Three Ways Communities Are Sharing Data to Improve Health

As communities form multi-sector collaborations work to share and use data that supports their ability to better understand and address the social determinants of health, there is a growing recognition of the need to expand partnerships and build capacity beyond the healthcare sector.

A primary aim of Data Across Sectors for Health (DASH) is to better understand local initiatives that are pioneering new ways to integrate data systems from multiple sectors with the ultimate aim of improving community health.

Through its Environmental Scan, the DASH National Program Office reviewed peer-reviewed and grey literature, interviewed experts and thought leaders, and conducted surveys to characterize and document existing and emerging multi-sector collaborations. The process of cataloging a vibrant set of data-sharing initiatives from across the United States revealed three main community health aims:

1. care coordination;

2. community health needs assessment (CHNA), as well as related planning, implementation and monitoring efforts; and

3. research and policy change.

This blog shares key insights related to each aim, and includes examples of initiatives that are integrating multi-sector data for these purposes.

1. Linking Individual Data to Provide Coordinated Care

For some people, especially those with complex health and social needs, coordination of medical and community services can be the difference between life and death. As these high-need individuals are often among the most costly, many communities are linking social services and clinical services data to provide more comprehensive care that proactively identifies patients’ needs in order to deliver the right services at the right time. While these efforts focus primarily on individual and family needs, they can have community-wide health and cost implications. For this reason, Accountable Care Organizations (ACOs) and social service agencies often join forces to improve care coordination for the most vulnerable populations.

Altair Accountable Care for People with Disabilities is a DASH grantee that, with funding and leadership from Lutheran Social Services of Minnesota, is integrating behavioral health providers into the state health information exchange (HIE) to improve care coordination for patients with disabilities. By incorporating disability-competent providers into the ACO and HIE, and creating an interface to help behavioral health providers manage patient care, Altair aims to provide holistic social, behavioral, and clinical care to a high-need and high-cost population.

2. Using Aggregated Geographic Data for Needs Assessment, Planning, and Monitoring

Linking different aggregate data sources to reveal information about a specific geographic area or sub-population can help create a more comprehensive understanding of how a number of different social, environmental, and health factors contribute to the health of a community. Measuring health at a more granular level allows community stakeholders to identify disparities, target and plan more effective interventions, and monitor improvement over time. Such projects can be undertaken by health systems involved in community health needs assessments, public health departments, and other community partnerships.

The NYC Department of Health and Mental Hygiene, a DASH grantee, is linking new and existing data sources from multiple sectors at the neighborhood tabulation area (NTA) level. The NTA level (approximately 30,000 residents on average) is more granular than already-existing community district level (more than 100,000 residents on average). Preliminary analysis of the linked data has already revealed pockets of disparity in certain health outcomes that were not apparent when data was analyzed on the community district level. NTA-level data will allow for more focused health interventions, more efficient use of city resources to address disparities, and more sophisticated analysis of the effects of proposed policies. A subset of NTA-level data will also eventually be available to the public through an online portal to help empower community action around health issues.

3. Combining Data for Research, Policy, and Advocacy

Data can also be integrated to analyze and research issues related to health equity, social determinants of health, and vulnerable populations. The findings can be applied to community engagement efforts, advocacy agendas, and policy decisions. Academic institutions are often key partners in these initiatives, which sometimes produce data infrastructure that can be re-purposed or applied to address related questions.

A DASH grantee, the Allegheny Data Sharing Alliance for Health is building a connected data warehouse that aggregates multi-sector data on cardiovascular risk factors. Once data is collected, it will be exported to a modeling platform, FRED, designed by the University of Pittsburgh’s Graduate School of Public Health, and house a geographically-accurate risk model for cardiovascular disease in Allegheny County. This infrastructure will facilitate the investigation of potential interventions and policies aimed at reducing cardiovascular disease and enable assessment of their relative impact. Having built this capacity, the Allegheny County Health Department is actively exploring possible collaborative efforts to address other major health conditions of concern to the community.

But Wait, There’s More!

Of course, not all initiatives fit neatly into one of three aims and, as collaborations and data systems mature, data-sharing activities can support multiple simultaneous objectives. For example, the Vanderbilt University Department of Health Policy aims to help coordinate care as well as impact research and policy as part of its project supported through the Community Health Peer Learning Program. Aggregating data from an array of sectors (e.g., public health, social services, criminal justice) the team is working to create an infant mortality risk model for the surrounding area, which has a disproportionally high infant mortality rate. This risk model will be used to identify specific at-risk mothers and infants and refer them to culturally appropriate interventions, introducing a care coordination element.

There are currently other projects working primarily on one community health aim, but many may eventually expand to support others. For example, Altair ACO’s primary use case involves care coordination, but when data is aggregated in the future, they will be able to evaluate care and inform policies and initiatives.

All three models presented have the potential to transform the health of communities. Greater attention to and investment in data systems, and the increasing proliferation and maturity of collaborations, should enable both policy and systems-level changes in the health landscape, and improve the lives of millions.