How we uncover barriers to health


Ethan M. Berke, MD, MPH, and Brian Solow, MD, FAAFP, discuss new ways to use nontraditional sources of data about a person’s health to improve care delivery.

When doctors and other health providers engage patients in conversations and build a trusting relationship, they can uncover social needs that impact health and access to care. Information about social determinants of health (SDOH) — such as lack of housing or food insecurity —when combined with medical records and transaction information, paints a more complete picture about an individual’s, or a population’s, overall well-being.

As the American Medical Association and UnitedHealthcare introduce more than 20 new ICD-10 codes related to SDOH, we can quantify social and medical risks in ways that make health care more sustainable. This allows for the opportunity to:

  • Create new risk and health care models
  • Incentivize and reward health systems for different diagnoses not currently captured or well understood

These new codes provide a more in-depth look at the various factors affecting health. Instead of relying only on traditional information sources, analytics and artificial intelligence (AI) can now be used to organize and measure nontraditional data in ways that can improve health outcomes.

Listening for social risks

There is significant value in listening to what patients and their family members share. The information is incredibly important and helps to fully understand lifestyle and health conditions in order to develop successful care plans.

SDOH data empowers health care systems  to find ways to care for people more fully. And when documented using ICD-10 codes, physicians add structure to this SDOH data, traditionally captured in unstructured clinical notes.

To promote widespread use of the new ICD-10 codes, the American Hospital Association (AHA) now allows any clinician involved in a patient’s care to report codes related to SDOH, not just physicians. Nurses, social workers, case managers and even discharge planners can easily document critical information about a patient’s circumstances from their routine screening processes and interactions with patients.

After all, every patient is different. The care we provide should be, too.

When we collect information about the SDOH, we create the opportunity to uncover issues related to:

  • Economics
  • Food insecurity
  • Housing
  • Mental health
  • Social environments or social isolation

This data can be used to create tools and care plans to address larger determinants impacting overall health and medical outcomes. Without overgeneralizing the data, it can be used to successfully uncover a person’s or population’s medical and social factors and needs — thus leading to many beneficial insights.

Housing as health care

Let’s look at how this data can be put to use by looking at one particular barrier to health: housing insecurity. When a physician codes an overweight patient with hypertension, food instability and a lack of stable housing, the information creates an opportunity to address an individual’s overall needs. In some cases, patients might be facing hurdles that need to be addressed first (such as a permanent home), before steps can be taken to improve their health (such as eating a healthy diet and reducing anxiety and stress). Similar to Maslow’s hierarchy of needs, such instances illustrate how a person might prioritize the need for basic housing and a place to safely store their belongings over whether or not they take their blood pressure medicine or other health care needs.

The AHA says that individuals experiencing housing instability are more likely to have limited access to preventive care, in addition to infectious diseases and chronic health conditions.

The data is so convincing that health care organizations – who only have so many dollars to spend, and historically have split that budget between patient care and administration – are investing in affordable housing:

  • Health plans are looking to manage health more effectively, and in turn reduce costs related to utilization and post-treatment complications. One health plan found that in a program for people in a managed care plan who otherwise would be homeless, emergency room admissions dropped 60% and total cost of care dropped 50% for enrollees.
  • Providers are also getting in the game, with great success. Harbor Place is a former 59-room motel that now provides temporary, emergency housing to patients without stable housing who are discharged from the University of Vermont Medical Center in Burlington Patients receive social and health care services from several partner agencies. UVMC compared hospital use and costs for these patients before and after being at Harbor Place and found:
  • Before their stay, these patients required 95 inpatient admissions at an average cost of $13,000, for a total cost more than $1.2 million.
  • After leaving, the same patients required 30 inpatient admissions at an average cost of $7,000, for a total cost of $220,000.
  • Emergency department use dropped from 161 visits to 94.

Guiding patient care

SDOH and other nontraditional sources can be a challenge for doctors to manage. EMRs are not built to easily streamline all the information they contain. It’s going to take time to understand how it all fits together. Technology and analytics enable us to organize and analyze larger, diverse data sets in useful and meaningful ways.

We continue to discover the most important components that help us develop care plans to best meet the needs of the whole person. As we uncover this data we use it as a guide to take care of a person’s medical issues as well as the other critical parts of their life.

Learn more about next steps in using SDOH data to help improve clinical, financial and operational performance in our recent article “Social determinants: Completing the health data picture.”

About the authors: 

DrEthanBerkeImageEthan M. Berke, MD, MPH
Chief Medical Officer, Population Health
Vice President, Clinical Innovation

Ethan M. Berke, MD, MPH, is the chief medical officer, Population Health Solutions, and vice president of clinical innovation at Optum. In this role, Ethan serves as the clinical lead for provider and health-system-focused solutions that improve the care of patients in the context of their community, and help the health system provide the highest quality care, at the lowest cost, with an exemplary patient and provider experience, no matter what payment system is utilized. He joined Optum after serving as the medical director of clinical design and innovation at Dartmouth-Hitchcock Health System. At D-HHS he was also chief medical officer of ImagineCare, a 24/7 nurse-led, coordinated care model that leverages remote medical sensing and machine learning analytics.


Brian K. Solow, MD, FAAFP

Chief Medical Officer, Optum Analytics

Brian K. Solow, MD, FAAFP, is the chief medical officer of Optum Analytics, which leverages advanced data, analytics techniques and cross-industry health care expertise to serve payer, provider and life sciences clients. He was previously the chief medical officer at OptumRx, coordinating clinical activities related to the development, enhancement and implementation of clinical programs supporting formulary management for OptumRx clients. Before joining Optum, Dr. Solow was an active member of a physician-owned medical group. He holds clinical faculty appointments at both the University of California, San Francisco, and the University of Southern California School of Pharmacy.

Leave a Reply