LAUSR

Kawasaki disease (KD) is an acute febrile vasculitis of childhood of unknown cause.1 As the coronavirus disease 2019 (COVID-19) pandemic continues, many countries have implemented nonpharmaceutical interventions (NPIs) to mitigate the spread of the virus. In Korea, NPIs such as mandatory mask wearing, school closure, and testing and isolation of symptomatic individuals were implemented in February 2020, with good compliance.2 On-site school reopening was postponed until May 20, 2020, and schools were reopened gradually until June 6, maintaining strict policies such as physical distancing and restrictions on private education and group activities.2 This nationwide intervention provided a unique opportunity for us to investigate the collateral effect of NPIs on the incidence of other diseases. We aimed to investigate whether NPIs were associated with a change in KD incidence in Korea, the country with the second-highest KD incidence rate worldwide. In this retrospective, ecologic study, data on all KD cases among children 0 to 19 years of age from January 2010 to September 2020 were collected using the National Health Insurance Service database. The NPI period was defined as February through September 2020. The KD incidence rate in the NPI period was compared with the annual mean incidence rate during the same months in the pre-NPI period and with the incidence rate predicted by the autoregressive integrated moving average model. KD was defined as a case where the International Classification of Diseases–10 diagnostic code for KD (M30.3) and the prescription code for intravenous immunoglobulin (IVIg) were claimed together. IVIg-resistant KD was defined as recrudescent or persistent fever 24 to 48 hours after the first IVIg infusion, requiring an IVIg prescription nonconsecutively for at least 3 days. These data are licensed for this analysis only, and the institutional review board approved the study. A total of 53 424 cases of KD were identified; among these, 83% occurred in children <5 years of age, and 17% were resistant to IVIg. In the pre-NPI period, the annual mean incidence was 48.1 per 100 000 persons and the monthly mean incidence was 4.1/100 000 (Figure [A]). After NPI implementation, KD incidence declined sharply; from April onward, it plateaued (Figure [B]). KD incidence is typically seasonal, with a minor peak in the late spring and summer and a larger one in the winter; however, this seasonal pattern was absent in the NPI period. When considering the months from February to September, the incidence in the NPI period was 18.8/100 000, corresponding to 60% of the mean incidence (31.5/100 000) in the pre-NPI period (P=0.085) and 58% of the predicted incidence (32.2/100 000 [95% CI, 46.6–77.9%]). The incidence in children between 0 and 4 years of age, the group with the highest incidence rate, was 80.0/100 000, significantly lower than the