Top 3 Data Challenges to Addressing the Social Determinants of Health

Data collection and sharing

A lack of standardization in data collection and sharing has significantly impacted organizations’ ability to address the SDOH. In a 2019 report from the Department of Health and Human Services (HHS), stakeholders noted that various methods of gathering SDOH information can hinder initiatives and projects centered around the non-clinical elements that impact health.

“The growth in the use of SDOH data has yielded a diverse array of definitions and terms for the various categories and metrics of SDOH data,” HHS wrote.

“The differences in how providers collect housing data, for example, can include different definitions, metadata, and measurement. Moreover, a wide variety of screening tools are currently used in clinical settings to capture data about the social determinants of health.”

To overcome these issues, HHS said that stakeholders should develop a centralized repository of open-source assessment tools and social screening platforms. This will help identify best practices and build consensus among industry leaders.

Within individual organizations, providers may be able to use advanced technologies to identify key terms in the EHR that are associated with the SDOH.

A 2017 study from Massachusetts General Hospital (MGH) showed that natural language processing (NLP) techniques could help extract meaningful socioeconomic data in the EHR and predict psychosocial risk.

Because SDOH information is usually collected within the EHR in an unstructured format – as free-text clinical notes or in the form of zip codes, missed appointment records, or payment patterns – tools like NLP could help providers obtain the SDOH information needed to refer patients to critical social services.

As providers in the clinical care space work to integrate socioeconomic information with medical data, privacy and security are also top of mind for industry stakeholders. A recent report from the National Academies of Sciences, Engineering, and Medicine stressed the need for data sharing standards and increased interoperability around SDOH.

“The interoperability landscape is already working to build a trust framework and set of incentives for sharing data between health and social care systems,” the report said.

“As such, going forward the healthcare and social sectors will need to agree upon a shared trust framework for data exchange, sharing, and use. Part of building a transparent trust framework will be the need to have meaningful informed consent for individuals.”

Limited research scope, limited data

Closing gaps in healthcare and reducing disparities among different racial and ethnic groups is a big part of adequately addressing individuals’ social determinants. However, failing to collect information about minority and underserved populations can prevent providers from delivering comprehensive care to minority patients.

“National standards of care are largely based on Caucasian patient data,” Amy M. Andrade, MS, assistant vice president of research at Meharry Medical College and founding director of Meharry Medical College’s Data Science Institute, told HealthITAnalytics.com in an interview.

“However, this information isn’t applicable to all patients, and we can’t rely on that information alone to care for our diverse populations, which may respond differently to various treatments.”

Broadening the scope of medical research to include more diverse patient cohorts will help eliminate care gaps and uncover the causes of medical disparities. At Meharry Medical College, researchers and staff are working to leverage socioeconomic data to improve health outcomes for diverse patients.

Using a database that tracks health and environmental data at the neighborhood level, including air quality metrics, crime statistics, access to affordable housing, and availability of grocery stores, researchers are uncovering new insights related to diabetes, obesity, and hypertension.

These conditions disproportionately affect minority and underserved populations, and having access to this data could help organizations determine disease trends among different groups of patients.

“The Institute’s database will help our researchers identify and understand patterns in social and environmental triggers for these chronic diseases, and ultimately inform recommendations and solutions to address them,” explained Andrade.

In addition to accessing more data on minority and underrepresented populations, the industry should make a focused effort to fund research projects that explore ways to reduce disparities in healthcare.

In a recent blog published in Health Affairs, Lynne Garner, Nancy C. Yedlin, and Jennifer Salopek of the Donaghue Foundation discussed the need for research institutions to recognize potential implicit biases in grantee selection processes.

“To reduce potential bias in all forms, including our affinity for organizations or investigators we’ve worked with before, and to eliminate implicit gender or racial bias, we will continue to use a blinded process for reviewing Letters of Intent, the initial phase of our application review process,” the authors wrote.

Finding the time and tools to address SDOH data

Although providers recognize the important role social determinants play in overall health, many still lack the time and resources necessary to adequately address their patients’ social needs.

A 2018 survey from the American Academy of Family Physicians (AAFP) showed that 80 percent of family physicians feel they don’t have the time to discuss social determinants during routine consults. Sixty-four percent said they don’t believe they have the staff or resources to do anything about social risk factors even if they can identify them.

“One of the major obstacles providers face is connecting with the individual and extracting the contextual information that they can’t get from a healthcare claim,” Sanket Shah, Clinical Assistant Professor at the University of Illinois at Chicago Master of Science in Health Informatics and Health Information Management, told HealthITAnalytics.com.

“The straightforward questions that providers are accustomed to asking – such as when a patient’s last visit was or when their last A1C reading was – are important. But it's the non-clinical things that come up that provide a holistic picture of the individual.”

To encourage more conversation around SDOH, and to better address individuals’ social needs, providers could ask questions that go beyond just physical health.

“Providers may want to consider expanding some of the techniques and approaches that they’re accustomed to when engaging with their patient population. They could start asking the questions that lead to pertinent social determinants information, and change their style a little bit,” Shah said.

Organizations could also use real-time data to better evaluate patients’ SDOH. The North Carolina Department of Health and Human Services (NC DHHS) recently deployed a patient intake platform that sends real-time alerts to providers and care coordinators about patients’ individual social needs.

As part of the project, North Carolina also deployed the first statewide coordinated network with a shared technology connecting healthcare and social service providers, making it easier to connect patients with resources to address their social needs.

“To advance the health and well-being of North Carolinians, we need to build a coordinated, whole person-centered system that addresses both medical and non-medical drivers of health,” DHHS Secretary Mandy K. Cohen, MD, said at the time.

“This project will make it easier for doctors and other healthcare providers to ask patients about their non-medical health needs, which are a critical component of their overall health.”

Recognizing and understanding individuals’ SDOH will lead to more comprehensive, holistic healthcare, but stakeholders across the care continuum will need to address several data-related challenges before this vision can become a reality. By assessing the data they have, as well as the data they need to collect, organizations can more fully integrate social determinants information with clinical care delivery.