The Inaccuracy of ICD-10 Coding in Revision Total Hip Arthroplasty and Its Implication on Revision Data

The International Statistical Classification of Diseases, 10th Revision (ICD-10), was adopted by the United States on October 1, 2015 and expanded coding from 3800 codes with the International Statistical Classification of Diseases, Ninth Revision, procedure code system (ICD-9-PCS) to 73,000. The increase in number of codes was designed to create more accurate representations of procedures like revision total hip arthroplasties (rTHAs). However, many worry that the increased complexity leads to more inaccurate coding. The purpose of this study is to determine the accuracy of ICD-10-PCS coding for rTHA and discuss the implications on registry data.

The rTHA databases at 2 large, academic medical centers were retrospectively reviewed for all rTHAs between October 1, 2015 and July 3, 2019. The laterality and specific revised components were recorded and compared with the ICD-10-PCS codes used for each procedure. The accuracy of ICD-10-PCS codes relative to the surgical record was determined using coding guidelines published by the American Joint Replacement Registry (AJRR).

Overall, 895 cases were reviewed. Replacement coding was 22% accurate (195 of 895). For removal and replacement coding, accuracy dropped to 17% (152 of 895). All procedures had at least 1 rTHA trigger code that would signify correctly to AJRR that an rTHA occurred.

In this study, the percent of correctly coded rTHA was low. All rTHA procedures had at least 1 AJRR trigger code; therefore, an rTHA would have been appropriately captured by AJRR. But these inaccuracies should make one pause when using ICD-10-PCS procedural data to try to evaluate specific rTHA details from administrative claims databases and ward against expanding ICD-10-PCS as a means to collect implant survival and registry data.