Solar radiation and water vapor pressure to forecast chickenpox epidemics.

The clear seasonality of varicella infections in temperate regions suggests the influence of meteorologic conditions. However, there are very few data on this association. The aim of this study was to determine the seasonal pattern of varicella infections on the Mediterranean island of Mallorca (Spain), and its association with meteorologic conditions and schooling. Data on the number of cases of varicella were obtained from the Network of Epidemiologic Surveillance, which is composed of primary care physicians who notify varicella cases on a compulsory basis. From 1995 to 2012, varicella cases were correlated to temperature, humidity, rainfall, water vapor pressure, atmospheric pressure, wind speed, and solar radiation using regression and time-series models. The influence of schooling was also analyzed. A total of 68,379 cases of varicella were notified during the study period. Cases occurred all year round, with a peak incidence in June. Varicella cases increased with the decrease in water vapor pressure and/or the increase of solar radiation, 3 and 4 weeks prior to reporting, respectively. An inverse association was also observed between varicella cases and school holidays. Using these variables, the best fitting autoregressive moving average with exogenous variables (ARMAX) model could predict 95 % of varicella cases. In conclusion, varicella in our region had a clear seasonality, which was mainly determined by solar radiation and water vapor pressure.