Bringing Researchers and Community Developers Together to Revitalize a Public Housing Project and Improve Health

Abstract

Billions of dollars are invested annually to improve low-income neighborhoods, the health impacts of which are rarely studied prospectively. University of California researchers and Mercy Housing/The Related Companies formed a “Learning Community” with the dual goals of examining the health impacts of a large-scale San Francisco redevelopment project and informing the development team of best public health practices. Early experiences highlight challenges and opportunities, including differences in stakeholders, incentives, and funding, the strengths of local partnerships, and fresh insights from new analytic tools and perspectives. Research suggests interventions that ameliorate upstream causes of poor health would save health care dollars, but policy makers must incentivize collaboration in order to spread a Learning Community model.

Introduction

For decades, the seemingly intractable negative health outcomes associated with urban poverty have been of considerable research interest; however, there has been limited progress in solving this public health problem. The challenge to policymakers goes far beyond addressing the behaviors and health problems of individuals in urban poverty, but extends to entrenched contexts of poverty in which geographic and economic isolation, low-quality housing, crime, and truancy are common and longstanding. In recent years, $50–$80 billion in combined public and private funds have been spent annually in the U.S. to revitalize low-quality housing and impoverished neighborhoods;^1,2 However, the direct effects of these massive investments on health is rarely evaluated.

Here we discuss a collaboration between Mercy Housing California (Mercy) and The Related Companies (Related) – national, non-profit and for-profit developers – and researchers at the University of California, Berkeley and San Francisco. This collaboration was established with the dual goals of examining the health impacts of a large-scale redevelopment project and informing the development team of best public health practices. We believe that by prospectively following Sunnydale residents and examining changes in health status in a community kept intact as developers transform an area of concentrated poverty into an integrated, mixed income neighborhood, we may add substantial insights into the health effects of anti-poverty efforts.

Two worthwhile lessons learned at this early point in the collaboration are: 1) Researchers and developers have different but complementary skills and expertise that can work together to thoughtfully develop and implement a plan of transformation for this neighborhood in collaboration with community stakeholders, and 2) The long period leading up to the start of construction can be used to establish baselines, plan initiatives in collaboration with community members and stakeholders, and seek funding for their implementation.

By forming a “Learning Community” in the setting of a large-scale redevelopment, we hope to provide a model for developers and public health researchers who want to harness the immense resources allocated to community revitalization efforts. Here we briefly review research on the intersection of housing, community development and health, then describe the opportunities for developers and researchers created by this collaboration, the challenges we have encountered thus far, and the potential local- and national-level policy implications of our work.

Housing, Community Development and Health

Epidemiologic research on neighborhood and community characteristics examines how the spatial, built, and natural environments impact the distribution of disease.^{3, 4} Quality housing is an important aspect of neighborhoods and a necessity in our daily lives. Furthermore, where you live can determine access to important financial and social resources.^{5, 6} Areas with poor quality housing often have fewer resources for physical infrastructure (i.e., parks, streetscapes, utility services)^{4, 7} and education,^{8, 9} limited economic opportunities,⁶ and higher levels of racial and economic segregation,^{5, 8} all of which impact health.^10–12 Living in high poverty areas has been associated with lower self-reported health,¹³ higher risk of death,¹⁴ poor emotional and behavioral outcomes for children and adults,^15,16 and higher rates of childhood obesity and other health problems.^10,17

Beginning in the late 1970s, federal and state governments implemented several programs to de-concentrate areas of extreme poverty in an effort to improve the quality of life for low-income urban residents. In what came to be known as “dispersal”, residents were given the opportunity to move to middle income neighborhoods. In the Gautreaux Assisted Housing Program, more than 7,100 families in Chicago public housing took the opportunity to move into desegregated areas elsewhere in the metro area.¹⁸ This program showed positive results when residents moved to less poor communities, but because of stringent eligibility criteria and participant self-selection, the generalizability of the Gautreaux findings are unclear.¹⁹ The first randomized trial of dispersal began with the Moving To Opportunity program in 1992.²⁰ The results for residents moving from high- to low-poverty communities in this more rigorous program were mixed, although interestingly health – and particularly mental health – did significantly improve.²¹ Overall the outcomes spurred more debate in the literature over whether moving away from high-poverty neighborhoods was a net positive for the residents.^22,23

HOPE VI and HOPE SF

In 1992 the federal government created the HOPE VI program (Housing Opportunities for People Everywhere). The goal of HOPE VI was to demolish the most severely distressed public housing developments and replace them with mixed-income housing and enhanced social and physical environments.²⁵ However, this program has been controversial because residents were mandated to move and many of those displaced did not return to the newly renovated housing. These residents lost their social networks,^24,26,27 and experienced few health benefits.²⁸ Furthermore, the HOPE VI Panel Study followed displaced, original residents and found many experienced multiple chronic problems such as high blood pressure, diabetes, arthritis, and asthma.²⁹

Modeled after the HOPE VI program, but with a specific emphasis on maintaining residents in-place within their community, the locally-funded San Francisco initiative, HOPE SF, aims to transform severely distressed public housing into integrated mixed-income neighborhoods (see http://hope-sf.org/about.php). HOPE SF intends to create healthy environments by addressing not only physical development (e.g. unit mix and community spaces) but also human development, by improving access to social services and investing in community capacity. The program’s goal is to reintegrate each site and its resident community into the surrounding neighborhood fabric, ending decades of isolation, and create a new financial model for public housing revitalization applicable nationwide.

Sunnydale

Sunnydale is a 50-acre, 785-unit, San Francisco public housing complex built in the 1940s and home to 1,700 residents. The demographics of Sunnydale illustrate the major challenges facing its revitalization. Of the current residents, more than 900 (53%) are youth or seniors. The median annual household income is $12,750 with a 75% unemployment rate and a median adult education level of 5^th grade. Truancy is a major problem with many children dropping out before middle school. There is heavy drug and alcohol use and three local gangs vie for turf, forcing Sunnydale residents indoors. Many residents deal with untreated mental, physical and behavioral health issues and most have never received professional care. Although nearly all residents are Medicaid eligible, only 16% are enrolled. Thus, first identification of severe problems often occurs through intervention by Youth or Adult Protective Services, too frequently resulting in eviction.

In 2007, Mercy and Related were selected to revitalize Sunnydale. Before the planning began, the developers opened an office on the Sunnydale property with three staff members who provide daily contact and services to the residents. An 18-month engagement process with over 500 local residents produced a community master plan that includes replacement of all existing public housing and the addition of 900 subsidized or market-rate units; a Life Center with a fitness facility and family programming; space for a health clinic, arts program and early childhood education; 6.5 acres of new green space; a farmers market; and 8,000 square feet of retail (banks, restaurants and grocery).

Developer Opportunities through Collaboration with Researchers

Our collaboration impacts the developers’ plans for both human capital and real estate development in multiple ways. First, the researchers are an interdisciplinary team from urban planning, epidemiology, and public health, participating early on to influence how the development proceeds. For example, the Sunnydale master plan includes new parks, community garden and orchards, which were incorporated based on community input and green urban design principles. Now that the initial master planning is completed, the research team will help the developers design programs in these parks to improve individual health through exercise and nutrition, as well as community health and social cohesion.

Second, the researchers can help the developers consider ideas before major investments are made. For example, the master plan includes a significant amount of space for neighborhood programs and retail. Though not entirely in keeping with the developers’ urban design goals, the plan also includes a 10,000 square-foot health clinic. The clinic is a large investment that will require significant financial resources to build and operate. One question we have is whether an on-site clinic is the best way to improve health outcomes for residents or will it duplicate resources already available that can be accessed in a less costly way? The researchers and developers will work together to determine whether and how to include this clinic and its value to the Sunnydale redevelopment.

Third, the research team provides resources and expertise that the developers don’t have. The developers are conducting needs assessments of the existing Sunnydale households with advice from the researchers on assessment design. The researchers provide much needed help in dealing with the complexity of the assessment domains and data analysis. The researchers were able to obtain funding for a summer intern to collate and map initial data collected by the developers. The researchers also provide us their knowledge of the literature and advise on best practices. One researcher’s graduate student surveyed the literature for best practices in developing social cohesion in transitioning communities and provided a set of recommendations that were incorporated into the developers’ work.

Last, our collaboration with researchers has the potential to demonstrate, with rigorously collected and analyzed data, that our real estate and human development plans for Sunnydale will result in improved social and health outcomes for the residents. This will be extremely valuable during interactions with local and federal policy makers and philanthropic organizations that distribute competitive funds for community development and health innovation.

Gains to Researchers through Collaboration with Community Developers

HOPE SF is distinct from prior large-scale redevelopment efforts in its commitment to keeping the community intact and its greater emphasis on developing human capital and improving access to social services. Because neighborhood conditions will improve around the residents of Sunnydale, our early collaboration with the developers presents a unique opportunity to prospectively study the health effects of a neighborhood redevelopment plan. Through this work, we hope to provide an important comparison to the dispersal approach for de-concentrating poverty and begin to answer the question: “Does neighborhood revitalization lead to better health outcomes for the community’s original residents?”

Another benefit to collaboration with Mercy and Related is access to their on-site community-based team. Their knowledge and long-standing presence will facilitate planning outreach with the residents and save valuable time and research funds. Further, by building on their strong ties to local officials, we will be able to collect and integrate information from across levels of analysis that are often examined separately. Combining information obtained from residents with local administrative data (e.g. high school dropout rates, interactions with the justice system, and emergency room visits) will allow a multi-level and cross-sectoral examination of the impact of redevelopment on both individual-level health indicators and community-level costs associated with the use of public resources.

Finally, these collaborations have the potential to introduce new funding streams outside traditional public health or National Institutes of Health sources. For example, partnering with Mercy and Related we were able to apply for large-scale funding through the MacArthur Foundation and are positioned for new funding for research on health and the built environment offered by the United States Department of Housing and Urban Development.

Challenges to Collaboration: Differences in Methods, Culture and Responsibilities

While we see substantial benefits for both researchers and developers through working together, we have also encountered challenges. Cultural differences such as layers of bureaucracy and development deadlines (zoning, conditional use, building permits, etc.) do not always mesh with academic expectations or timeframes. In one instance, San Francisco planned a community needs assessment, but multiple agencies were required to approve any changes to the questionnaire. As a result, the researchers’ were unable to take advantage of the existing survey by adding additional items for analysis.

In another example, data routinely collected by cities and developers is different than data required by researchers. For instance, while property managers are interested in who is on the lease and if the rent gets paid on time, it is possible that the names on the lease do not fully represent the actual residents. However, as researchers we are interested in an exact count and a much more precise picture of who is living in the community. Thus, using existing data on the resident population may not be adequate for the researchers’ analytical needs.

Another challenge is that both parties are responsible to different stakeholders. Developers are responsible to an array of advisory boards, regulators, and funders/investors potentially including a mix of government, foundations, banks and stockholders. In contrast, after meeting university standards for ethics and funding, researchers run relatively little risk of top-down disapproval from their institution. Developers are also more likely to be the visible players in the community, held accountable by residents, local policymakers and community constituencies in a way that behind-the-scenes researchers may not be.

Finally, while both researchers and community developers are responsible for the privacy and comfort of residents, perceptions of what is considered intrusive in the research effort may differ. For example, some research questions could require multiple visits with residents, the collection of personal socio-demographic, social network and health history information, or biological samples (e.g. saliva or blood to assess stress biomarkers). Finding a comfortable compromise that meet’s everyone’s needs will require discussion and negotiation.

Policy Implications

Policymakers should demand that the health impact of billions of dollars in investment no longer go unexamined. We would advocate that alongside each large-scale community development effort, policymakers request that a Learning Community be created to track and evaluate health impacts. These Learning Communities, comprising developers and researchers with input from residents, local government, and other stakeholders, would collect data and complete analyses to generate knowledge on the impact of redevelopment efforts.

This new information has the potential to improve program efficiency and cost-effectiveness by increasing the positive impact of community development on important outcomes related to health, education, and public safety. Data and analyses from such collaborations would be critically valuable to policy makers working to determine where to spend limited funds and to researchers choosing the direction of their inquiries.

In addition, data gathered in the context of Learning Communities could enhance the coordination of government agencies at all levels. At the federal level, large-scale efforts are already underway to break down silos and improve inter-agency communications. For example, with a goal of ensuring that affordable housing, transportation, and energy efficiency are linked, the federal “Sustainable Communities” program has brought together Housing and Urban Development, the Department of Transportation, and the Environmental Protection Agency.³⁰ Similarly, the Departments of Education, Justice, and Health and Human Services are working with the Department of Housing and Urban Development to coordinate investments in neighborhoods of concentrated poverty through the “Choice Neighborhoods Initiative.”³¹ But in order to organize their efforts, federal and local collaborations require empirical research – of the type possible through Learning Communities – to demonstrate strong associations between housing and outcomes relevant to other government agencies (e.g. education, health, and criminal justice). Greater coordination between agencies will, in turn, improve the capacity for further collaboration.

To encourage collaboration, it is critical that funding creates incentives for joint efforts. For example, a new $25 million fund proposed by the United States Department of Housing and Urban Development Office of Policy Development and Research encourages partnering with other agencies to study linkages between health and the built environment.³² Similarly, private initiatives could play a role in supporting developer/researcher collaboration. For example, the Health Impact Group, funded by the Pew Charitable Trusts and Robert Wood Johnson Foundation, has supported health impact assessments (HIAs) of development projects across the country (see www.healthimpactproject.org).

If research finds that improved housing also leads to better school outcomes or reduced criminality, policymakers could propose funding for future work jointly between housing and other government agencies such as justice or education. Similarly, if the work of Learning Communities demonstrates that development is a cost effective way to alleviate health burdens, then local and state health agencies can draw on that evidence to determine whether investing in such community projects is an effective way to “bend the cost curve” of skyrocketing healthcare expenditures. A similar analysis might show that an increase in spending for community revitalization results in net savings – a positive return on investment – but in other areas of government (For a discussion of balancing those costs and benefits at the federal budget level, see Xavier Briggs’ article in this issue).

We believe that increased investment from private capital could also be driven by developer-researcher collaboration. Socially-motivated or mission-related investors – usually foundations, high net-worth individuals, or other institutional investors – might consider community development a worthwhile investment if convincing links to health outcomes were made. For example, a recent randomized study of a new light rail line in Charlotte, NC found that residents who used the train service lost nearly seven pounds over only a few months compared to residents who continued their usual commutes.³³ While bonds sold to fund this light rail line likely appealed to environmentally-motivated investors, the explicit link to an important health outcome suggests that health-motivated investors might be interested in similar projects in the future.

Last, this type of joint research could aid in the development of Social Impact Bonds – or other pay-for-performance programs – as a way to bring new sources of capital to community development. Investors in this vehicle provide money for prevention programs, then share in the savings the government accrues downstream. The key to making this type of investment viable is data that prove cost savings to government.³⁴

Next Steps and Conclusions

Having established a stable Learning Community to explore the health impacts of the Sunnydale redevelopment, our next steps include building on existing community outreach efforts to conduct focus groups with residents (youth and adults) with the goals of developing an on-the-ground perspective of residents’ health related concerns, determining their openness to participating in health surveys, and integrating a focus on health into the community development process. To facilitate this process, we will add experts in community-based participatory research to our Learning Community. We will use the results from our focus groups and community outreach efforts to inform and augment a city-funded, community wide survey and draw on resources at the University of California to analyze results and describe the baseline health characteristics of Sunnydale residents. These results will feed back to development planning and be used to refine existing human capital investments.

Establishing Learning Communities comprising academic researchers and community developers, like the one we have described in this article, can provide new insights into improving both neighborhoods and health. While issues remain to be resolved in this collaboration – including differences in culture, incentives, responsibilities, and adequate funding streams – the opportunities these partnerships present to both parties and to policymakers are substantial. Community developers bring creative financing, local partnerships, and an ability to improve neighborhoods on a large scale. Researchers bring new insights, analytical skills, and provide valuable feedback to improve this work, especially in terms of improving health. Working together, the sum of these two is much greater than the parts.