Non-linear effect of temperature variation on childhood rotavirus infection: A time series study from Kathmandu, Nepal

Abstract

Introduction: This study aimed to investigate the effects of temperature variability on rotavirus infections among children under 5 years of age in Kathmandu, Nepal. Findings may informinfection control planning, especially in relation to the role of environmental factors in the transmission of rotavirus infection. Methods: Generalized linear Poisson regression equationswith distributed lag non-linearmodelwere fitted to estimate the effect of temperature (maximum,mean and minimum) variation onweekly counts of rotavirus infections among children under 5 years of age living in Kathmandu, Nepal, over the study period (2013 to 2016). Seasonality and long-term effects were adjusted in the model using Fourier terms up to the seventh harmonic and a time function, respectively. We further adjusted the model for the confounding effects of rainfall and relative humidity. Results: During the study period, a total of 733 cases of rotavirus infection were recorded, with amean of 3 cases per week. We detected an inverse non-linear association between rotavirus infection and average weekly mean temperature, with increased risk (RR: 1.52; 95% CI: 1.08–2.15) at the lower quantile (10th percentile) and decreased risk (RR: 0.64; 95% CI: 0.43–0.95) at the higher quantile (75th percentile). Similarly, we detected an increased risk [(RR: 1.93; 95% CI: 1.40–2.65) and (RR: 1.42; 95% CI: 1.04–1.95)] of rotavirus infection for both maximum and minimum temperature at their lower quantile (10th percentile). We estimated that 344 (47.01%) cases of rotavirus diarrhoea among the children under 5 years of age were attributable to minimum temperature. The significant effect of temperature on rotavirus infectionwas not observed beyond lag zero week. Conclusion: An inverse non-linear association was estimated between rotavirus incidence and all three indices of temperature, indicating a higher risk of infection during the cooler times of the year, and suggesting that transmission of rotavirus in Kathmandu, Nepal may be influenced by temperature.