Health-related social needs play a crucial role in determining the overall health outcomes of patients. Addressing issues such as housing instability, food insecurity, transportation barriers, and financial strain is essential for improving the well-being of individuals. A recent study conducted by the Regenstrief Institute and the Indiana University Indianapolis Richard M. Fairbanks School of Public Health has delved into the most effective approach to predicting the likely need for health-related social services among patients.
The study focused on identifying patients in the emergency department (E.D.) who may require assistance with health-related social needs in the near future. Researchers compared the use of machine learning algorithms to extract relevant information from electronic health records (EHR) with traditional patient-completed screening surveys. The goal was to determine which method was more accurate in identifying E.D. patients who were likely to need social services within the next 30 days.
The results revealed that a machine learning predictive model, which leveraged various robust EHR data sources such as scheduling information and clinical notes, outperformed the screening questionnaire model in predicting the future need for health-related social services. This finding highlights the potential of utilizing existing data within EHR systems to effectively identify patients in need of assistance.
Dr. Joshua Vest, the senior author of the study and a research scientist at Regenstrief Institute, emphasized the importance of access to information in delivering quality care. He stated that developing tools integrated into EHR systems could streamline the process of identifying and addressing health-related social needs for patients.
Despite the success of the machine learning model, both predictive models demonstrated biases. They were more effective at identifying White, non-Hispanic patients with health-related social needs compared to patients from other racial and ethnic backgrounds. This disparity underscores the need for more inclusive and equitable approaches to addressing social determinants of health.
The emergency department serves as a critical setting for screening patients with health-related social needs, as many vulnerable individuals seek care in this setting. Dr. Olena Mazurenko, the lead author of the study and an associate professor of health policy and management, highlighted the importance of identifying and addressing social needs to prevent patients from repeatedly seeking care in the E.D. due to unmet social challenges.
In addition to improving patient care, collecting information on health-related social needs has become a necessity for healthcare providers due to regulatory requirements from organizations like the Centers for Medicare and Medicaid Services (CMS) and The Joint Commission. These mandates underscore the importance of integrating social determinants of health into clinical practice to enhance overall patient outcomes.
The study, titled “Comparing the performance of screening surveys versus predictive models in identifying patients in need of health-related social need services in the emergency department,” was published in PLOS ONE. The findings highlight the potential of machine learning in identifying patients with health-related social needs and the importance of addressing these needs to improve health outcomes for all individuals.
