Can the American Joint Replacement Registry Utilize Administrative Claims Data to Accurately Classify Revision Total Hip Arthroplasty (THA) Surgical Diagnoses?

• Published in: The Journal of arthroplasty 2023-07, Vol.38 (7), p.S179-S183.e2
• Main Authors: Wilson, Jacob M., Broida, Samuel E., Kremers, Hilal Maradit, Browne, James B., Springer, Bryan D., Berry, Daniel J., Lewallen, David G., Bedard, Nicholas A.
• Format: Article
• Language: English
• Subjects: AJRR; Arthroplasty, Replacement, Hip; Humans; ICD-10; Periprosthetic Fractures - diagnosis; Periprosthetic Fractures - epidemiology; Periprosthetic Fractures - surgery; periprosthetic joint infection; Registries; Reoperation; Retrospective Studies; revision; total hip arthroplasty; United States
• ISSN: 0883-5403; 1532-8406; 1532-8406
• DOI: 10.1016/j.arth.2023.04.021

The American Joint Replacement Registry (AJRR) is a valuable tool for studying revision total hip arthroplasty (rTHA). Currently, International Classification of Diseases-10 (ICD-10) codes are utilized by the AJRR for classifying surgical diagnoses. However, the validity of this methodology is unknown. The purpose of this study was to determine the accuracy of these codes, as used by AJRR, in classifying rTHA diagnoses. We identified 908 rTHAs performed at our institution from 2015 to 2021 using our total joint registry (TJR). Revision diagnoses were obtained from the TJR, which contains prospectively recorded surgical diagnoses collected by trained abstractors, independent from ICD-10 data. The ICD-10 diagnosis codes, as submitted to AJRR, were retrieved for the same procedures. The accuracy of ICD-10 codes for classifying rTHA diagnoses as septic versus aseptic, instability, aseptic loosening, and periprosthetic fracture was assessed using Cohen’s Kappa statistic, sensitivity, and specificity. Concordance between AJRR-submitted data and TJR for classifying rTHA as septic or aseptic was excellent (96.9%; k = 0.87). Agreement for aseptic diagnoses varied from very good for instability (k = 0.76) and loosening (k = 0.67) to moderate for periprosthetic fracture (k = 0.54). Specificity was high (>96%) for all 3 aseptic diagnoses, but sensitivity was lower at 74%, 68%, and 44% for instability, loosening, and periprosthetic fracture, respectively. The AJRR submitted ICD-10 data correctly classifies the infection status of rTHA procedures with outstanding accuracy, but the accuracy for more granular diagnoses was variable. These data demonstrate the potential for diagnosis specific limitations when utilizing ICD-10 administrative claims for registry reporting.