Development and validation of computable social phenotypes for health-related social needs

Objective Measurement of health-related social needs (HRSNs) is complex. We sought to develop and validate computable phenotypes (CPs) using structured electronic health record (EHR) data for food insecurity, housing instability, financial insecurity, transportation barriers, and a composite-type measure of these, using human-defined rule-based and machine learning (ML) classifier approaches.Materials and Methods We collected HRSN surveys as the reference standard and obtained EHR data from 1550 patients in 3 health systems from 2 states. We followed a Delphi-like approach to develop the human-defined rule-based CP. For the ML classifier approach, we trained supervised ML (XGBoost) models using 78 features. Using surveys as the reference standard, we calculated sensitivity, specificity, positive predictive values, and area under the curve (AUC). We compared AUCs using the Delong test and other performance measures using McNemar's test, and checked for differential performance.Results Most patients (63%) reported at least one HRSN on the reference standard survey. Human-defined rule-based CPs exhibited poor performance (AUCs=.52 to .68). ML classifier CPs performed significantly better, but still poor-to-fair (AUCs = .68 to .75). Significant differences for race/ethnicity were found for ML classifier CPs (higher AUCs for White non-Hispanic patients). Important features included number of encounters and Medicaid insurance.Discussion Using a supervised ML classifier approach, HRSN CPs approached thresholds of fair performance, but exhibited differential performance by race/ethnicity.Conclusion CPs may help to identify patients who may benefit from additional social needs screening. Future work should explore the use of area-level features via geospatial data and natural language processing to improve model performance. Health-related social needs (HRSNs), such as food insecurity and housing instability, can impact patients' health. For health systems to address these needs, they need an effective way to measure them. The standard approach to measurement of HRSNs, surveying patients, is challenging due to the time and resources needed to survey each patient. Toward an alternative approach, we sought to determine if patient information from their electronic health records (EHRs) could serve as a "computable phenotype" (a representation of patient characteristics using data that is combined into a set of features and logical expressions) to identify patients with HRSNs. Using 2 different approaches to developing potential computable phenotypes for HRSNs (a human-defined rule-based approach and a machine learning approach), we found that the computable phenotypes in the current study were poor-to-fair at accurately identifying patients with HRSNs and that the phenotype performance was poorer for patients who were non-White and/or Hispanic. Future work could seek to improve the computable phenotypes by including additional data, such as clinical notes.