Estimation of the incidence of Kawasaki disease in Taiwan. A comparison of two data sources: nationwide hospital survey and national health insurance claims

Abstract

Background: Kawasaki disease (KD), first described by Dr. Tomisaku Kawasaki in 1967, was found for the first time in Taiwan in 1976. It continued to occur in increased numbers. For the study of incidence rates and epidemiological features of KD, we conducted five nationwide hospital surveys (NHS) in 1987, 1992, 1994, 2001 and 2008, respectively. We estimated also the annual incidence rates of KD during 1996-2007, based on the National Health Insurance (NHI) database, which had been implemented since 1995, covering 98% of the population in Taiwan.

Methods: The annual incidence rates of KD during the twelve years, from1996 to 2007, estimated by the NHS and the NHI claims were compared, analyzed and discussed.

Results: During 1996-2007, a total of 9,938 cases of KD were reported by the Departments of Pediatrics of all hospitals surveyed, and a total of 11,849 cases of KD were claimed in the NHI database. The annual number of cases and incidence rates of KD based on NHI claims constantly surpassed those by the NHS. The ratio of the two incidence rates varied from 1.10 to 1.33. They were well correlated (r = 0.902, p < 0.001) with a linear equation, NHI = 16.07 + 0.93*NHS. The changes in annual incidence rate by the NHI were mean 1.149, p = 0.07, 95% CI -0.082 - 2.382, and those by the NHS were mean 1.562, p <0.001, CI 0.656 - 2.468.

Conclusion: The annual incidence rates of KD can be estimated by the NHI claims and by the classic NHS. The values estimated by the NHI claims constantly outnumbered those by the NHS. Some pitfalls involved in the NHI claims are discussed.

Keywords: Kawasaki disease; National Health Insurance claims; Taiwan; incidence of KD; nationwide questionnaire hospital survey.