Data sharing and integration: Why it’s worth it

Key Takeaways

Counties hold a veritable treasure trove of administrative data on the individuals and communities they serve. These data are routinely collected in the course of service delivery and, taken together, form a rich picture of people’s experiences and trajectories. The challenge is that, all too often, each government agency only has access to the data collected through its program provision — just one puzzle piece in that much larger picture.

When counties start to bring these pieces together, important connections are brought to light. For example, behavioral and mental health supports may help a mother maintain employment and avoid interaction with the justice system.

Learn More

• AISP website
• NACo’s upcoming webinar on Sept. 26 at 3 p.m. ET

Access to a high-quality early learning or home-visiting programs may help a family cope with stress or trauma and avoid referral to the child welfare system. More stable housing may increase a child’s attendance and achievement in school. It may have even greater impact if it’s paired with convenient access to preventative medical care. Without the capacity to share and link data across programs, we’ll simply never know.

This realization has motivated many counties to rethink their traditional, siloed approach and pursue data sharing and linkage for interdisciplinary projects and program evaluations.

This work is often slow going, as counties must contend with the quirks of legacy systems, the inertia of business as usual, the risk aversion of legal counsel and the limited capacity of over-burdened staff.

And yet, there are success stories.

In Cuyahoga County, Ohio, data from Case Western’s ChildHood Integrated Longitudinal Data System proved essential in designing, implementing and evaluating a new county program to support mothers experiencing homelessness who also have children in the child welfare system. The unique Pay for Success initiative aims to reduce the number of days that children spend in foster care and is projected to save taxpayers over $4 million dollars.

In Philadelphia, the Data Management Office and partners at the University of Pennsylvania were able to map early childhood risk factors, including issues such as inadequate prenatal care, homelessness, lead exposure, and child maltreatment and compare the geographic distribution of these risks to the geographic distribution of high-quality early childhood programming. Their analysis was used to advocate for a new city-wide beverage tax to support expanded public pre-K. Once the tax passed, the analysis then informed the targeting of over 2,000 new program slots to high-need, low-supply neighborhoods.

So, how can a resource-strapped county with aging technology systems get started?

Here are a few lessons we’ve learned from counties that are leading the way:

• Remember that data sharing is as relational as it is technical. Rather than rushing to procure new systems, we recommend starting with a cross-agency retreat to surface common priorities and research questions.
• Start small. Once you’ve surfaced a cross-agency priority, start with a coalition of the willing and focus on demonstrating value. Public or foundation funding for a single ad-hoc data-sharing project may be leveraged to support basic infrastructure that is adaptable for future uses.
• Tap local universities. When county resources are scarce, academic partners can lend valuable research capacity and expertise.
• Focus on strong and inclusive governance. Clearly defined policies and procedures to support decision-making, routine meeting structures, and well-documented proceedings all help to foster a culture of trust, collaboration, and openness.
• Emphasize transparent communication and stakeholder engagement. The linking and use of sensitive personal data are governed by local, state and federal privacy laws, as well as rigorous technical safeguards. Nevertheless, individuals and communities will have questions about how their information will be used and protected. Counties should lean into opportunities to talk about why data are necessary for social policy improvement and make time to address stakeholders’ concerns, expectations and priorities.
• Set data standards. Focus on linking only data elements that are both relevant to the social problem at hand and of sufficient quality to provide insight. This will require you to define “data quality” and sustain a continued exchange between those who are building the system or data model (generally agency analysts and technologists) and those who work most closely with the data day-to-day (practitioners and program staff).
• Tap into a peer learning community. While every county is unique, no one is well-served reinventing the wheel. We recommend that counties beg, borrow and steal from their NACo colleagues, as well as resources like Actionable Intelligence for Social Policy (AISP).

Check out the AISP website for free resources and information about its Learning Community, and tune in to NACo’s upcoming webinar on Sept. 26 at 3 p.m. ET featuring AISP and its partners from the Broward County Data Collaborative. The county will discuss how it is leveraging AISP’s technical assistance and other resources to formalize a data governance process, engage stakeholders and advance a race equity agenda through community-participatory research.