Developing Physical Activity Referral Criteria For Patient Data Extraction From An Electronic Health Record System

PURPOSE: With greater integration of the physical activity (PA) vital sign in electronic health record (EHR) systems, providers can efficiently identify inactive individuals and make referrals to existing PA resources. However, not all inactive patients should be considered eligible. The purpose of this study was to describe the generation of International Classification of Diseases, 10th revision (ICD-10) codes to use as inclusion and exclusion criteria during EHR data extraction in determining whether patients should have received a PA referral. METHODS: Exercise is Medicine Greenville® (EIMG®) is an established PA referral program in the Prisma Health system and was used for this process. The inclusion criteria for referral to EIMG® are based on five primary diagnoses (e.g., physical inactivity, obesity, type II diabetes, hypertension, and dyslipidemia). Each diagnosis was mapped to all relevant ICD-10 codes. Determining the exclusion criteria began by identifying diagnoses that would disqualify an individual from participation in the program based on established EIMG® policies. Additional exclusion criteria included contraindications to PA by the American College of Sports Medicine’s Guidelines for Exercise Testing and Prescription and the Physical Activity Readiness Questionnaire for Everyone. Additionally, seven practicing physicians, familiar with EIMG®, were asked to identify exclusionary diagnoses they considered when making a PA referral. All the diagnoses were combined and duplicates were eliminated. The list was then refined through a collaborative effort of individuals with varying backgrounds (e.g., lifestyle medicine experts, health care professionals). Each diagnosis was then mapped to all relevant ICD-10 codes, then discussed again by the research team, leading to a final list of exclusion criteria. RESULTS: The five primary inclusion criteria diagnoses were mapped to 35 individual ICD-10 codes. Thirty-six exclusion criteria diagnoses were mapped to 109 truncated ICD-10 codes. CONCLUSIONS: The inclusion-exclusion criteria may be used to inform others to identify patient eligibility in referral to PA programs during EHR data extraction. The list of codes can be utilized world-wide as more work is done in the identification and treatment of physical inactivity. Research reported in this poster was supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health under award number R56HL157218