Integrated Data Systems and Outcomes-Oriented Contracting: A Powerful Combination for Improving OutcomesPart II

In Part I of this post, we described how outcomes-oriented contracting can leverage linked data to inform government spending and support better service delivery and outcomes for individuals, families, and communities. We also described how Integrated Data Systems (IDS) can supply the critical data needed throughout the outcomes-oriented contracting process. In this post, we dive deeper into outcomes-oriented contracting as a high-value use case for IDS-- providing examples from the Empowering Families cohort, outlining key considerations for building IDS, and discussing opportunities to combine these two strategies to enhance your own agency’s use of data.

For an agency administering or delivering social services, IDS becomes particularly important if you are interested in:

  • Measuring how a program or intervention impacts a person’s life trajectory across multiple domains. For example: How do early care and education programs impact a child’s socio-emotional development, health, and education outcomes over time?
  • Learning how a program or intervention impacts different members of the same family. For example: What are the impacts of child care subsidy on both parents’ education and employment, and child development outcomes?
  • Tracking the cumulative impact on an individual or members of a family receiving multiple services  across agencies. For example: How can programs like supportive housing and preventative health services be used in concert to promote better outcomes for child welfare involved families?

If these types of questions are of interest, we recommend you begin by considering the data governance, legal framework, technology, and human capacity that would be required for your agency to take a more cross-agency and outcomes-oriented approach to service delivery.

Data Governance: Data sharing and integration is best supported by a defined governance structure and foundational agreements. Clearly defined policies and procedures  support privacy protections and collaborative decision-making around data and data access. Governance should focus on supporting a culture of trust and collaboration among data owners and other key stakeholders, such as service providers, around the table. For example, when using linked data to understand the achievement of impact metrics or opportunities for continuous improvement in outcomes-oriented contracts, an advisory group made up of service providers and other community partners may play an important role. As part of the Empowering Families cohort, the Department of Human Rights in Iowa worked to enable the use of integrated data to improve and support an outcomes orientation for the statewide Family Development and Self-Sufficiency Program (FaDSS). Their governance structure includes a community-advisory team that is consulted on a project-by-project basis to ensure partners with relevant program and policy expertise have a voice in project design and implementation.

Legal Framework: High-functioning IDS usually rely on two or more foundational agreements to allow for routine data sharing in accordance with all relevant privacy laws. An “enterprise” memorandum of understanding (eMOU) serves as the foundational agreement between the parties that are contributing data to the IDS and the party that is doing the linkage. Once the eMOU is in place, a data sharing agreement (DSA) or business associate agreement (BAA)  between a data partner and the linking partner is often used to outline technical specifications of transfer and storage of identifiable data prior to linking. A data use license (DUL) is used to set forth the terms and conditions under which an analyst, researcher, or evaluator may access anonymized data from the IDS for a specific purpose. This multi-tiered process can substantially reduce the time and energy spent executing “one-off” agreements for data sharing. Colorado has followed this model; so far, six state agencies have signed on to an eMOU allowing for secure sharing of data into a centralized linking hub (Linked Information Network of Colorado - LINC), which produces anonymized datasets for approved end users. LINC enables agencies and their research partners to better understand service users and their outcomes using cross-agency, cross-program, and two-generation analyses.

Technology and Data Model: A high-functioning IDS links only data elements that are both relevant to the social problem or evaluation question at hand and of sufficient quality to provide insight. Sites must therefore first define “data quality” and then establish strategies by which quality may be built into the processes of measurement, collection, record transfer, and analysis. This will require continued exchange between those who are building both the system and data model (generally agency analysts and technologists) and those who work with the data on a day-to-day basis (generally program staff and provider partners). This coordination ensures that the purpose or question guides the technical approach to linkage and analysis, not the other way around. In Massachusetts, the state’s shared data infrastructure has been used to link and leverage high-quality data on short- and long-term education, income, and employment outcomes for the individuals and families most reliant on public assistance. This linked data allows the Commonwealth’s Learn-to-Earn (LTE) initiative to track utilization patterns on specific, shared indicators of interest and better support client outcomes.

Human Capacity: Systems don’t make meaning from data, people do. Data sharing, linkage, and use in decision-making require a sustained investment in agency staffing for this work. Likewise, moving to an outcomes orientation in contracting requires agencies to invest time and capacity in building a shared culture of trust and data use with partners. Providers, in particular, can serve as an important bridge between state agencies and their clients and communities. Agencies and providers should work together to lean into opportunities to talk about why data are necessary for policy and program improvement and smarter resource allocation—and also make time to listen to and address stakeholders’ concerns, expectations, and priorities around ethical data use and data privacy. Some organizations, like the Children’s Services Council of Broward County, FL, have worked to involve clients in the process of formulating research questions and agendas and setting outcomes-oriented program goals. They established the Broward Data Collaborative (BDC), an inter-agency group of state and county stakeholders focused on the sharing and use of community data. BDC aims to provide the capacity to pair quantitative IDS data with information on provider and participant lived experiences to drive funding decisions and promote equitable outcomes for children, families, and communities.

For the Empowering Families cohort of state agencies, and for social services agencies in general, IDS and outcomes-oriented contracting represent a powerful combination for improving outcomes. Together, they enable agencies to link and utilize cross-agency data to monitor and measure utilization and impact across programs, across an individual’s life trajectory, and across families.

For agencies interested in developing IDS to measure and manage toward improved social service outcomes, the Social Impact Partnerships to Pay for Results Act (SIPPRA) is a competitive source of funding that can support these efforts. SIPPRA is a 100 million dollar fund, held by the U.S. Treasury, available to state and local governments that are implementing programs (projects and/or feasibility studies) that produce defined and rigorously-measured outcomes and cost savings. The Round 2 NOFA for feasibility study funding is anticipated to be released in Fall 2019.

If you are interested in learning more about building cross-agency data infrastructure, please contact Della Jenkins at adeliaj@upenn.edu, and for more information about outcomes-oriented contracting or the upcoming SIPPRA Round 2 funding opportunity, please reach out to Oscar Benitez at obenitez@thirdsectorcap.org.

Note: Empowering Families has been funded in part by a Social Innovation Fund (SIF) grant from the Corporation for National and Community Service (CNCS).

About AISP

Actionable Intelligence for Social Policy (AISP) is an initiative housed at the University of Pennsylvania that focuses on promoting secure, cross-sector data sharing for policy analysis and program reform. We help state and local governments and their partners build the technical and human capacity for collaborative and efficient social problem solving. To learn more, visit www.aisp.upenn.edu.

About Third Sector

Third Sector is a nonprofit consultancy transforming the way communities connect people with vital social services. We use public funding, data, and incentives as levers to transform how governments, service providers, and stakeholders work together to deliver services that measurably improve people's lives. To learn more, please visit www.thirdsectorcap.org or send an email to info@thirdsectorcap.org.