A systematic review of validated methods for identifying Kawasaki disease using administrative or claims data

Purpose: To identify and assess algorithms used to identify Kawasaki syndrome/Kawasaki disease in administrative and claims databases.

Methods: We searched the MEDLINE database from 1991 to September 2012 using controlled vocabulary and key terms related to Kawasaki disease. We also searched the reference lists of included studies. Two investigators independently assessed the full text of studies against pre-determined inclusion criteria. Two reviewers independently extracted data regarding participant and algorithm characteristics.

Results: Our searches identified 177 citations of which 22 met our inclusion criteria. All studies used algorithms including International Classification of Diseases, Ninth Revision (ICD-9) code 446.1 either alone, or with evidence of intravenous immunoglobulin (IVIG) administration, or with ICD-10 code M30.3. Six studies confirmed diagnoses by medical chart review. Three of these six studies reported validation statistics, with positive predictive values of 74%, 84%, and 86%, respectively.

Conclusions: All studies that reported algorithms used either the ICD-9 code 446.1 either alone, with evidence of IVIG administration or with ICD-10 code M30.3. The ICD-9 code 446.1 alone produced positive predictive values of 74%, 84%, and 86% in separate studies in Georgia and California. The sensitivity of these codes to detect Kawasaki disease is unknown, as no sampling of medical records for missed true cases of Kawasaki disease was done. Further research would be helpful to determine whether the relatively high positive predictive values found in southern California are seen elsewhere and also to evaluate the performance of other codes to identify cases of Kawasaki disease and the sensitivity of the narrow algorithms that have been used to date.