The Accuracy of the ICD-10 Diagnostic Codes for Kidney Transplant Rejection and Failure
1Medicine, University of Ottawa, Ottawa Hospital, Ottawa, ON, Canada, 2Medicine, University of Ottawa, Ottawa Hospital Research Institute, Ottawa, ON, Canada, 3Clinical Epidemiology, University of Ottawa, Ottawa Hospital Research Institute, Ottawa, ON, Canada, 4Medicine, University of Montreal, Ottawa, ON, Canada, 5Hematology, University of Ottawa, Ottawa Hospital Research Institute, Ottawa, ON, Canada
Meeting: 2020 American Transplant Congress
Abstract number: B-217
Keywords: Graft failure, Kidney transplantation, N/A, Rejection
Session Information
Session Name: Poster Session B: Quality Assurance Process Improvement & Regulatory Issues
Session Type: Poster Session
Date: Saturday, May 30, 2020
Session Time: 3:15pm-4:00pm
Presentation Time: 3:30pm-4:00pm
Location: Virtual
*Purpose: Clinical research requires that diagnostic codes captured from routinely collected health administrative data accurately identify individuals with a disease. In this study, we examined the diagnostic characteristics of the ICD10 codes for kidney transplant rejection (T86.101) and for kidney transplant rejection or failure (T86.1).
*Methods: We conducted a single institution retrospective study of all adult kidney transplant recipients from 2002 to 2018. Manual chart review was undertaken for each individual to determine rejection at any point in the post-transplant period. Graft loss was captured through the transplant clinic database at our centre, as well as chart review. For each recipient, we determined from our institution’s data warehouse the first date an ICD10 code T86.1 or T86.101 was recorded as a hospital encounter discharge diagnosis. With chart review as the gold standard, we determined the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for the ICD codes T86.1 and T86.101.
*Results: We identified 1,258 kidney transplant recipients during the study period. From chart review, the prevalence of rejection was 15.6% and the prevalence of graft loss (death excluded) or rejection was 20.0%. For the code T86.101, the sensitivity was 71.7% (95% CI 65.2 to 78.3), specificity 96.7% (95.7 to 97.8), PPV 79.0% (72.9 to 85.2), and NPV 95.2% (94.0 to 96.5). For T86.1, the sensitivity was 73.9% (95% CI 68.0 to 79.7), specificity 84.8% (82.6 to 87.0), PPV 50.5% (45.0 to 56.0), and NPV 93.9% (92.4 to 95.5).
*Conclusions: We showed that 79% of individuals receiving a T86.101 code for kidney transplant rejection truly had a rejection episode. Since PPVs vary based on prevalence of disease and our centre’s prevalence of rejection is similar to that found in most transplant centres, the ICD10 T86.101 code is a reliable way of identifying kidney transplant recipients with rejection using administrative health data. Future studies could examine algorithms combining additional diagnostic or procedure codes to improve the ability to correctly identify individuals with kidney transplant rejection using health administrative data.
To cite this abstract in AMA style:
Massicotte-Azarniouch D, Sood M, Fergusson D, Chassé M, Tinmouth A, Knoll G. The Accuracy of the ICD-10 Diagnostic Codes for Kidney Transplant Rejection and Failure [abstract]. Am J Transplant. 2020; 20 (suppl 3). https://atcmeetingabstracts.com/abstract/the-accuracy-of-the-icd-10-diagnostic-codes-for-kidney-transplant-rejection-and-failure/. Accessed May 14, 2025.« Back to 2020 American Transplant Congress