Climatic factors and rotavirus infections among children under five years old in Bangladesh: time-series analysis

Abstract

BackgroundDiarrhoeal disease is a significant public health challenge in Bangladesh and the association between climate and gastroenteritis is unknown. This study aimed to investigate the effects of climate conditions on rotavirus incidence among children under five years of age using countrywide rotavirus surveillance and climate data to inform public health policy concerning prediction and management of future infections.MethodsTime-series regression analysis was conducted on weekly rotavirus notifications, maximum temperature, relative humidity and rainfall between July 2012 and June 2017. A Poisson regression model with distributed lag linear terms were used to examine the association between climate factors and rotavirus notifications, adjusted for the presence of long-term and seasonal trends using flexible spline functions, over 0-8 week time-lags.ResultsUnadjusted analysis - each 1°C increase in maximum temperature was associated with lower incidence of rotavirus infection (IRR 0.870, 95% CI 0.851-0.890). Adjusted analysis - the direction of the association reverses (IRR 1.046, 95% CI 1.012-1.081) with the number of rotavirus notifications increasing by 4.6% [1.2-8.1] for each 1°C rise in maximum temperature at lag 0 week. Analysis further adjusted for relative humidity (allowing for expected non-linearity) - the association persists (IRR 1.046, 95% CI 1.004-1.089).Conclusions Variations in temperature and rotavirus infection may be confounded by seasonality and long-term trends. Our results are consistent with previous studies suggesting that higher temperatures are associated with increased risk of rotavirus infections. However, the magnitude of this effect may be smaller than previously estimated, particularly when intra-country geographical variations are taken into account. Considering the high prevalence of rotavirus infection in children under five years, understanding the role of climatic variables in disease incidence may inform targeted rotavirus prevention strategies, given the expected rises in annual temperature associated with climate change.