Pay-for-Success Financing and Population Health

https://wrenchinthegears.com/wp-content/uploads/2020/12/Pay-for-Success-Financing-and-Population-Health-Financing-Population-Health-Improvement-NCBI-Bookshelf.pdf Financing Population Health Improvement: Workshop Summary. Show details Roundtable on Population Health Improvement; Board on Population Health and Public Health Practice; Institute of Medicine. Washington (DC): National Academies Press (US); 2015 Jan 6. ContentsHardcopy Version at National Academies Press Search term < PrevNext > 5Pay-for-Success Financing and Population Health Continuing the theme of finding creative ways to finance population health and align incentives to focus on performance, the workshop's third panel discussed a relatively new approach that is being put into action to fund population health: using social impact bonds to provide capital in a scheme known as pay-for-success financing. Social impact bonds allow philanthropic funders and private investors to pool capital for social programs, with the loans repaid by the government only if the funded initiative achieves agreed-on results. Megan Golden, a fellow at New York University Wagner School's Innovation Labs and the Institute for Child Success, provided an overview of the role that pay-for-success financing can play in population health. Robert Dugger, founder and managing partner of Hanover Provident Capital, LLC, and Rick Brush, founder and chief executive officer of Collective Health, then gave their perspectives on this new approach to financing population health and discussed examples of how it has been used so far. The presentations were followed by an open discussion period, moderated by Andrew Webber, a member of the workshop planning committee and chief executive officer of the Maine Health Management Coalition. Go to: OVERVIEW The idea of paying for success in the population health field is not new—David Kindig introduced the idea in 1997 in the book Purchasing Population Health—but according to Megan Golden two trends and one problem came together to produce the field's current focus on this new financing mechanism. The first trend was the rise of impact investing, the potential of which was quantified in a 2010 report from J.P. Morgan and the Rockefeller Foundation that estimated that some $400 billion to $1 trillion would be available to achieve results in social programs (O'Donohoe et al., 2010). The second trend was the increasing interest among foundations, government, and service providers in measuring the impact of spending. The problem, Golden said, was the same one that this workshop had been discussing, which is that government is spending a great deal of money reacting to problems after they have happened and has little money left over to spend on preventing those problems from arising in the first place (Golden and Waters, 2014). Golden explained how pay-for-success financing works. Investors put up money to implement cost-effective programs on a large scale, and then the government contracts to pay back the investors with a small premium after the programs have demonstrated that they achieve some predetermined outcomes. She further explained that there may be an intermediary that manages the project and contracts with the investors, the government, and service providers to implement the interventions. There is also an impartial evaluator who determines whether the outcomes are achieved by using accepted research evaluation methodology and comparing those outcomes to some kind of a matched comparison or control group. “The result is a win-win-win-win situation,” Golden said. Communities and individuals benefit from more effective services yielding better results. Nonprofit groups benefit by having access to upfront funding that enables them to scale programs. Government wins with the development of more cost-effective services and better results. And investors win by having the ability to make a positive impact on a community while achieving modest returns on their investment. The first example of pay-for-success financing in the United States was launched in 2012 to fund a New York City program aimed at reducing recidivism among 16- to 18-year-olds leaving city jails. For this project, social impact bonds were used to fund implementation of a proven cognitive behavioral intervention for the 3,000 or so young people passing through the city's jail each year. Goldman Sachs invested $9.6 million with a 6-year loan, and the degree of reduction in recidivism will determine what the city will pay back to the investor. For example, if recidivism falls by 20 percent or more, the city will make the maximum payment of $11.7 million. The breakeven point is a 10 percent reduction in recidivism, while the city would only pay $4.8 million if the reduction in recidivism only reaches 8.5 percent. While these payment terms provide a reasonable return if the program is successful, the risk on the downside was too high, so a foundation provided a $7.2 million loan guarantee so that the most Goldman Sachs will lose is $2.4 million. Currently, nine pay-for-success projects have been completed worldwide—four in the United States, four in the United Kingdom, and one in Australia. No final results are available for any of these projects. Nonetheless, Golden said that population health seems to be a good candidate for this type of financing because there is a strong evidence base for at least some interventions and their benefits exceed their costs, producing harvestable cost savings. Recently, she completed a feasibility study for a South Carolina–based organization called the Institute for Child Success that wants to scale the Nurse–Family Partnership program, which aims to improve outcomes for young children in that state (Institute for Child Success, 2012). Because the study supported the feasibility of such a program, South Carolina is now working to put together a pay-for-success scheme to finance it. South Carolina is particularly interested in improving the lives of its youngest citizens, Golden said, because the state ranks 45th in overall child well-being. Home-visiting programs have demonstrated improved outcomes for children in a number of studies, and South Carolina, with federal funding, has implemented several of these visitation programs but not at a scale that would produce a big impact. According to Golden's assessment, 11,500 very poor women in the state could benefit each year, but even after an expansion in 2012 the state's program could only serve 568 new families annually. By examining the available data and working with providers on the ground, Golden developed an expansion strategy that could realistically come to serve 2,750 new families phased in over 3 years at a cost of $21 million and with a projected savings of $52 million. About two-thirds of those savings would come from Medicaid funds. “That's about $5 in savings for every $2 invested,” Golden said, “which makes for a pretty attractive investment opportunity.” Putting together a pay-for-success arrangement requires a specific metric on which to condition payments, so Golden focused specifically on health outcomes even though other benefits were expected in areas such as child abuse, education, and criminal justice. In the home visiting feasibility study, the metric Golden and colleagues chose to use was the reduction in preterm birth rates—a selection that was based an examination of the available state Medicaid data on adverse birth outcomes and other health outcomes for the first 2 years of life. Based on the research, Golden and colleagues plan to phase in the program over 3 years, with a focus on one to two health outcomes chosen by the state, which will only pay if there are improvements in the participant group, in comparison to others. At the time of the workshop, the pay-for-success transaction to finance this program was in the process of being put together, and the deal is expected to involve a mix of commercial and philanthropic capital. Golden said that while a reduction in preterm births will not by itself pay for the program in the short term, there are enough data available on the longer-term savings that come from reducing preterm births to indicate that this investment will be worthwhile for South Carolina (IOM, 2006). In closing, Golden said that there are challenges to using this financing mechanism but that there are also many reasons for enthusiasm. “The reason I'm excited about it,” she said, “is that it gets multiple sectors together to focus on and be accountable for outcomes, and it's a mechanism to operationalize that shift of funding from remediation to a place farther upstream.” Go to: SOCIAL IMPACT INVESTMENT AND POPULATION HEALTH What is the most important product of our society? “A ready-for-life 18-year-old,” said Robert Dugger in stating the central tenet underlying pay-for-success financing for early childhood health and development projects. He noted that two organizations that he helped establish, the Human Capital Economic Opportunity Working Group and ReadyNation, are particularly interested in supporting these kinds of projects. The Human Capital Economic Opportunity Working Group comprises about 400 researchers across all fields of health, psychology, behavior, and economics who are concerned about producing ready-for-life 18-year-olds, while ReadyNation focuses on building business leadership support for this type of work. Both groups operate from a conclusively demonstrated premise that investments in children at the earliest months and years of life have higher returns than investments made later in life and that these returns continue to accumulate over very long periods of time. “The breakthrough that has occurred over the last few years,” Dugger said, “is the awareness that you don't have to wait 15 years until you get drops in teen pregnancy, drug use, and adolescent crime, to get higher rates of return.” The South Carolina program that Golden discussed is one example of how this idea is being translated into practice. Although there is a significant body of research showing that early interventions can produce satisfactory rates of return on investment, Dugger questioned the time frame over which those returns are actually realized. For example, a study of a program to create strong nurse–family relationships for high-risk pregnant mothers in Virginia calculated that this program had a 1-year return of 26 percent based on a reduction in the need for neonatal care (Greene, 2008). But are the savings realized immediately? “You know almost immediately that neonatal intensive care use is going to drop, but we still have a neonatal intensive care unit that is being amortized and nurses and doctors that are under multiple-year contracts,” Dugger said. “So even though we reduce intensive care use immediately, we don't really get the full savings immediately.” This is one aspect of pay-for-success financing that has not been completely worked out yet, he said. Turning to the specifics of how pay-for-success projects are developed and implemented, Dugger explained that they start with a feasibility study that underlies the entire transaction for a specific intervention in a specific population. “A pay-for-success project is like a mortgage financing,” he said. “You don't do a mortgage financing on the average house; you do it for a specific house. And just as you do for a mortgage, there has to be an appraisal, not for the average house, but for that specific house in that neighborhood.” Conducting a feasibility study requires data detailing past performance of the specific intervention in a community similar to the one that will receive the intervention, and the lack of these data is the first problem encountered when trying to put together a pay-for-success project. “The data demands for social impact financing are very high, and the later in a child's life that the intervention is occurring, the higher the data challenges are,” Dugger said. Assuming that the data are available and that the feasibility study provides a good case for funding an intervention, the next step in the process involves many contracts. In most cases, there are contracts with the state, contracts with a third party evaluator, and contracts with the organizations that will realize the savings.”There are important contractual issues here that are only now beginning to be understood,” Dugger said. “This is an important reason why there are so few social impact finance transactions.” Dugger spoke about his own experience, helping to finance a project in Utah that aims to expand quality pre-kindergarten programs for 600 children. The United Way of Salt Lake City provided the regional vision; Voices for Utah Children conducted the feasibility study; and Goldman Sachs, the United Way, and a group of philanthropists, including Dugger and the Pritzker Family Foundation, provided $20 million in financing. Financial engineering was done by Voices for Utah Children, Goldman Sachs, Imperium Capital, and the Pritzker Children's Initiative, while the United Way of Salt Lake City, acting as the intermediary, managed the funds. The project was launched in June 2013. Go to: IMPACT INVESTING FOR BETTER HEALTH AND FINANCIAL OUTCOMES One of the challenges that come with investing in health, Rick Brush said, “is that what really matters to health is a very complex, intertwined constellation of factors that are embedded in our communities, our social life, our economic fabric, and our culture. So how do we invest when we know, for example, that education matters more to health than health care, when we know that relationships, social connections with our friends and families, and our support networks, and our community matter so much to health that they outweigh risk factors like obesity? How do we realign the health care financing system to begin to address these larger factors?” One way to think about this challenge is to view it through the lens of an investor with $2.7 trillion—the amount of the nation's annual health care expenditures—to invest. The first step would be to use data to identify what really matters to health and what initiatives would create the greatest impact, after which one would place strategic bets based on estimated return on investment. Ideally, there would be a portfolio of short- and long-term investments, each with specific metrics for improvement and each evaluated using sound methodologies that can attribute outcomes to these investments. Brush noted that the Memphis program that Gunderson and Cutts described earlier in the workshop was supported by Cigna, a major insurer in the Memphis area, and he suggested that companies in the insurance industry should be seen as potential financial partners in pay-for-success programs. Other potential partners include self-insured employers, which actually account for a large percentage of the health care dollars being spent; foundations that are looking to make philanthropic contributions and returns on investment; high-net-worth individuals; and commercial investors. In bringing all of these different sources of investment together, it will be important to use a progressive reinvestment strategy that can take short-term cost savings out of the health care system and dedicate a portion to paying back investors through shared agreements among all of the payers that are benefitting from these savings. “So as we reduce Medicaid costs or we reduce costs for a self-insured employer,” Brush said, “they keep some of the savings, pay back a portion of savings to investors that provided the upfront capital, and, ideally, reinvest a portion of savings in expanded or longer-term health improvement programs.” As an example of a potential application of health impact financing, Brush described an effort to reduce costs associated with childhood asthma in Fresno, California, that is being supported by the California Endowment. Previous studies have shown that a home-based intervention can return $5.30 to $14 for each dollar invested (Nurmagambetov et al., 2011) and decrease asthma-related emergency room visits by 68 percent and hospitalizations by 84.8 percent (Woods et al., 2012). Brush said that while this demonstration project will aim to benefit 200 children, the estimated population of children in Fresno that would benefit from its widespread implementation would be around 16,000 children, given that about 20 percent of children in Fresno have been diagnosed with asthma, with the actual percentage likely to be even higher. The demonstration targets specific neighborhoods where the incidence rate is as high as 40 percent. Fortunately, Brush said, there are easily implemented home-based strategies to improve air quality that have demonstrated dramatic reductions in asthma-related emergencies for children. In the current project, the evaluators are using insurance claims data to measure the actual cost savings to the state's Medi-Cal program, which have been estimated to be between $1,000 and $5,000 per year per child enrolled in the program. The project has engaged Clinica Sierra Vista, a network of federally qualified health clinics, as its clinical partner and the Central California Asthma Collaborative as its community partner. The program pays for a range of services, including home visits by community health workers, repairs to fix leaky roofs and to remove and prevent mold at an average of cost of $850 per home, HEPA (high-efficiency particulate air) vacuum cleaners, and integrated pest management, and it replaces asthma-triggering home cleaners with less caustic cleaning products. Brush closed his comments by describing the work of a new nonprofit organization called HICCup, the Health Initiative Coordinating Council, which was founded by angel investor Esther Dyson. HICCup is launching a national competition in 2014 called The Way to Wellville, in which five communities with populations under 100,000 will compete to win the HICCup Prize for the greatest cost-effective improvement in five measures of health over 5 years. HICCup will support the five Wellville communities with a network of partners and investment models that produce better health with financial returns. Go to: DISCUSSION Session moderator Andrew Webber opened the discussion by asking the panel how quickly the tipping point might come—if it ever will—for pay-for-success financing to gain critical mass and become an important and widely used mechanism for funding projects to improve the public's health. He noted that the nation has been waiting for at least 15 years for the transition to a pay-for-performance medical system to reach a tipping point, and that has yet to happen. Brush answered that “it is still early in the application of pay-for-success and impact investing in health care,” and he noted that there are still many complexities that have to be addressed, such as whether benefits will aggregate to a single payer or a dozen payers in a community, how to attribute an outcome to an intervention, and how to capture and access sources of repayment. He noted that the complexities are particularly challenging when dealing with governments, which is why he believes that self-insured employers may be an important test bed for this mechanism. In contrast, Dugger said he has no doubt that pay-for-success financing has reached the tipping point and that it will take off as an important mechanism for health improvement efforts because of the increasing availability of data that can be used to complete the feasibility studies needed to set up this type of financing scheme. “The Affordable Care Act is making many things clear that were not clear before, such as the discrepancy between the contributors to health and what we actually pay for,” he said. He also said that managed care companies are already using interventions to reduce their own costs and pocket the savings and that he considers such efforts to be examples of internally funded pay-for-success financings. He acknowledged that the contractual infrastructure needed to facilitate these transactions is still in its infancy, and he predicted that it will take at least a decade to assemble the legal architecture that will enable pay-for-success to become widespread. Golden added that there is a huge pipeline of projects in the works, with many states and counties having issued requests for information or requests for proposals to procure pay-for-success contracts. “I think there's a wave coming,” she said. “The first several deals will focus on programs that have a rock solid evidence base.” She cautioned, though, that the field will have to expand its focus from those programs with a substantial supportive evidence base to the kind of work that Brush is doing with demonstration projects or to projects that examine multiple outcomes. Golden also said the White House, the Office of Management and Budget, and the U.S. Department of Treasury have all expressed great interest in this financing mechanism and have instructed all federal agencies to try to take advantage of it. Given the need for data to demonstrate the feasibility of projects, David Kindig asked, is there some way to use public–private partnerships to broaden the financing base for the research that needs to be done to develop the evidence base for the next generation of initiatives? Both Brush and Dugger thought that would be possible, particularly for specific feasibility studies. “I think we can put together packages of investors, philanthropists, and state and federal funding for feasibility studies,” Dugger said. In fact, he said he believes that there will be a proliferation of such arrangements. “I can see many organizations coming into existence to solve different parts of this problem.” Dugger said he got involved in social investing because of the realization that other countries, such as those in East Asia and Northern Europe, invest far more in human capital than the United States does and that we are not producing enough “life-ready 18-year-olds” to sustain our gross domestic product. As a result, he formed a partnership and developed the Invest in Kids Working Groups to promote investments in human capital. Marthe Gold of the City College of New York asked about the necessity of randomized clinical trials to convince businesses and investment firms to enter pay-for-success contracts. Both Dugger and Golden said that not all businesses or existing pay-for-success programs require trials before deciding to invest. Dugger went on to say that some of the investors are compelled enough by moral aspirations and thus do not rely solely on data analysis. Copyright 2015 by the National Academy of Sciences. All rights reserved.