How Multi-Sector Collaborations Are Navigating Decisions About Data

It’s becoming increasingly clear that finding solutions to society’s greatest public health challenges will require bringing a broad range of stakeholders and sectors together to unleash the full potential of their data. These data-driven community collaboratives are essential vehicles for helping decision-makers understand and address the social and environmental factors that influence health.

In a previous blog post, we discussed tips for building and sustaining multi-sector partnerships to share data. Once collaborations are in place, a critical and early step is to identify the specific data sources, types, and fields that will enable the desired progress toward shared goals for population health. Incorporating data that builds on the existing goals and priorities of various sectors is a critical element for ensuring the sustainability of a collaboration over the long haul.

But determining how and when community collaborations decide what data should be included, and who should make those decisions, is a challenging process that—if not well navigated—can foster ill-will and lead to setbacks. We asked a group of projects from All In: Data for Community Health, including 10 projects from Data Across Sectors for Health and 15 projects from AcademyHealth’s Community Health Peer Learning (CHP) Program, to share their processes and suggestions for making decisions about data.

We heard broad consensus on one essential ingredient: all discussions about data should start with a clear articulation of why it is needed to support the specific population health objective. This can serve as a foundation for action and provide a shared rationale for doing the hard and often time consuming work required.

Who decides?

During the planning stages of their initiatives, All In project teams engaged a diverse set of decision-makers to provide input on what data and metrics should be included. Examples of stakeholders involved in this deliberative process include:

  • Project managers/team members
  • Data scientists/analysts
  • Community partners (from a variety of different sectors)
  • Project evaluators
  • Expert consultants
  • Legal advisors
  • Funders

Because data decisions can significantly shape future work, engaging a wide range of community leaders, experts, and partners early on in the project can help ensure continued investment and participation in the initiative down the road. It’s also essential to include individuals representing the intervention’s target population. For example, a data sharing program that aims to reduce the rate of unnecessary hospitalization among asthmatic children should include asthma patients and their caregivers, as well as the community-based organizations working on their behalf.

How are decisions made?

Based on existing evidence and national guidelines

When embarking on a data sharing project, many teams first scan the existing literature to find out if other local initiatives or national organizations have already identified useful data sources or metrics. The DASH Environmental Scan is just one example of research that can provide a national context for multi-sector data sharing projects. Many All In projects reviewed the available evidence as well as national guidelines, frameworks, or benchmarks to provide a starting point for thinking about what data sources could help achieve their goals.

“Decisions on what to include in the electronic shared plans of care are informed by national guidelines on care coordination as well as specific needs identified by our community planning group.”

-University of Vermont (CHP)

“The actual measures – metrics in the DASH framework – are pulled from the evidence base, if available, or aligned with other national initiatives or benchmarks.”

-Center for Health Care Services (DASH)

“We are using a theoretical framework for the factors that influence cardiovascular outcomes. In addition, the stakeholder group has recommended additional data that will make the resulting data warehouse more valuable to them from their perspective.”

-Allegheny County Health Department (DASH)

In collaboration with community partners

Community partners from different sectors can provide project leaders and data scientists with useful information that can help inform the decision-making process. Convening partners around the table in discussions about which datasets are the best indicators for understanding a health condition is critical because these other sectors play a vital role in developing and implementing interventions to address the issue.

“HealthInfoNet engages its partners, such as Community Action Agencies, addressing the social determinants of health and health care providers. This allows us to make shared decisions about what data do incorporate into the HIE for access in clinical care and population analytic tools hosted by HealthInfoNet.”

-HealthInfonet (DASH grantee)

“To identify data that would be included in our pilot data build, the Healthy Living Collaborative asked our Principal Investigator to conduct key informant interviews with partner organizations to determine the data they had available and the data/types of analysis that would be of use.”

-Providence Center for Outcomes Research and Education (CHP)

“We have a project manager who is working with our community partners to understand what data is available for analysis, and we have a separate track evaluating availability of data within city agencies. A core project team reviews these memos and a data team evaluates the feasibility of including each data set in the analysis.”

-Baltimore City Health Department (DASH)

Key considerations

When selecting data for your project, here are some important actions to consider:

  • Establish the “why,” clearly articulating how the data will contribute to your shared goals for population health.
  • Ensure that important stakeholders are meaningfully involved in the “what.” Engage them in the decision-making process from the beginning.
  • Learn from others who have been down the same road by researching existing data sources or metrics.
  • Iterate with collaborators to identify the “sweet spot” of data access/use that satisfies their needs but is also feasible to achieve given the context and constraints.

Learn more

As multi-sector collaborations across the country develop data-driven solutions to complex health problems, All In will continue to facilitate the sharing of best practices and lessons learned. Here are three ways you can learn more about current efforts to improve community health through multi-sector data sharing.