Information on the nonmedical aspects that influence health outcomes, often called social determinants of health, is commonly collected at medical appointments. But this information is often recorded as text inside the clinical notes written by physicians, nurses, social staff, and therapists.

Researchers from Regenstrief Institute and Indiana University Fairbanks School of Public Health recently published one in all the primary studies by which natural language processing was applied to social determinants of health. The researchers developed three recent natural language processing algorithms to successfully extract information from text data related to housing challenges, financial stability and employment status from electronic health records.

“Health and well-being are usually not nearly medical care. Mostly, they’re about our behaviors, the environment, our social connections,” said Regenstrief Institute Research Scientist and Fairbanks School of Public Health faculty member Joshua Vest, PhD, who led the study. “Increasingly more healthcare organizations are having to cope with social determinants since it is aspects like financial resources, housing, and employment status that basically drive costs that make people unhealthy. The challenge for health care organizations is effectively measuring and identifying patients with social risks in order that they will intervene.”

“Our work helps advance the sector in each application and methodology. Natural language processing has been applied to quite a few conditions previously, but that is one in all the primary papers that applies it to social determinants of health. We demonstrated that a comparatively simplistic natural language processing approach could effectively measure social determinants as an alternative of using of more sophisticated deep learning and neural network models. These later models are powerful but complex, difficult to implement, and require a variety of expertise, which many health systems do not have.”

We purposely designed a system that would run within the background, read all of the notes and create tags or indicators that claims this patient’s record comprises data suggesting possible concern a couple of social indicator related to health. Our overall goal is to measure social determinants well enough for researchers to develop risk models and for clinicians and healthcare systems to find a way to make use of these aspects – housing challenges, financial security and employment status – in routine practice to assist individuals and to supply a greater understanding of the general characteristics and desires of their patient population.”

Joshua Vest, PhD, Regenstrief Institute Research Scientist and Fairbanks School of Public Health faculty member

Information indicating social needs may be extracted for a lot of varieties of data in an electronic medical record, including information on patient occupation, health insurance coverage, marital status, size of household, address (low versus high crime area) and frequency of address changes.

Previously, Dr. Vest and colleagues, including Regenstrief Institute Vice President for Data and Analytics Shaun Grannis, M.D., created an app they named Uppstroms, Swedish for upstream, and successfully demonstrated that it could use structured data to predict patients in need of a referral to a social service comparable to a nutritionist.

Source:

Journal reference:

Allen, K. S., et al. (2023) Natural language processing-driven state machines to extract social aspects from unstructured clinical documentation. JAMIA Open. doi.org/10.1093/jamiaopen/ooad024.