An alternative comprehensive index to quantify the interactive effect of temperature and relative humidity on hand, foot and mouth disease: A two-stage time series study including 143 cities in mainland China

Abstract

BackgroundComprehensive indices have been used to quantify the interactive effect of temperature and humidity on hand, foot and mouth disease (HFMD). The majority of them reflect how weather feels to humans. In this study, we propose an alternative index aiming to reflect the impacts of weather on HFMD and compare its performance with that of previous indices. MethodsWe proposed an index defined as the product of temperature and a weight parameter raised to the rescaled relative humidity, denoted by THIa. We then compared its model fit and heterogeneity with those of previous indices (including the humidex, heat index and temperature) by a multicity two-stage time series analysis. We first built a common distributed lag nonlinear model to estimate the associations between different indices and HFMD for each city separately. We then pooled the city-specific estimates and compared the average model fit (measured by the QAIC) and heterogeneity (measured by I2) among the different indices. ResultsWe included the time series of HFMD and meteorological variables from 143 cities in mainland China from 2009 to 2014. By varying the weight parameter of THIa, the results suggested that 100% relative humidity can amplify the effects of temperature on HFMD 1.6-fold compared to 50% relative humidity. By comparing different candidate indices, THIa performed the best in terms of the average of the model fits (QAIC = 9449.37), followed by humidex, heat index and temperature. In addition, the estimated exposure-response curves between THIa and HFMD were consistent across climate regions with minimum heterogeneity (I2 = 65.90), whereas the others varied across climate regions. ConclusionsThis study proposed an alternative comprehensive index to characterize the interactive effects of temperature and humidity on HFMD. In addition, the results also imply that previous human-based indices might not be sufficient to reflect the complicated associations between weather and HFMD.