Advancing Community Outcomes Through Equitable Data-Sharing and Partnerships

In the last few years, Philadelphia has seen unacceptable community-level outcomes in terms of gun violence and drug deaths and remains the poorest large city in the US, with a 22% poverty rate. The city has devoted increased resources and capacity to address these complex challenges.

Money alone is not the answer, and no single sector—government, for-profit, or nonprofit—can solve these problems by itself. Another tool is data. City leaders can use multiple types of data thoughtfully to help develop a unified strategy to make progress and deliver services more effectively.

In Philadelphia, decision-making across all levels of government is in the process of becoming more data-driven.¹ City agencies and community-based organizations (CBOs) are developing dashboards^2,3,4,5 and producing data-driven reports to define problems more clearly and measure their own success.^6,7 However, most of these efforts are siloed and, at times, duplicative. There is a critical need for service providers—both government agencies and CBOs—to utilize data collaboratively with greater transparency, equity, and justice. We must make significant advances in the shared use of data, evidence, and evaluation to improve outcomes.

To accomplish this goal, the City of Philadelphia will have to further develop its currently decentralized data systems and focus on building a data culture inside and outside government. Several successful city initiatives have demonstrated what can be done when integrated data is shared and analyzed. But there have not been enough of them. We urge city leaders to take bold action to streamline the data sharing process, make the process more accessible for community-based organizations, and help build community-centered data initiatives.

Specifically, the mayor should consider the following actions:

• Expand Philadelphia’s current capacity to analyze integrated data, identify client-level service gaps, and evaluate the impact of existing programs.

• Signal the importance the mayor puts on these efforts by creating a new Chief Analytics Officer position.

• Strengthen data partnerships across government agencies, the school district, community-based organizations, and higher education institutions to measure collective impact and progress, as well as identify specific areas for improvement.

• Help build community-centered data initiatives that enable community-based organizations to collect, own, receive, and analyze city data.

Improving the city’s performance in data integration and usage requires political will—to overcome the reluctance of some city departments to cooperate with one another; to negotiate data-sharing arrangements with state, local, and community-based agencies; and to further increase spending on systems and training. That political will can come only from the mayor. In this paper, we explore how renewed attention to data-sharing can help leaders inside and outside of local government address Philadelphia’s most pressing challenges.

Data integration is the process of matching administrative service records and other data from city agencies and CBOs to better understand service beneficiaries and government operations—and to address program gaps and opportunities revealed through thoughtful data analysis.

An integrated data system involving various city agencies can ensure that applicants or participants in one program (housing, for example) are automatically enrolled in other programs (like food assistance) for which they are eligible. This results in increased benefits for residents and increased efficiencies for programs.

Philadelphia’s Data Management Office was established over two decades ago, and since then has made incremental progress in developing a data system to support policy initiatives—mostly focused on health and human service programs—by assessing their impact and developing grant proposals across different agencies. As of now, however, the city’s data system and protocols are not well-integrated across the entire government. Further investments at the local and state level are needed to address data barriers. These investments would help make sure that data is collected consistently so that it can be a reliable source for program and policy decisions and is maintained to protect confidential information. Without these increased investments, Philadelphia’s ability to make and measure progress across social, economic, educational and health outcomes will continue to be limited.

In March 2022, the Data Management Office was restructured by executive order as the Office of Integrated Data for Evidence and Action (IDEA). The office’s goal is to bring together, in one place, linked and secure data to answer questions about the design and efficacy of city programs. It has a budget of $1.2 million and a staff of eight—not enough to address the magnitude of the challenge.

Using de-identified city data, the city partnered with the Penn Child Research Center to identify:

1. neighborhoods with low concentrations of quality childcare slots; and
2. neighborhoods with high negative health and socioeconomic indicators for caregivers and children who were most in need of quality pre-K slots. This analysis helped prioritize neighborhoods for pre-K investments.

Data integration is the process of matching administrative service records and other data from city agencies and CBOs to better understand service beneficiaries and government operations—and to address program gaps and opportunities revealed through thoughtful data analysis.

Using de-identified city data, the city partnered with the Penn Child Research Center to identify:

(1) neighborhoods with low concentrations of quality childcare slots; and

(2) neighborhoods with high negative health and socioeconomic indicators for caregivers and children who were most in need of quality pre-K slots. This analysis helped prioritize neighborhoods for pre-K investments.

Another example is more recent. To protect public health, the city decided early in the pandemic to stop water shutoffs to households that did not pay their bills. In the spring of 2022, the shutoff moratorium ended. At the same time, the city pledged that it would strive to protect low-income households from shut-off if they were experiencing financial hardship.

To understand whether existing protections were sufficient and equitable, IDEA was asked by the Mayor’s Office to help the Philadelphia Water Department and Water Revenue Bureau by: 1) identifying low-income households based on individuals’ participation in income-restricted programs; and 2) providing descriptive analyses of the accounts experiencing, at risk of, and protected from shutoff.

The resulting data helped protect nearly 100,000 vulnerable accounts (more than 18,000 of which would have otherwise been shut off for nonpayment during 2022) and informed further changes to the city’s water policy in 2023.

The project required a high level of collaboration from the Managing Director, Finance Director, Mayor’s Policy Office, Health and Human Services, Revenue Department, Water Department, and the Department of Public Health, and was managed by IDEA staff. When questions arose about balancing different policy objectives (e.g., protecting vulnerable families vs. raising sufficient revenue to invest in repairing water infrastructure), the Mayor’s Office helped resolve the issues and ensure that data-sharing agreements moved forward.

These examples demonstrate the kind of work Philadelphia could do more of if it devoted more resources to the effort. For example, with an integrated data system, residents and program customers could be automatically enrolled in benefit programs, eliminating the need for multiple application processes.

One-off projects within a decentralized system can only help the city find solutions that are specific to an individual agency. We need to shift our focus to collective impact, with a proactive approach to improving service delivery and measuring impact.

The county’s Department of Human Services (DHS) Data Warehouse is considered a national model for data integration and analysis at the local level, integrating over 21 categories of data both internal and external to the county.⁹ Unlike data warehouses that are developed for research purposes or “one-shot” analyses, this one supports integrated client service and decision-making.

Using the warehouse’s secure database, DHS caseworkers in Allegheny County can, for instance, identify jailed residents who have experienced homelessness and received mental health treatment within six months prior to their jail stay. This allows caseworkers to more precisely direct services to improve mental health outcomes, reduce homelessness, and prevent recidivism. It allows the county to improve caseworkers’ ability to perform their jobs, enrich the capacity of providers to manage and administer programs and services, and evaluate the effectiveness and quality of policies and operations.

The cost of using and maintaining the Data Warehouse—which includes close to 30 analysts, seven data leads, an analytics manager and technical support—is about $6.5 million annually, far more than what Philadelphia spends on IDEA. This represents less than 1% of DHS’s total budget, and it happened because of leadership, a supportive political climate, and trust.10 Getting non-Allegheny County government entities to share their data required data-sharing agreements with Pittsburgh Public School District and numerous other school districts and education entities within the county, including charter schools. These agreements allow for a unified focus on improving educational and well-being outcomes for school-aged children involved in human services.

DHS can determine the number of students in a school district engaged in active mental health services and compare that to the number of parents associated with child welfare allegations, investigations, or cases. This statistical analysis then enables DHS and the district to create and implement appropriate interventions for students and households and evaluate those strategies in real time. The data can be filtered by race, neighborhood, and other demographic variables to reduce inequality and develop a holistic and unified strategy.

Right now, we have multiple agencies addressing critical issues in siloes. Progress depends on agencies working toward the same goals and measuring the same things. To address both the root causes and the symptoms of a problem—such as housing, mental health, and employment opportunities—coordination across government agencies and the nonprofit sector is essential.

To expand the city’s capacity to successfully integrate and analyze client-level data from an expanded number of municipal and partner agencies, Philadelphia’s next mayor will need to make a significant investment in staff capacity and data analytics infrastructure.

To address concerns around privacy and equity, as well as to ensure that data is collected in a timely, accurate and complete manner, the city will need to establish governance and management systems that are transparent and shared with the public. Investments are needed in leadership and staff capacity, as well as citywide standards and principles. So, too, are executive level mandates.

Leadership and Staff Capacity
While the interagency effort to protect at-risk Philadelphia families from water shut-offs in 2022 succeeded, it demonstrated that the city needs more analytical capacity and resources to be able to conduct similar efforts that are larger in scale or occur simultaneously.

To make this possible, the new administration should create a Chief Analytics Officer position that reports directly to the mayor. The officer would develop and oversee a research, data analysis and evaluation strategy to support the mayor’s policy goals. This includes identifying the research and data analysis questions tied to the administration’s priorities, finding the data sets needed to conduct the analyses, overseeing the development of data-sharing agreements, and directing the work of data analysts. The city should also hire additional analysts to strengthen the staff capacity within IDEA and across agencies delivering critical services.

Until now, data work has largely been done on a department-by-department basis, with ad-hoc involvement from community-based organizations. Arguably, this is why the city is failing to make as much progress as it should.

Citywide Standards & Principles
Besides investing more in leadership and staff capacity, the city should institute a set of interagency policies and procedures that govern data management, including standards for data collection integration, and reporting. Currently, different city departments can have vastly different data collection standards and tools, some of which are rooted in 25-year-old legacy systems that are difficult to use.

Executive-Level Mandates
The next Mayor should mandate the use of basic evaluation activities across all city agencies and contracted partners. This can lead to better understanding of gaps in service delivery and program impacts for specified initiatives and policy priorities.

A robust data-sharing agreement with Pennsylvania’s Department of Labor and Industry would allow the city to integrate residents’ income data, to which it does not currently have access. Similarly, data from the state Department of Human Services would allow the city to quickly identify which residents are enrolled in benefit programs administered by the state.

Establishing such partnerships can present challenges. Linking individual records across city agencies requires data-sharing agreements and strictconfidentiality of records in line with privacy regulations, including the Health Insurance Portability and Accessibility Act (HIPAA) and the Family Educational Rights and Privacy Act (FERPA).

When sharing data with other agencies or organizations, for example, this data would likely need to be de-identified and aggregated to address privacy or confidentiality risks. City leaders will need to work closely with the Law Department to establish new data-sharing agreements or memorandums of understanding with external partners such as the Philadelphia School District, the First Judicial District, and the Philadelphia Housing Authority. This work is labor-intensive, and the city would undoubtedly need more than its current capacity to do it.

Furthermore, local organizations and grassroots groups are generally not “at the table” when it comes to data access and ownership. Representatives of neighborhoods being studied should have access to the data about their communities so they can help craft policies.

The city should view community advisory boards (CABs) and community research review boards (CRRBs) as key components of its approach to making data and research initiatives more equitable on the neighborhood level. CABs advise researchers in recruitment and retention of community participants, as well as interpretation of results;¹² CRRBs help ensure representation and participation of community members in research projects and can help negotiate access to city-owned data and ownership of data generated by research projects for CBOs.

The non-profit, volunteer-led Bronx Community Research Review Board (BxCRRB) in New York City is one example.¹³ It seeks to ensure that clinical research addresses the needs and protections of Bronx residents who participate in the research. Its involvement has resulted in increased trust between the community and researchers.

The BxCRRB reviews applications and advises researchers on how to make their work inclusive and beneficial for all, as well as how to enter the community in a sensitive and respectful way. The board also works to educate residents about research, allowing community members to take a more active role in the projects. The board has the capacity to enter into data-sharing agreements with the New York City government and maintain datasets that can be used and shared with other CBOs.

Similar efforts are underway in Philadelphia, where Nueva Esperanza, Inc. (Esperanza)—a faith-based, multiservice nonprofit organization located in the Hunting Park neighborhood¹⁴—is working to establish a community advisory or research review board.

Once activated, Esperanza’s CRRB would support the organization’s health equity work, which is to be housed in the new Esperanza Institute for Latino Health Equity (EILHE). The CRRB could help the EILHE conduct community-based participatory research and access existing community-level data to better address health inequities affecting Hispanic/Latinos and others in Hunting Park.

The next Mayor of Philadelphia should support Esperanza and other CBOs in their efforts to establish CABs and/or CRRBs. By improving data-sharing between government and community-based organizations, the city can promote more ethical and equitable data practices that are in service of residents’ health and well-being. This kind of cooperation will give CBOs more decision-making power regarding the evaluation and improvement of critical services for their communities

This paper benefited from insights from several practitioners locally and from across the Commonwealth of Pennsylvania. It also had the support of a wonderful team of editors, including Larry Eichel and Elinor Haider of the Pew Charitable Trusts, Caitlin O’Brien and Joe Pyle of the Scattergood Foundation, and Ellen Hwang. The authors are grateful to all those who contributed.