Social Determinants of Health in EMS Records: A Mixed-methods Analysis Using Natural Language Processing and Qualitative Content Analysis

Introduction: Social determinants of health (SDoH) are known to impact the and well-being patients. However, information regarding them is not always collected in healthcare interactions, professionals well-trained or equipped address them. Emergency medical services (EMS) uniquely positioned observe attend SDoH because their presence patients’ environments; however, transmission that may be lost during transitions care. Documentation EMS records helpful identifying addressing insecurities improving outcomes. Our objective this study was determine adult understand how such referenced, appraised, linked other by personnel.Methods: Using for patients 2019 ESO Data Collaborative public-use research dataset using a natural language processing (NLP) algorithm, we identified free-text narratives containing documentation at least one from categories associated with food, housing, employment, insurance, financial, social support insecurities. From NLP corpus, randomly selected 100 each qualitative content analysis grounded theory.Results: Of 5,665,229 analyzed 175,378 (3.1%) were as reference SDoH. References those centered around topics accessibility, mental health, physical substance use. There infrequent explicit references records, but some relationships between could inferred contexts. Appraisals housing mostly negative. Narratives including financial less negatively while insurance neutral related operations procedures.Conclusion: The infrequently documented records. When they included, explicitly often appraised professionals. Given unique position share information, should trained understand, document, practice.