Primary Care and Population Health: Fourth in a Five-Part Series

Feb 6, 2013, 9:00 AM, Posted by Dave Chokshi, Nicholas Stine

Nicholas Stine, MD, and Dave Chokshi, MD, MSc, writing on behalf of the New York Academy of Medicine Primary Care and Population Health Working Group.

The New York Academy of Medicine is the National Program Office for the Robert Wood Johnson Foundation Health & Society Scholars program, which works to reduce population health disparities and improve the health of all Americans. The New York Academy recently conducted a survey of 17 thought leaders in primary care and population health. In the fourth of five blog posts, we share a synthesis of what those leaders had to say. All quotes are printed with permission.

file Nicholas Stine and Dave Chokshi

The Role of Primary Care and Clinical Incentives: Most discussants agreed that primary care and the patient-centered medical home (PCMH) movement are important foundations upon which to build broad-based population health activities. While acknowledging that these systems are already over-burdened with clinical responsibilities—and that current incentive structures are poorly aligned to accomplish this goal—many cited the degree of overlap in the missions of primary care and public health institutions as a starting point.

Community Care of North Carolina (CCNC), for example, evolved slowly and steadily over 25 years from a clinical quality network to a statewide multi-sectoral public-private partnership based on the PCMH. Allen Dobson, MD, described the key components as: the formation of cooperative provider networks; introduction of population management tools; case management; and data infrastructure with rapid feedback to providers. “Community Care is bottom-up and physician-led with respect to quality improvement…but because the collaboration includes the public health department, we are also looking at population metrics regardless of whether or not we’re managing that population.” Financing is organized using flexible per-member-per-month allotments that allow networks to put resources into quality measurement.  Dobson cited external evaluations demonstrating that the overall project is cost-saving, with CCNC responsible for nearly $1.5 billion in lower costs from 2007-09.

While the PCMH provides fertile ground for the development of a population health-oriented approach, most discussants were quick to point out that additional changes in measurement and incentive structures would be necessary to foster these efforts on a broader scale. “We’ve been very successful getting NCQA [National Committee for Quality Assurance] certification,” noted Henry Chung, MD, of Montefiore. “The real challenge is bridging the gap between certification and really making an impact on the population.”

The experience of Lloyd Michener, MD, at Duke echoes this distinction, highlighting the importance of community-based intervention:  “Our effort grew out of the provider community, but when we first started we didn’t see any changes in health care outcomes—for example on ER visits for asthma. First, we got all of the provider groups to agree to one standard protocol for asthma, which had no effect on ER rates. Then we started partnering with the community as well, and that’s when metrics started to budge. The community efforts led to a striking—as high as 80 percent—reduction in ER utilization.”

Michener also emphasized the utility of mapping interventions.  He described a sophisticated data merge funded by a National Institutes of Health Clinical and Translational Science Award grant via collaboration among the county health department and the Schools of Medicine and Environment at Duke.  Health department, Medicaid, crime rate, and health care system data were aggregated and de-identified, with mapping capabilities overlaid.  Thus far, geomapping analyses of ED utilization, teen pregnancy, HIV incidence, violent crime, and substance abuse have informed community discussions—and the design of targeted interventions for specific neighborhoods, such as community health worker placement.

An associated challenge is precisely which metrics to monitor.  “Metrics are in a very early stage, as most leaders in the field would agree,” according to David Stevens, MD. “But we need to start measuring and learn as we go.” Most discussants agreed with the strategy of starting with intermediate measures to push clinical systems into the community. For example, while merely assessing tobacco use status in the office may not go far enough, holding providers accountable for tobacco-related deaths from the outset may be challenging.  Thus rates of tobacco use in the community may be the more appropriate compromise measure.

Visit the RWJF Human Capital Blog on Friday for Part Five in this series, which will look at how best to address social determinants of health. See all the posts in the series here.

This commentary originally appeared on the RWJF Human Capital Blog. The views and opinions expressed here are those of the authors.