Developing and Evaluating Mappings of ICD-10 and ICD-10-CM codes to Phecodes

Many studies of Electronic Health Record (EHR) data utilize custom-developed aggregations of billing codes enabling clinical and genetic research, including phenome-wide association studies (PheWAS). One such grouping is the phecode system, originally developed for PheWAS. Phecodes were built upon the International Classification of Diseases, version 9, Clinical Modification (ICD-9-CM). However, many healthcare systems across the world use ICD-10 and ICD-10-CM, and the United States switched from ICD-9-CM to ICD-10-CM in 2015. Here we present our work on developing and validating the mappings for both ICD-10 and ICD-10-CM to phecodes. We first assessed the coverage of both the ICD-10 and ICD-10-CM phecode maps in two large databases: Vanderbilt University Medical Center (VUMC) using ICD-10-CM and the United Kingdom Biobank (UKBB) using ICD-10 codes. We then evaluated the validity of the map for ICD-10-CM by comparing phecode prevalence between ICD-9-CM and ICD-10-CM derived phecodes at VUMC. Approximately 90% of all instances of ICD-10-CM codes and 84% of ICD-10 codes were successfully mapped to phecodes. To demonstrate the utility of the ICD-10-CM map, we further performed a PheWAS using ICD-9-CM and ICD-10-CM maps. This work provides an initial high-coverage map of ICD-10 and ICD-10-CM to phecodes. These codes are publicly available to aid in EHR-based investigation.