Portland, Ore.-based OCHIN provides a cloud-based version of the Epic specifically designed for smaller, independent, and community-based health care organizations. Its national network supports over 44,000 providers at more than 2,200 care delivery sites across 42 states.

OCHIN notes that stressors such as housing instability, food insecurity, chronic stress and financial strain can make it harder to keep appointments, take medications or follow care plans. Over time, these stressors can contribute to higher blood pressure, worsening diabetes and other conditions that increase heart risk. This project seeks to improve risk estimates by combining clinical information with patients’ lived experiences, and to ensure the research translates into better care.


Healthcare Innovation: What are some ways that new AI models and EHR data sets can improve on traditional calculators of heart disease risk?

Schmidt: Studies show that the risk of a cardiovascular disease (CVD) event (e.g., stroke, heart attack) is strongly associated with the social, economic, and environmental factors that shape our lives. Traditional CVD risk calculators don't take many of these factors, such as housing stability, into consideration when calculating someone's risk, but studies show that they are strongly associated with varying levels of risk. And we know that EHR data, such as a chart note by a provider indicating “patient can’t afford meds,” can signal underlying social, environmental, and economic factors. Our hope is that responsible AI can use these clues to better assess risk for CVD.

HCI: Has earlier research demonstrated that factors such as social determinants impact heart disease?

Schmidt: In short, yes. Prior research has indeed demonstrated that socioeconomic and environmental factors can impact cardiovascular disease. This is especially concerning in rural and lower-resourced communities where patients often face additional barriers, such as lack of transportation and long distance to specialists.

HCI: How did the partnership between OCHIN and Mass Chan Medical School come together? Is the work grant-funded?

Schmidt: UMass and OCHIN have been research partners for years, and when UMass Chan Medical School began designing this project, they asked OCHIN to be a partner. This work is funded by the National Institutes of Health National Heart, Lung, and Blood Institute.

HCI: How will OCHIN member clinics use these indicators in their care coordination with patients?

Schmidt: OCHIN members can use these socioeconomic indicators to help determine if their patients are at higher risk for stroke and heart attack, and to direct resources accordingly. For example, if a provider learns that a patient has lost their housing or is unable to afford healthy food, they might recognize these obstacles to self-care and connect the patient with additional community resources.



HCI: Are OCHIN member clinics now regularly capturing social determinant data in ways that they weren’t a few years ago? Alderson: OCHIN members have been at the forefront of screening patients for nonclinical drivers of health outcomes for years, with more than 456,000 screenings conducted within OCHIN Epic so far in 2026 alone. These screenings are important in helping providers connect their patients to much needed resources to improve their health and well-being, such as transportation or housing. 

HCI: Does the fact that OCHIN has a large data warehouse of community health center patients make this type of research easier to do? 

Alderson: The OCHIN-led ADVANCE data warehouse supports research on healthcare and outcomes among patients everywhere — including those who have historically been difficult to reach for research purposes, such as those in rural areas. Having OCHIN network patients included in research ensures that the insights learned drive improvements in patient care in communities across the country, including rural and lower-resourced communities.