Enhanced phenotypes for identifying opioid overdose in emergency department visit electronic health record data

• Published in: JAMIA open 2023-10, Vol.6 (3), p.ooad081
• Main Authors: Ward, Ralph, Obeid, Jihad S, Jennings, Lindsey, Szwast, Elizabeth, Hayes, William Garrett, Pipaliya, Royal, Bailey, Cameron, Faul, Skylar, Polyak, Brianna, Baker, George Hamilton, McCauley, Jenna L, Lenert, Leslie A
• Format: Article
• Language: English
• Subjects: Computational linguistics; Drug overdose; Drugs; Electronic health records; Electronic records; Emergency medical care; Emergency service; Evaluation; Hospitals; Language processing; Machine learning; Medical records; Narcotics; Natural language interfaces; Overdose; Phenotype; Research and Applications; Surveillance; Vital signs
• ISSN: 2574-2531; 2574-2531
• DOI: 10.1093/jamiaopen/ooad081

Abstract Background Accurate identification of opioid overdose (OOD) cases in electronic healthcare record (EHR) data is an important element in surveillance, empirical research, and clinical intervention. We sought to improve existing OOD electronic phenotypes by incorporating new data types beyond diagnostic codes and by applying several statistical and machine learning methods. Materials and Methods We developed an EHR dataset of emergency department visits involving OOD cases or patients considered at risk for an OOD and ascertained true OOD status through manual chart reviews. We developed and validated prediction models using Random Forest, Extreme Gradient Boost, and Elastic Net models that incorporated 717 features involving primary and second diagnoses, chief complaints, medications prescribed, vital signs, laboratory results, and procedural codes. We also developed models limited to single data types. Results A total of 1718 records involving 1485 patients were manually reviewed; 541 (36.4%) patients had one or more OOD. Prediction performance was similar for all models; sensitivity varied from 94% to 97%; and area under the receiver operating characteristic curve (AUC) was 98% for all methods. The primary diagnosis and chief complaint were the most important contributors to AUC performance; primary diagnoses and medication class contributed most to sensitivity; chief complaint, primary diagnosis, and vital signs were most important for specificity. Models limited to decision support data types available in real time demonstrated robust prediction performance. Conclusions Substantial prediction performance improvements were demonstrated for identifying OODs in EHR data. Our e-phenotypes could be applied in surveillance, retrospective empirical applications, or clinical decision support systems. Lay Summary Methods to accurately identify opioid overdose (OOD) cases in electronic healthcare record (EHR) data are important tools for surveillance, empirical research, and clinical interventions needed to help mitigate the opioid crisis. We sought to improve existing OOD EHR phenotypes through 2 innovations: (1) incorporating new data types beyond diagnostic codes and (2) applying several advanced statistical and machine learning methods. We first developed an EHR dataset of 1718 emergency department (ED) visits that was a mixture of 621 OOD cases and 1097 non-OOD visits involving patients considered at high risk for an overdose. These non-overdose pat