In 2014, the Commonwealth Center for Governance Studies released the report Improving Community Health Through Hospital-Public Health Collaboration: Insights and Lessons Learned from Successful Partnerships (Prybil et al., 2014). An overview of the report was provided by Lawrence Prybil, principal investigator for the study and the Norton Professor in Healthcare Leadership and associate dean of the College of Public Health at the University of Kentucky. Nicole Carkner, executive director of the Quad City Health Initiative, then shared the perspective as 1 of the 12 partnerships featured in the report. Following the presentations, a discussion was moderated by Sunny Ramchandani, commander and medical director of the Healthcare Business Directorate at the Naval Medical Center San Diego. Box 4-1 provides an overview of session highlights.
Prybil began by explaining that this study of hospital–public health partnerships stemmed from a conversation at the Keeneland Conference on Public Health Services and Systems Research in 2012. Rich Umbdenstock, president of the American Hospital Association, Paul Jarris, executive director of the Association of State and Territorial Health Officials (ASTHO), and Robert Pestronk, executive director of the National Association of County and City Health Officials (NACCHO), wanted to identify, examine, and extract lessons learned from successful operational
Key Themes of the Session on Hospital–Public Health Collaboration
- The Prybil et al. (2014) study identified eight characteristics of successful hospital–public health partnerships
- Having one or more anchor institutions, as in the case of the Quad City Health Initiative, is extremely important for sustaining collaboration (Carkner, Prybil)
- Why invest in collaboration? No one can improve the community’s health alone. A broad collaborative infrastructure and capacity are required (Prybil)
- Community-based partnerships are “a way to . . . galvanize communities, citizens, businesses, schools, and other parties to pay more attention collectively to improving the health of the community” (Prybil)
- To convince other potential partners, the message is that the work of collaboratives such as Quad City Health Initiative is about community development and community involvement (Carkner)
- The contributions of the hospital and health system community are necessary, but not sufficient, and other partners are needed (Prybil)
partnerships involving hospitals and public health departments focused on improving the health of their communities. They approached Prybil and Douglas Scutchfield of the University of Kentucky to conduct the study.
The first step was to find examples of successful hospital–public health collaboratives, as no list of such partnerships existed, Prybil said. The researchers developed a set of eight core characteristics of successful partnerships and broadly disseminated an announcement inviting nominations of partnerships with those characteristics for participation in the study (summarized in Box 4-2 and discussed in full, including 27 indicators, in Prybil et al., 2014). He pointed out that the characteristics of successful partnerships developed for the study were similar to those discussed by Jarris and others during the workshop, such as the importance of culture and trust. “Partnership” referred to a situation where independent parties came together to jointly address a common purpose and therefore included organizations calling themselves alliances, consortia, and many other designations. More than 160 nominations were received from across 44 states. These were examined to identify a set for study that were “highly successful” relative to the predefined core criteria. Sixty-three nominees were contacted for additional information about metrics and impact, and out of 17 finalists, 12 partnerships were
Core Characteristics of Successful Hospital–Public Health Partnerships Identified by Prybil and Colleagues (2014)
- Vision, Mission, and Values—The partnership’s vision, mission, and values are clearly stated, reflect a strong focus on improving community health, and are firmly supported by the partners.
- Partners—The partners demonstrate a culture of collaboration with other parties, understand the challenges in forming and operating partnerships, and enjoy mutual respect and trust.
- Goals and Objectives—The goals and objectives of the partnership are clearly stated, widely communicated, and fully supported by the partners and the partnership staff.
- Organizational Structure—A durable structure is in place to carry out the mission and goals of the collaborative arrangement. This can take the form of a legal entity, affiliation agreement, memorandum of understanding, or other less formal arrangements such as community coalitions.
- Leadership—The partners jointly have designated highly qualified and dedicated persons to manage the partnership and its programs.
- Partnership Operations—The partnership institutes programs and operates them effectively.
- Program Success and Sustainability—The collaborative partnership has been operational for at least 2 years, has demonstrated operational success, and is having positive impact on the health of the population served.
- Performance Evaluation and Improvement—The partnership monitors and measures its performance periodically against agreed upon goals, objectives, and metrics.
SOURCE: Prybil et al., 2014, p. 6.
ultimately selected to be studied in depth. These 12 were from 11 states across the country, with very different missions, he said.
The study process included a 2-day site visit to each of the 12 partnerships. Prybil noted with gratitude the cooperation, candor, and interest the study team experienced at all of the sites. Site visits included one-on-one, confidential, in-depth interviews; small group discussions; and review of documents and records. The final phase of the study involved processing, verifying, tabulating, and analyzing the data collected.
The study also examined research on partnerships with a focus on the extent to which they survived and succeeded or did not. Studies suggest that about half of partnerships in general, across a variety of sectors, survive and succeed, Prybil said. A more in-depth analysis of these studies shows that partnerships that display many of the core characteristics out-
Recommendations from Prybil and Colleagues (2014)
- To have enduring impact, partnerships focused on improving community health should include hospitals and public health departments as core partners but, over time, engage a broad range of other parties from the private and public sectors.
- Whenever possible, partnerships should be built on a foundation of preexisting, trust-based relationships among some, if not all, of the principal founding partners. Other partners can and should be added as the organization becomes operational, but building and maintaining trust among all members is essential.
- In the context of their particular community’s health needs, the capabilities of existing community organizations, and resource constraints, the parties who decide to establish a new partnership devoted to improving community health should adopt a statement of mission and goals that focuses on clearly defined, high priority needs and will inspire community-wide interest, engagement, and support.
- For long-term success, partnerships need to have one or more “anchor institutions” with dedication to the partnership’s mission and strong commitment to provide ongoing financial support for it.
- Partnerships focused on improving community health should have a designated body with a clearly defined charter that is empowered by the principal partners to set policy and provide strategic leadership for the partnership.
- Partnership leaders should strive to build a clear, mutual understanding of “population health” concepts, definitions, and principles among the partners, participants, and, in so far as possible, the community at-large.
lined in Box 4-2 have a much higher potential for success, up to 80 percent (Prybil et al., 2014, see p. 101, footnote 22). This finding was reinforced by the current study, Prybil said.
After reviewing the findings from the interviews, small group discussions, site visits, and documents, the study team arrived at 11 evidence-based recommendations that reflect the lessons learned from the 12 partnerships studied (see Box 4-3). Prybil noted that none of the 12 partnerships met all of the characteristics and all of the indicators. No partnership does, he added. The question is how substantially they display the characteristics they do possess.
- To enable objective, evidence-based evaluation of a partnership’s progress in achieving its mission and goals and fulfill its accountability to key stakeholders, the partnership’s leadership must specify the community health measures they want to address, the particular objectives and targets they intend to achieve, and the metrics and tools they will use to track and monitor progress.
- All partnerships focused on improving community health should place priority on developing and disseminating “impact statements” that present an evidence-based picture of the effects the partnership’s efforts are having in relation to the direct and indirect costs it is incurring.
- To enhance sustainability, all partnerships focused on community health improvement should develop a deliberate strategy for broadening and diversifying their sources of funding support.
- If they have not already done so, the governing boards of nonprofit hospitals and health systems and the boards of local health departments should establish standing committees with oversight responsibility for their organization’s engagement in examining community health needs, establishing priorities, and developing strategies for addressing them, including multi-sector collaboration focused on community health improvement.
- If they have not already done so, local, state, and federal agencies with responsibilities related to population health improvement and hospital and public health associations should adopt policy positions that promote the development of collaborative partnerships involving hospitals, public health departments, and other stakeholders focused on assessing and improving the health of the communities they serve.
SOURCE: Prybil et al., 2014, pp. 39–44.
The Quad City Health Initiative1 was 1 of the 12 partnerships selected for participation in the study by Prybil and colleagues. Carkner explained that, despite the name, the Quad Cities include five cities on the border of Iowa and Illinois that are home to about 317,000 individuals (Davenport and Bettendorf, Iowa; Moline, East Moline, and Rock Island, Illinois). These five cities on the Mississippi River have a long history of collaboration, she said.
1 The Quad City Health Initiative oversees current projects on health promotion (especially in the context of workplace wellness), mental health (including enhancing public awareness and improving service integration), and tobacco (expanding smoke-free policies and access to smoking cessation services). For more information, visit the initiative’s website http://www.genesishealth.com/qchi (accessed June 18, 2015).
The Quad City Health Initiative started in 1999, when leaders in the community were inspired by work happening in other communities around the country and the Healthy Communities Movement. As told by Carkner, one charismatic leader, an internal medicine physician who was passionate about improving the health of the community, approached his board colleagues at the two local nonprofit health systems, and together they created a community board they named the Quad City Health Initiative. For the first few years, the board operated as a voluntary association of organizations and individuals who were interested in improving the health and quality of life of the Quad City area. Later, the two nonprofit health systems became the founding sponsors of the Quad City Health Initiative, providing seed money to create an office to support the work of the initiative. Carkner joined the initiative in 2001 as its first staff person.
Carkner described several of the characteristics of the partnership that contribute to its success. The vision of the Quad City Health Initiative is to be the community’s leader for collaborative action on health, she said, and the mission of the organization is to create a healthier community. There are three core values: commitment, collaboration, and creativity. The organization thought it was important to include creativity in order to “give ourselves permission” to try approaches that have not been tried before, she explained. Commitment is part of the original ethos for the Quad City Health Initiative and emphasizes that, together, the community can accomplish whatever it sets out to. She added that, 15 years later, the founding health system partners continue to provide most of the operating support for the initiative. The current governance structure is a 25-member community board. Carkner noted that the initiative is not a separately incorporated entity. The work of the initiative is housed at and conducted with in-kind and administrative support from the two local founding health systems.
The foundation for all of the collaborative program or issue-based work of the initiative has been joint community needs and community health assessment planning. This is a bi-state process, which has been going on since 2002. The initiative now has a staff of two: Carkner and a colleague. The power of the Quad City Health Initiative, Carkner said, is a strong multisector, cross-sector network of more than 130 volunteers representing more than 60 organizations around the community. Carkner noted that language from the collective impact literature (e.g., about being a backbone and providing an infrastructure for collaboration) has helped new partners understand what the Quad City Health Initiative is about and why it is so important to creating health in their community. Carkner
said that for many people working at a local level, it is difficult to identify peers around the country with whom to share information and experiences. Being part of this study has been valuable for the Quad City Health Initiative because it has helped to establish such a peer group.
During the open discussion, participants expanded on the topics of securing initial and sustainable funding, engaging partners and stakeholders, setting clear goals for the collaborative, and measuring impact.
One of the lessons partnerships can take from the study by Prybil and colleagues (2014) is how to enhance their chances of success, perhaps using the study findings and recommendations as a checklist, Ramchandani suggested. He asked Prybil and Carkner to elaborate on some of the challenges facing hospital–public health collaborations, and funding was the first topic raised.
Support and sustainability start with having local leadership commit to the partnership with stable support over time, Carkner said. Central to the success of the Quad City Health Initiative has been the anchor leadership of the two nonprofit health systems and their sustained leadership and support over time. They provide the backbone of support that allows for the staff positions and the office to support the work of the board. There has also been support from other community organizations and from individuals, she noted, and the approach to fundraising is still evolving. During this past year, there was, for the first time, 100 percent participation by all of the community board members and/or their organizations in supporting the initiative. The Quad City Health Initiative has also been successful in obtaining grants at the regional and national levels and has just become part of the Centers for Disease Control and Prevention Partnerships to Improve Community Health awardee cohort.
Prybil pointed out that the two health systems partnering with the Quad City Health Initiative are business competitors, but they have agreed to cooperate and support the initiative. The two chief executive officers and their boards believe that the Quad City Health Initiative is good for the whole region. Competitors can be collaborators toward a common goal. Reiterating the findings of the study, Prybil said that partnerships such as the Quad City Health Initiative are helped enormously by having one or more anchor institutions. The anchor institution does not have to be a hospital or health system; it could be a major employer or another party who can make a longer-term commitment of financial
resources to underpin the foundation for the initiative. Both health plans and employers are direct beneficiaries of successful partnerships, Prybil said. Nevertheless, he noted, although many of the partnerships studied have health plan staff in volunteer roles or on the board, few health plans were engaged in helping to support the partnerships financially. Similarly, many employers have allowed their staff to contribute as volunteers, but few major employers have made substantial financial investments in the partnerships.
When approaching foundations, major employers, health plans, and other potential financial partners, it helps significantly to have an impact statement that documents what that partnership is achieving in relation to the investments that are made in it, Prybil said. He noted that among the arguably very successful partnerships studied, relatively few had developed a business case around their impact on the health of the community in relation to their costs. As initiatives try to broaden their base of funding, evidence-based impact statements can be very important in making the case for employers and for health plans to choose to make an investment in the partnerships. Similarly, there must be an ability to link metrics and measures to the elements the partnership is trying to influence. If the partnership cannot demonstrate to current funders that they are making a difference on their priority elements of health, it is going to lose support.
Ramchandani asked what would inspire a health system to invest in a collaboration rather than funding initiatives for their own local population. Carkner responded that in the Quad Cities, both nonprofit health systems interact with the same population base in the community. In addition, the Quad City Health Initiative benefits, and has benefitted from, other community collaborative efforts. Many of the senior leadership who serve on the health initiative board serve on many other community boards together as well, providing opportunities to work together to tackle difficult challenges across sectors. Partners are investing not only in creating health or in supporting the work of one organization but also in creating a broad collaborative infrastructure and capacity in the community. It often takes a charismatic leader to see beyond the borders of the institution. As an example, Prybil mentioned an initiative to make Cheshire County, New Hampshire, the healthiest community in America by 2020. This was the vision of Art Nichols, chief executive officer of Cheshire Medical Center, who saw that even excellent care for individual patients in the hospital was never going to impact the overall health of the community. He rallied the community around this vision, and the challenge now is how to keep track of progress and measure success. Another incentive for joining a collaborative, Ramchandani added, is that no one entity (health care systems, nonprofit organizations, governing agencies)
can do this alone. We are far more effective and efficient when we partner together.
Carkner noted the importance of alignment at a national level with regard to funding. She described the community’s past experience in having received several grants through various organizations that were similar in purpose but were just distinct enough that some confusion occurred at the local level in terms of who was leading on an issue. She stressed that the Quad City Health Initiative welcomes support from organizations outside of the community, but it is most helpful when the partners can design how they work together to respond to community needs.
Bob Griss of the Institute of Social Medicine and Community Health asked how governments at the federal and state levels could leverage their major investments in health care delivery to create incentives for these collaboratives. He recalled the health systems agencies of the 1970s that were focused on collaboration between systems. Prybil responded that recommendation number 11 of the study calls for governments at all levels to adopt policy positions that promote the development of collaborative partnerships focused on the health of the community. The report included examples from Maryland and New York of state-level policy encouraging collaboration between public health and hospitals. He suggested that other public agencies, such as the Centers for Medicare & Medicaid Services, could think about how they might encourage the development of such partnerships. Ramchandani pointed out that the Center for Medicare & Medicaid Innovation has pilot programs in this area.
Prybil added that nearly 70 percent of America’s hospitals are now a part of health systems, and health systems should be encouraged to invest financially in successful partnerships. Community-based partnerships are not the ultimate solution to the country’s health care problems, Prybil said. They are, however, a way to promote a stronger focus on community health and to galvanize communities, citizens, businesses, schools, and other parties to pay more attention collectively to improving the health of the community. Prybil noted that most of the partnerships studied are lightly funded, with little public or private insurer money (with the exceptions of the collaborations involving Kaiser and Maine Health as anchor institutions). Seventy percent of the funding of these partnerships comes from the hospital partners, with another 10 percent from private foundations. Hospitals and health systems are required to produce community benefit, and investing in collaboratives such as the Quad City Health Initiative certainly is a community benefit. Nevertheless, hospitals should not be expected to bear 70 percent of the cost of managing these partnerships, he said, especially when businesses and health plans will ultimately benefit—there needs to be a more balanced funding picture.
David Kindig of the University of Wisconsin School of Medicine and Public Health, asked about the types of partners involved beyond hospitals and public health and what types of organizations are most common as anchor institutions. The Quad City Health Initiative board has ex officio positions for a number of organizations in the community that it determined should always be present at the table, Carkner responded. These include the hospital systems, public health departments, and other community health organizations, as well as permanent positions for the chamber of commerce and representatives from the Rock Island Arsenal, the local metropolitan transit and planning associations, and organizations such as the local United Way and YMCAs. The balance of the 25-member board is structured to include rotating representatives from other sectors of the community, including the education sector, the private/business sector, and other community leaders. For more detail, Carkner referred participants to the discussion of the governance structure of the Quad City Health Initiative in the report (Prybil et al., 2014, pp. 69–71). The structure reflects a health-in-all-policies approach in including cross-sector community stakeholders, she said.
Prybil said that all of the 12 partnerships studied have a broad array of partners who participate at various scales and levels (e.g., anchor institutions that have made a major financial commitment, principal partners who provide support in other ways, other smaller partners, and volunteers). He added that initiative directors, such as Carkner, become very good at managing volunteers and engaging them effectively to contribute to the work of the partnership.
Prybil recommended that partnerships start out with a smaller group of organizations that are committed to the issue, have established trust, and have experience in collaboration. As the organization becomes established, it can then gradually absorb and orient new partners. A collaboration can also be “a revolving door,” as leaders and partners come and go. Partnerships are flexible as a form of organization, he said, but somewhat less durable than corporate entities.
Mary Pittman of the Public Health Institute asked about the next generation of leadership for these partnerships, including youth engagement. Carkner said that local city representatives and local elected officials are now part of the Quad City Health Initiative partnership, which paves the way for engagement of other sectors (e.g., housing and transit) in the health agenda. Becoming institutionalized as an organization helps with recruitment. Carkner observed that the longevity of the partnership for the Quad City Health Initiative has made it easier for her to recruit people at all levels and across sectors to participate in the initiative than when it was new. She noted that the numerous coalitions, teams, and work groups
find opportunities to engage their emerging leaders in the activities of the initiative. The hope is that as their career grows, their participation in the initiative will grow as well. Youth have been engaged through community advocacy such as the tobacco control initiatives.
Ron Bialek of the Public Health Foundation raised the issue of recognizing and addressing “the elephants in the room.” Carkner agreed and said it is very important to be sure that partners and stakeholders have said everything that they believe they need to say. For her initiative, strategic decisions are made by consensus to ensure that all partners believe that their opinions are heard and that their perspectives are represented. This has sometimes meant walking away from opportunities, because there has not been the consensus needed to pursue a specific idea, but the initiative does not want to sacrifice long-term partnerships for short-term objectives.
Marthe Gold from the City College of New York asked what messages could help to bring along partners that are not yet convinced of the need for public health and health systems to collaborate. Carkner said that the Quad City Health Initiative has worked to engage the local private sector and to help partners outside of the health care sector understand their role in creating health and how improved health has an overall positive impact on the region’s economic development and growth. The message they try to convey is that the Initiative’s work is about more than just public health issues in the classic sense; it is about community development and community involvement. Prybil suggested that more people from the hospital and medical community now acknowledge that doing what has always been done is important, but is not sufficient. There is a need to focus on the populations they are serving and reach out to other parties (e.g., schools, public health) that share a common interest in creating a healthier community and a culture of health. Most health care expenditures have been on acute and chronic care, which has not worked very well, he said. Policy-level support for health care–public health collaboration and more interest by health plans and big employers could help to accelerate population health improvement. We need to change our investment portfolios, he said.
The Setting of Goals
Mary Lou Goeke of the United Way of Santa Cruz County, California, asked about commonalities across the 12 partnerships studied in how they set the goals, including engaging their community residents in goal setting. Goals and objectives for the Quad City Health Initiative have been largely based on community health assessments, Carkner said, which have relied very heavily on Behavioral Risk Factor Surveillance System–
type indicators. Over time, team leaders and volunteers have developed process-type measures to assess whether the desired progress toward the goals was being achieved within the timeline set. Prybil added that all 12 of the partnerships studied based their goals on assessment of community needs. What the partnerships have chosen to focus on varies tremendously, he observed, from very specific health care issues to very broad missions. Regardless of their chosen focus, each must define its mission clearly and define measures to assess progress, he said. To help partnerships and other parties who are trying to improve the health of their communities, the report offers guidance, evidence-based recommendations, and concrete examples of metrics and approaches that have worked (see Prybil et al., 2014, pp. 98–99).
Measurement of Impact
Panelists expanded on the topic of measuring impact, including the challenges of sharing data. Carkner said that the Quad City Health Initiative conducted its first bi-state community health assessment in 2002 as its first joint project, with additional assessments in 2007 and 2012, and one is planned for 2015. Joint planning among partners has allowed everyone to work from common information about health status, and that planning process naturally leads to discussion about implementation and strategy selections once the assessment piece is complete, she said. She noted that indicators related to tobacco use have improved over time, but they continue to seek improvement on other indicators that are of common interest across communities, such as those related to physical activity, nutrition, and behavioral health. Still, the process of jointly measuring and assessing has built a foundation for working together.
Impacting outcomes in the right direction is a long-term effort that requires sustained support. Most of the partnerships face challenges in clearly discerning what health outcomes they want to achieve in their communities, Prybil noted. Partnerships that fail to achieve their goals do not clearly establish what they are trying to influence in terms of sustainability and generation of funding, he added. Those who select the most challenging outcomes to influence need to focus very clearly on some intermediate measures that they can achieve, measures for which the evidence shows that if they can “move” them, they will eventually achieve broader outcomes. As an example, he referred participants to the discussion of the Detroit Infant Mortality Reduction Task Force in the report by Prybil et al. (2014). This initiative, sponsored by the four health care systems and many other parties in the Detroit metropolitan area, is focused on reducing infant mortality in inner-city Detroit by identifying intermediate measures to address (Prybil et al., 2014, pp. 81–84).
Paul Mattessich, executive director of Wilder Research, asked about assessing impact in very large collaborative settings, such as Million Hearts, a federally funded program of states working with local community programs, or the Institute of Medicine roundtable and the Primary Care and Public Health Collaborative, which are national organizations with thousands of members collaborating on population health. Prybil said that his study of successful partnerships was focused on the community level; however, many of those characteristics of success could apply to a much broader range of partnerships, perhaps with some adjustments. He reiterated that, at any level, unless there is clarity of the mission and measures (i.e., the specific change to be achieved), it is difficult to ever measure success or impact.
Kindig alerted participants that the roundtable will discuss the issue population health metrics at a workshop in July 2015.2
2 Information on roundtable activities and reports can be found at http://iom.nationalacademies.org/Activities/PublicHealth/PopulationHealthImprovementRT.aspx.
This page intentionally left blank.