Enterprise HHS Made Real: 7 Steps to Data Governance for Meaningful Outcomes (Part Two)

In part one of our blog series, we discussed how health and human services (HHS) programs can achieve better outcomes through a holistic approach. These successful approaches are built on insights from a broad range of data about the agencies’ clients, ultimately aimed at helping individuals and families achieve sustained independence. The value of insights can be applied to several timely challenges. For example, our colleagues recently explored strategies to mitigate the opioid crisis in the blog series “Our Collective Role in Combating the Opioid Abuse Crisis,” and concluded that the epidemic is best answered by a coordinated effort among a diverse group of agencies. This coordination is driven by insights from client data, such as that discovered by Dr. Rahul Gupta and the West Virginia Department of Health and Human Services, as documented by in a recent Ohio Valley Resource article, Painful Lessons: Using Data On Overdose Deaths To Combat Opioid Crisis

In this blog, we look at the ways that proper analysis of shared data from multiple sources and successful data governance can inform successful program delivery through meaningful insights.

Creating a Functional Governance Structure to Share Data

Research shows that the likelihood of success improves when groups of community-based stakeholders regularly share data and information, coordinate services, and collaboratively organize around outcomes. A cross-agency data governance structure is critical to ensure data sharing is performed properly, and agencies today are primed to capitalize on data sharing. In fact, in 2017, Heathdata.gov announced that it now has 3,000 datasets, most publicly available via API access. As the HHS community comes together to share data, data steward workgroups comprised of representatives from participating organizations are pivotal given the standardization and compliance-related considerations in the public sector. To be effective, the data governance and sharing model should address reporting, data quality, and data security and privacy, alleviate community challenges around organizational engagement, support the HHS community in emphasizing the value of data sharing, and enable the flexibility to adapt to new strategies.

How can organizations get started implementing a successful data governance structure?

  • Implement Data Governance Structures, Standard Operating Procedures, Security, and Processes to support sustainable success of data sharing.
  • Implement a staged approach to data sharing that results in either a distributed system access solution where information is accessed directly from the source systems or develop a centralized data warehouse where participant-level data is consolidated for ease of access.
  • Formalize Change Management Structures to support continued engagement with the community through all stages of the future data sharing model development and supporting activities.
  • Influence and adopt federal and state laws that support the active sharing of information to coordinate care, while also safeguarding privacy.
  • Explore potential funding mechanisms to establish a financially sustainable data sharing program.
  • Develop a communication strategy to inform citizens, agencies, and other community-based stakeholders about the importance and benefits of data sharing.

Lessons in Cross-Agency Data Sharing

As we’ve highlighted, state and local government agencies are increasingly tapping into various datasets to help them solve business problems. In our experience, what’s arguably more useful is ‘blending’ relevant datasets together to create richer sets of insights. In working with a state department of elder affairs, we blended survey data with relevant data sets to model nursing home placement, giving the agency a predictive tool to rank survey respondents based on their likelihood to be admitted in the future. With more needs than resources available, our client can be more targeted and objective in making decisions about prioritizing care.

However, simply aggregating the data is not always the answer in getting better outcomes. Often, it’s important to be able to critically look at the data and identify the processes and services requiring focused attention. For a state child welfare agency, we supported key stakeholders and partners including a network of agencies, communities, providers and contractors. To advance its child welfare system, the state recently launched a program organized around the following tasks:

  • Produce an assessment of key program stakeholders in providing an integrated system of care.
  • Monitor and measure the use of resources, the quality and amount of services provided, and child and family outcomes.
  • Develop and maintain an inclusive, interactive, and evidence-supported program of quality improvement that is informed by data analysis.

These tasks are statutorily required to produce key outcomes for children in the state, focused largely on safety, permanency, and well-being. Once the program is fully implemented, the child welfare community will operate in a cycle of accountability.

As pictured above, the cycle relies on five key activity phases focused on continuous quality improvement to ensure the organization is continuously learning and moving toward meaningful outcomes for children and their families. For this cycle to operate effectively, data governance that ensures data is clean and manageable is critical to the insights needed to drive outcomes.

In a third example, we recently worked with a state’s county mental health coalition to unlock what’s possible through coordination and governance. The coalition includes a diverse group of stakeholders representing hospitals, health, housing, and others. It focuses on data-sharing and evidence-based practices to identify gaps and recommend a sustainable continuum of care for this vulnerable population, so they can receive consistent treatment to maintain health and well-being.

Several case studies show there is a greater chance of positive mental health outcomes when these coalition stakeholders regularly share information, coordinate services, and collaboratively organize around outcomes. To that end, success required a strategic effort to coordinate data sharing and focus planning and delivery of services across all the stakeholders. In this initiative, we conducted an assessment of current data sharing capabilities and potential future data sharing opportunities, and options to operationalize systemic data sharing, as well as establish a data sharing vision and action plan to realize that vision.

As we’ve explored today, gathering and analyzing large, holistic data sets with diverse related elements helps to identify trends and patterns that can be used to develop processes and delivery focus to produce better outcomes. And, successful analytics requires an effective governance structure for managing the sharing of data from diverse organizations. In our next and final installment of this series, we will discuss what an enabling enterprise information technology infrastructure might look like and how the direction of federal oversight and funding support can make enterprise HHS a reality.

North Highland subject matter experts Theresa Brandorff, Stephen Easter, and Tina Worley also contributed to the development of this blog.