMH consider a short instrument that could be administered between cycles of the PPC-PCMH and in a comparison group of practices

14. 14 Opportunities in the Current Landscape  Support of PCMH by a broad group of stakeholders  Most pilots have evaluations built in  Collective endeavor to develop common measures may permit meta-analysis across pilots to increase ability to find effects  Variety of evaluation contexts will increase generalizability of findings