Effect of Climate on COVID-19 Incidence: A Cross-Sectional Study in Japan

Abstract

Background: Effect of meteorological factors such as air temperature, humidity, and sunlight exposure on transmission dynamics of novel coronavirus disease 2019 (COVID-19) remains controversial. We investigated the associationof these factors on COVID-19 incidence in Japan.Methods: We analyzed data on reverse transcription polymerase chain reaction confirmed COVID-19 cases foreach prefecture (total=47) in Japan and incidence rate was defined as the number of all reported cumulative casesfrom January 15 to March 17, 2020. Independent variables of each prefecture included three climatic variables(mean values of air temperature, relative humidity, and sunlight exposure), population elderly ratio, and the numberof inbound travelers from China during February 2020. Multivariable-adjusted Poisson regression model wasconstructed to estimate COVID-19 incidence rate ratio (IRR) of independent variables.Results: There was a total of 702 cases during the study period in Japan (population=125, 900,000). Mean±standarddeviation values of meteorological variables were 7.12°C±2.91°C for air temperature, 67.49%±7.63% for relative humidity,and 46.77±12.55% for sunlight exposure. Poisson regression model adjusted for climate variables showedsignificant association between the incidence and three climatic variables: IRR for air temperature 0.854 (95% confidenceinterval [CI], 0.804–0.907; P<0.0001), relative humidity 0.904 (95% CI, 0.864–0.945; P<0.0001), and sunlightexposure 0.973 (95% CI, 0.951–0.997; P=0.026).Conclusion: Higher values of air temperature, relative humidity and sunlight exposure were associated with lowerincidence of COVID-19. Public health interventions against COVID-19 epidemic in a country should be developedby considering these meteorological factors