Background: Seasonality is a characteristic of some respiratory viruses. The aim of our study was to evaluate the seasonality and the potential effects of different meteorological factors on the detection rate of the non-SARS Corona Virus detection by PCR. Methods: We performed a retrospective analysis of 12763 respiratory tract sample results (288 positive and 12475 negative) for non-SARS, non-MERS Corona viruses (NL63, 229E, OC43, HKU1). The effect of seven single weather factors on the Corona virus detection rate was fitted in a logistic regression model with and without adjusting for other weather factors. Results: Corona virus infections followed a seasonal pattern peaking from December to March and plunging from July to September. The seasonal effect was less pronounced in immunosuppressed patients compared to immunocompetent. Different automatic variable selection processes agreed to select the predictors temperature, relative humidity, cloud cover and precipitation as remaining predictors in the multivariable logistic regression model including all weather factors, with low ambient temperature, low relative humidity, high cloud cover and high precipitation being linked to increased Corona virus detection rates. Conclusions: Corona virus infections followed a seasonal pattern, which was more pronounced in immunocompetent patients compared to immunosuppressed. Several meteorological factors were associated with the Corona virus detection rate. However, when mutually adjusting for all weather factors, only temperature, relative humidity, precipitation and cloud cover contributed independently to predicting the Corona virus detection rate.
medrxiv Subject Collection: Infectious Diseases