The association of weather factors on dengue transmission in an urbanized tropical environment remains inconclusive. This study aims to assess the impact of weather factors on dengue incidence in Singapore between 2012 and 2019.
Data on weather variables (daily temperature, air quality, rainfall & wind speed) and weekly dengue incidence rates were collected from 1 January 2012 to 25 August 2019. Statistically significant correlated variables identified from cross-correlation analysis using Pearson’s correlation were examined in univariate ARIMA model. A distributed lag non-linear model (DLNM) with Quasi-Poisson model was established to assess any non-linear association between climatic factors and dengue incidence. The Quasi-Poisson model coefficients were evaluated using Quasi Akaike’s Information Criterion (QAIC). To validate the model, the data was split into testing and validation sets, with QAIC, Mean Absolute Error (MAE) & Root Mean Square Error (RMSE) reported.
Pollutant Standards Indices (PSI) greater than 100 was associated to lower weekly dengue incidence at a 5-week and a 7-week lag period. High wind speeds at 5-week lag time was also associated with reduced dengue transmission. Mean and minimum temperatures of 28℃ and 25℃ respectively were associated with reduced risk of weekly dengue transmission across all lags effect. Mean temperatures above 28℃ at a 1-week lag and maximum temperatures above 32℃ at an 11-week lag promoted dengue transmission. Rainfall was not correlated with dengue cases in Singapore.
Based on split-sample model validation, mean temperature was the best predictor of dengue (MAE: 43.15, RMSE: 51.39). Weather factors had varied influence on both pre-epidemic surge periods and non-epidemic periods, but had a stronger correlation with dengue transmission in non-epidemic periods.
Poor air quality and high wind speeds were associated with reduced risk of dengue transmission in an urbanized tropical environment. Only a limited temperature range promotes dengue transmission.