Abstract
Providers currently rely on universal screening to identify health-related social needs (HRSNs). Predicting HRSNs using EHR and community-level data could be more efficient and less resource intensive. Using machine learning models, we evaluated the predictive performance of HRSN status from EHR and community-level social determinants of health (SDOH) data for Medicare and Medicaid beneficiaries participating in the Accountable Health Communities Model. We hypothesized that Medicaid insurance coverage would predict HRSN status. All models significantly outperformed the baseline Medicaid hypothesis. AUCs ranged from 0.59 to 0.68. The top performance (AUC = 0.68 CI 0.66–0.70) was achieved by the “any HRSNs” outcome, which is the most useful for screening prioritization. Community-level SDOH features had lower predictive performance than EHR features. Machine learning models can be used to prioritize patients for screening. However, screening only patients identified by our current model(s) would miss many patients. Future studies are warranted to optimize prediction of HRSNs.
Highlights
Providers currently rely on universal screening to identify health-related social needs (HRSNs)
All models trained with Electronic Health Record (EHR) and Census features significantly outperformed the baseline Medicaid insurance status to determine presence of a HRSN as shown by the 95% confidence intervals (CI) of the Receiver Operating Characteristic (ROC) curves when compared to the baseline Medicaid decision
We found that the addition of readily available social determinants of health (SDOH) data at the community-level did not improve performance over data typically available in the EHR for predicting patient social needs status
Summary
Providers currently rely on universal screening to identify health-related social needs (HRSNs). To achieve more equitable health outcomes at lower c osts[17], healthcare systems should prioritize individual patients for social interventions[18]. Screening patients, those who are low-income and those at highest risk for adverse health outcomes, is an important step in addressing social needs[19]. An alternative approach to universal screening is to utilize patient risk scores or risk prediction models to identify and prioritize patients who are most likely to have HRSNs. Risk scores are already widely used in healthcare settings to predict a range of outcomes from specific disease conditions (e.g., cardiovascular disease) to hospital readmissions, healthcare cost, and ED utilization[34–38]. These studies attempted to predict social service referrals rather than whether the patient reported a social need
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have