Boston University School of Public Health and Sharecare Mine Data for Social Determinants

A new five-year partnership between the Boston University School of Public Health and Sharecare is launched.

Over the past decade, there has been a data explosion in healthcare, especially for tracking, defining, and addressing the social determinants of health (SDoH).

This has included the following:

• Predictive analytics to prioritize Medicaid programming;
• Risk stratification to enhance patient workflows; and
• Disease state surveillance data analysis to define funding and initiatives for chronic illness and behavioral health.

Some may claim that the boom is attributable to technology’s rise and integration within the industry. Others may see it as a way to isolate the costs of care and address value-based care. At the end of the day, expanded use of the robust data captured across the industry is one way for organizations to better understand, then rein in, costs related to the healthcare, and especially costs associated with the SDoH.

A new five-year partnership between the Boston University School of Public Health (BUSPH) and Sharecare, a digital health company, is set to mine SDoH data. For those requiring a brief primer in Data Mining 101, the method involves the practice of examining large databases in order to generate new information. Data mining is not new; it has been used across the healthcare sector, defining high-risk populations, their chronic diseases, and optimal interventions required. Opportunities abound for managed care providers, health systems, and other types of providers to simply get a better handle on the needs of their populations. In the scope of the SDoH, health system data may opt to use data mining to determine exactly what programming to targets, such as nutrition for mothers and newborns. By reviewing its data, a health system can identify the exact issues that matter most to their target community, if not assorted populations served.

The aforementioned partnership will develop a Community Well-Being Index from captured health data, which will highlight the assorted influences environmental circumstances may have on patients’ overall health and welfare. Large volumes of data already indicate how diverse cultures and individuals who reside in low-income areas experience poor health outcomes as a result. These individuals may also have trouble accessing healthy food, or simply walking to the pharmacy to get their prescriptions, due to other factors, such as high rates of crime and violence.

Boston University’s Biostatistics & Epidemiology Data Analytics Centers then will mine the information so researchers can generate action plans that enhance patient health and wellness outcomes. They plan to use 60 data sets that will span the now-famous five SDoH domains:

• Built environment; 
• Health and healthcare;
• Social and community context;
• Education; and
• Economic stability. 

It is expected that sometime in autumn 2019, Sharecare’s clients (e.g. Walmart, Lockheed Martin, Medicaid) should have access to data visualizations. These graphics will leverage the ability for clients to analyze their populations, allowing them to design unique strategies, programs, and initiatives to improve care.

I’ve become a fan of using solid industry evidence, when available. The Centers for Medicare & Medicaid Services (CMS) Office of Minority Health developed the Mapping Medicare Disparities tool: a public dataset that permits users to conduct individualized population health management analytics. Data exists at the county level for the following:

• Hospital readmissions;
• Mortality rates;
• Emergency department use; and
• Chronic illness prevalence and costs.

The Centers for Disease Control and Prevention (CDC) has extensive data sites to support addressing SDoH:

• Chronic disease indicators
• The Interactive Atlas of Health Disease & Stroke
• National Center for HIV/AIDs, Viral Hepatitis, STD, and TB Prevention Atlas
• The National Environmental Public Health Tracking Network
• Social Vulnerability Index

In addition, the CDC’s Behavioral Risk Factor Surveillance System is the largest continuously conducted health survey system in the world. Using telephone surveys, data is collected on health-related risk behaviors, chronic health conditions, and use of preventive services by U.S. residents across the 50 states, the District of Columbia, and three U.S. territories. We can expect to continue to see ongoing attention to this data as the industry strives to reconcile the triple-edged sword of costs, quality, and value versus volume in addressing the SDoH.