Day 2 of Datapalooza: More Providers Seeking Social Determinants Data | David Raths | Healthcare Blogs Skip to content Skip to navigation

Day 2 of Datapalooza: More Providers Seeking Social Determinants Data

April 28, 2017
| Reprints
Datapalooza also focused on the rise of machine learning, predictive analytics, the creation of learning health systems
Click To View Gallery

Healthcare Informatics made social determinants of health data one of its top 10 tech trends for 2017. The presentations at the 8th annual Health Datapalooza in Washington, D.C., this week confirmed that sharing data across sectors is becoming more important in a value-based payment world. There were many references to getting a 360-degree view of patients’ lives to better understand the context of the care they are receiving.

In an April 28 luncheon conversation, Mike Leavitt, former Secretary of Health & Human Services under George W. Bush as well as a former governor of Utah, said that social determinants data is at the the heart of the transition from fee-for-service to value-based care.

“We are going to see states contract not just for managed Medicaid programs, but also other human service programs that states conduct now. That is where experimentation on social determinants will happen,” Leavitt said. “Medicaid data will become among the most important tools we have as it becomes more directed toward managed care and social determinants.”

Dana Barnes, a strategy consultant on the Kaiser Permanente national delivery system strategy team, described how KP is piloting in Southern California a Total Health Action Tool that pulls medical data from the EHR and nonmedical measures from public data sources and maps them in a data visualization platform.

The tool includes 100 indicators and is helping Kaiser employees make decisions about marketing, planning new retail clinics, and understanding member transportation needs. Although the tool is currently only available to KP employees, Barnes said she hoped that the use cases that develop from it would help KP develop an open data strategy for collaboration with community groups.

The Datapalooza conference also featured a panel session on the benefits and challenges of connecting health providers and housing organizations.  One project in Cleveland has created an integrated an public health database that brings together medical records held by health systems on patients with COPD and conditions such as asthma with housing structure issues identified by the city such as mold, lead inspections and violations. That can lead to data-driven recruitment for home repairs.

Last year, Healthcare Informatics interviewed another of that panel’s presenters, Martin Love, CEO of the North Coast Health Improvement and Information Network (NCHIIN).  A nonprofit HIE in California, NCHIIN has built an interface between the Humboldt County Department of Health and Human Services Social Service Homeless Management Information System and North Coast’s HIE to share data from electronic health records and public and private records. Social services case managers will receive alerts about clients’ health center/hospital visits for follow-up care coordination. The project is part of the Community Health Peer Learning Program, a national peer learning collaborative managed by AcademyHealth through a $2.2 million award from the Office of the National Coordinator for Health Information Technology.

Sharing social determinant data was just one of many issues addressed during the two-day meeting. Datapalooza also focused on the rise of machine learning, predictive analytics, the creation of learning health systems, and more. In just one example, Terri Steinberg, M.D., chief health information officer and vice president of population health informatics at Christiana Care Health System in Delaware, described the Carelink CareNow platform it has created for care managers, which enhances care coordination and includes a predictive analytics risk score for patients. “We believed care managers shouldn’t look for things to act upon but be notified by systems,” she said. “This is not an EHR,” Steinberg added. “It is a completely different world from Epic and Cerner.” It is a highly customized suite of services, and it pings the care managers when there is something to deal with in real time.  Among the data sources feeding the platform are ADT feeds from the Delaware Health Information Network, the statewide HIE. The care manager may get a notice and contact the emergency department about a patient when the patient is still in the waiting room.

In addition, the conference featured several announcements of grants and challenges.

Dr. Siddika Mithani, president of the Public Health Agency of Canada, and Donald Wright, Acting Assistant Secretary for Health at the U.S. Department of Health and Human Services, announced a Healthy Behavior Data Challenge, which seeks to identify and evaluate new data sources and methods to enhance public health surveillance.

They said they are looking for innovators to propose creative new types of data and data sources that can be used to measure indicators of physical activity (e.g. daily number of steps), sleep (e.g. number of times awake per night), sedentary behavior (e.g. average number of hours per day spent sedentary), or nutrition (e.g. servings of fruits and vegetables consumed per day).

Selected submissions will be invited to implement their concept. The results will be compared to existing research, analysis, and/or surveillance outputs from the Public Health Agency of Canada (for the Canadian stream) or the Centers for Disease Control and Prevention (for the American stream). The most promising solutions will each receive a financial award and an opportunity to explore how the concept could be integrated into public health surveillance systems.