Volume 7: No. 6, November 2010
David A. Kindig, MD, PhD; Bridget C. Booske, PhD; Kirstin Q. Siemering,
DrPH; Brenda L. Henry, PhD; Patrick L. Remington, MD, MPH
Suggested citation for this article: Kindig DA,
Booske BC, Siemering KQ, Henry BL, Remington PL. Observations and
recommendations From the Mobilizing Action Toward Community Health (MATCH)
Expert Meeting. Prev Chronic Dis 2010;7(6):A124.
http://www.cdc.gov/pcd/issues/2010/nov/10_0132.htm. Accessed [date].
Introduction
In October 2009, authors, staff, and guest experts from the Mobilizing Action
Toward Community Health (MATCH) project and the Robert Wood Johnson Foundation,
the project’s funder, met in Madison, Wisconsin to discuss metrics, incentives,
and partnerships for population health improvement. Their essays were published
in this and the previous 2 issues of Preventing Chronic Disease
(www.cdc.gov/pcd/issues/2010/jul/toc.htm and www.cdc.gov/pcd/issues/2010/sep/toc.htm).
The plenary and small-group discussions were provocative and wide ranging. The
purpose of this commentary is to 1) summarize key themes from the essays and
meeting discussion and 2) present recommendations for future practice and
research regarding metrics, incentives, and partnerships to improve population
health.
Back to top
Discussion Themes
Metrics
Bilheimer and Pestronk presented commentaries on the metrics essays (1,2).
Meeting participants identified challenges related to population health metrics.
They recognized that the usefulness, reliability, and validity of metrics are
often compromised by limitations in available data. Examples of these
complicating factors include sparsely populated geographic areas, challenges
with survey methods (such as random-digit dialing in a cell phone era), and the
choice of unit of analysis.
Geopolitical areas such as counties or states are often used because they
are the focus of much of the available data, but these areas do not
necessarily reflect population health market areas where programs and
policies are implemented to improve health outcomes. Data intricacies add
complexity to analyses — as is illustrated by the fact that different health
determinants operate in different geographic areas (eg, school nutrition
policies are local, air quality policies are regional, and Medicare policies
are national).
Participants agreed that the population health field needs revised metrics to
address various goals.
- Population-based metrics to monitor changes in population health.
Most measures of population health (eg, those used in the County Health
Rankings)
are used to measure differences between geographic areas and often
combine several years of data to increase the precision of the estimates
(3). More precise metrics are needed to monitor trends over time and show
changes over short time frames in response to local-level changes in
programs and policies.
- Standard measures of health disparities within communities. Most
measures of population health can demonstrate disparities between
geographic areas (eg, the County Health Rankings), but more attention
needs to be focused on disparities within communities by using
different disparity domains such as race/ethnicity and socioeconomic
factors.
- Metrics that can be easily understood by the public and policy makers.
Many metrics that reflect the health of a population (eg, age-adjusted death
rates) are difficult to communicate to the public or to policy makers.
Approaches such as dashboards (which use graphics resembling gauges and
dial-type indicators) or rankings can improve communication and awareness or
generate action among targeted and broad audiences.
One participant suggested that, “A good measure makes you feel responsible
for taking action.” Another noted that measurement is an assertion of
responsibility; population health should be measured at appropriate levels so
that disparities are not masked and should include a wide set of measures so
that governments and other relevant entities (eg, business, education,
transportation) can take responsibility. Participants also preferred an
interpretable logic model so that audiences
understand the choice of metrics: Why is each measure important and what can be
done about it? What are the pathways, how can they be influenced, and at which
levels?
Incentives
McGinnis and Lewis provided commentaries on the essays that examined the use
of incentives to improve population health (4,5). Meeting participants discussed
the process of creating incentives to improve population health, and how
incentives should link to measures of desired outcomes. Although much of the
discussion focused on financial incentives, participants also addressed
nonfinancial incentives such as political gain or professional recognition. For
example, it was noted that California’s quality improvement in health care was
largely driven by public reporting and information sharing. The desire to
achieve such recognition on published lists may fuel innovative and sustained
change.
As a result of current private and public fiscal instabilities, perhaps
financial incentives should be directed toward identifying new resources or
redirecting existing ones. Would resources be one-time grants from government
and foundations, or would they be built into formulas like the community benefit
tax rules to ensure the long-term investments that would be needed?
Participants noted that incentives must be linked to individual or
organizational self-interests to affect change. Unfortunately, no consensus
exists on which specific incentives best motivate individuals, organizations,
and sectors and how factors such as values, ideology, and beliefs affect the
power of incentives at all levels. We need to better understand how incentives
have been used both successfully and unsuccessfully in education, welfare, and
other social systems. Although government entities generally adopt a directive (ie,
top-down) approach to incentives, incentives can also be effectively initiated
from the bottom up, in which individuals and investors decide how and where to
direct their resources.
Partnerships
Shortell and Bailey provided commentaries on the population health
partnership essays (6,7). Participants observed that partnerships are anything
but one-size-fits-all; they may be characterized across a spectrum of
collaboration ranging from cooperation to integration. Participants raised
various issues on the partnership theme.
- Identifying best practices in community partnerships. Given the
wide variability in partnership structure and function, participants wanted
to know if best-practice processes can be identified that apply across the
board (such as with respect to capacity building and strategic planning).
For example, do partnerships require a minimum level of formality to
effectively share power and drive action? What factors cause partnerships to
have a more formal structure and function?
- Sustaining partnerships. Participants wanted to know more about how
partnerships earn credibility and legitimacy over time and how community
institutions can prevent or resolve conflict that could hinder strong cross-sectoral
collaborations. For example, how are costs and benefits evaluated from the
perspective of prospective partners (transaction costs of formation vs
potential for synergy once established)?
- Balancing competing priorities. Participants asked how partnerships
could balance core competence (what they accomplish in an absolute sense
based on available expertise, skills, and resources) with comparative
advantage (what they can accomplish in a relative sense based on what they
do better than others). In addition, they wanted to know the degree to which
having a population health agenda shared (overtly or not) by sectors outside
health, what might motivate nonhealth sectors to come to the table, and
whether a multisectoral investment logic model could be developed for all
partners.
Participants noted that there is no substitute for effective leadership
throughout all phases of partnership. Without questioning the potential of
partnerships, they challenged the notion that partnerships are necessary for
improved population health. Participants did not doubt that multiple sectors
should be engaged in efforts to address the multiple determinants of health, but
several questioned whether improvement actually requires cross-sector work. In
other words, is it possible to effect substantial change through focused intrasector activity? One possible response is that the nature of the task at
hand often determines the level of cross-sectoral coordination required: solving
bigger problems is likely to require more interdependence, particularly the
sharing of resources.
Back to top
Recommendations for Practitioners
In breakout groups, participants identified 3 opportunities for future work
among practitioners: increasing investments in multiple determinants of
population health, establishing service bureaus to provide technical assistance,
and establishing an award for population health improvement.
Increase investments in the multiple determinants of population health
Discussion regarding investments centered on aligning resources and
incentives to drive investment in programs and policies that will improve health
outcomes and reduce disparities. Suggestions included developing investment
pools similar to those being tried by the California Endowment. The California
Endowment is using funds for intervention via multisectoral partnerships or
enhancing naturally occurring multisectoral initiatives. Such interventions
should require investments in the multiple determinants of health, including
income and educational policies and the built environment. To increase the
likelihood of success, meeting participants recommended focusing investments in
places where some partnership activity already exists and where infrastructure
is in place.
This recommendation has several challenges. For example, how should
investments be balanced between communities with the need and those with the
highest likelihood of success? Also, who will provide the necessary resources?
Although government, foundations, and business and community investments are
reasonable sources, discussion also focused on other sources that might be more
dependable and permanent, such as savings captured from waste on unnecessary
health care. Some discussion focused on the policy proposals for accountable
care organizations (ACOs) in Medicare, which could generate savings for
high-quality and low-cost care. Instead of only sharing savings with providers
and payers, a portion could be used as a community health dividend. The Vermont
Blueprint for Health (8) has used such an approach, and leaders in Minnesota
have called for nesting ACOs in accountable health communities. Participants
also suggested that the community benefit definition used by the Internal
Revenue Service be expanded to include the value of hospital investment in local
population health improvement that goes beyond charity care. The 2010 Patient
Protection and Affordable Care Law (Pub L No. 111-148) represents a step in the
right direction by requiring nonprofit hospitals to conduct a needs assessment
in consultation with the communities they serve at least every 3 years.
Establish technical assistance service bureaus
Many participants noted the lack of community capacity and expertise for
population health improvement activities such as using metrics to leverage
investment and create effective partnerships. Local or virtual technical
assistance could be provided to use data for health improvement, identify
evidence-based policies and programs, create processes to identify and implement
local interventions, set cost-effective priorities, and help community partners
recognize the need for cross-sector collaboration for health improvement. For
example, public and private funders could be more prescriptive in providing a
menu of evidence-based programs and interventions.
Establish a population health improvement award
The idea of a Baldrige-like (9) annual prize for communities excelling in
improving population health through creative use of incentives, metrics, and
partnerships was proposed. Participants noted that recognition of improvement
should take account of change over time and achievement or accomplishment at a
point in time.
Back to top
Recommendations for Researchers
Participants identified some major research needs and opportunities that
could move understanding and action forward in the population health field. They
included examining causal relationships between determinants of health,
increasing understanding of population health incentives, and increasing
understanding of population health partnerships.
Examine causal relationships between determinants of health
Participants recommended that funders should support research to examine the
cost-effectiveness of addressing different determinant categories and also
specific programs and policies. This research should also address secondary
health effects of nonhealth policies, for example by expanding the scope of
comparative effectiveness research to include determinants of health beyond
health care. In addition, research should be conducted to improve metrics that
can monitor changes in population health and to propose ways to balance
incentives for population health improvement. Researchers should also develop
more robust disparity measures for health outcomes and health determinants.
Increase understanding of population health incentives
Researchers should develop an expanded multisector population health
model so that leaders understand their roles, responsibilities, and most
cost-effective actions for population health improvement within and outside of
their own sectors. Research on these investments should also determine what
cross-sectoral financial and policy investment at the community level has been
successful in improving health. The information can then be used to develop
local (ie, substate) data sets for understanding these relationships.
Researchers should also determine the advantages and disadvantages of
applying incentives at different levels of aggregation (ie, individual vs
community vs organization), the advantages and disadvantages of using bundled or
unbundled metrics for applying incentives, and how to avoid poor performers
receiving penalties when they need resources to improve. Finally, research
should examine the scope of potential nonmonetary and monetary incentives for
population health in the United States and abroad.
Increase understanding of population health partnerships
Research should be conducted to better understand public- and private-sector
leaders’ attitudes toward population health improvement and tradeoffs. Where do
population health improvement and disparity reduction (in general) fall on their
priority list? Who (outside of the health community) is paying attention?
Research on partnerships should also identify the characteristics of
effective partnerships. How can they be developed, expanded, and sustained? Are
partnerships necessary for population health improvement, or can sectors operate
effectively alone? Which organizations are candidates to be integrators across
the population health model?
Back to top
Conclusion
The 2009 MATCH expert meeting generated thoughtful and stimulating discussion
around the essays presented in this and the previous 2 issues of Preventing
Chronic Disease. Far more questions were asked at the meeting than answered.
Through facilitated dialogue, participants offered wide-ranging ideas and
insights in the areas of metrics, incentives, and partnerships. The meeting
provided little time or space for many details; the format necessitated input in
rather broad brushstrokes toward the goal of building consensus for practice and
research priorities. As the essays and commentaries in this series attest,
improving population health will require effort on many fronts; no single track
to success exists. Whereas the challenges are substantial, the ideas shared here
should be reflected on, refined, expanded, and hopefully pursued through
empirical and applied efforts to improve population health.
Back to top
Acknowledgments
The October 2009 meeting was supported by the MATCH project with a grant from
the Robert Wood Johnson Foundation to the University of Wisconsin Population
Health Institute. We appreciate the assistance of Jessica Athens, Jenny Buechner,
Erika Cheng, and Joan Fischer.
Back to top
Author Information
Corresponding Author: David A. Kindig, MD, PhD, University of Wisconsin
School of Medicine and Public Health, Population Health Institute, 760 WARF, 610
Walnut St, Madison, WI 53726. Telephone: 608-263-4886. E-mail:
dakindig@wisc.edu.
Author Affiliations: Bridget C. Booske, Kirstin Q. Siemering, Patrick L.
Remington, University of Wisconsin, Madison, Wisconsin. Brenda L. Henry, Robert
Wood Johnson Foundation, Princeton, New Jersey.
Back to top
References
- Bilheimer LT. Evaluating metrics to improve population health. Prev
Chronic Dis 2010;7(4).
http://www.cdc.gov/pcd/issues/2010/jul/10_0016.htm.
- Pestronk RM. Using metrics to improve population health. Prev Chronic Dis
2010;7(4).
http://www.cdc.gov/pcd/issues/2010/jul/10_0018.htm.
- County health rankings. http://www.countyhealthrankings.org. Accessed
August 6, 2010.
- McGinnis JM. Observations on incentives to improve population health.
Prev Chronic Dis 2010;7(5).
http://www.cdc.gov/pcd/issues/2010/sep/10_0078.htm.
- Lewis S. Creating incentives to improve population health. Prev Chronic
Dis 2010;7(5).
http://www.cdc.gov/pcd/issues/2010/sep/10_0175.htm.
- Shortell SM. Challenges and opportunities for population health
partnerships. Prev Chronic Dis 2010;7(6).
http://www.cdc.gov/pcd/issues/2010/nov/10_0110.htm.
- Bailey SBC. Focusing on solid partnerships across multiple sectors
for population health improvement. Prev Chronic Dis 2010;7(6).
http://www.cdc.gov/pcd/issues/2010/nov/10_0126.htm.
- Hester J Jr. Designing Vermont’s pay-for-population health system. Prev
Chronic Dis 2010;7(6).
http://www.cdc.gov/pcd/issues/2010/nov/10_0072.htm.
- The Malcolm Baldrige National Quality Improvement Act of 1987. Pub Law
100-107. http://www.nist.gov/baldrige/about/improvement_act.cfm. Accessed
August 6, 2010.
Back to top
Comment on this
article at PCD Dialogue
Learn more about PCD's
commenting policy
