An accurate early warning system to predict impending epidemics enhances the effectiveness of preventive measures against dengue fever. The aim of this study was to develop and validate a forecasting model that could predict dengue cases and provide timely early warning in Singapore. We developed a time series Poisson multivariate regression model using weekly mean temperature and cumulative rainfall over the period 2000–2010. Weather data were modeled using piecewise linear spline functions. We analyzed various lag times between dengue and weather variables to identify the optimal dengue forecasting period. Autoregression, seasonality and trend were considered in the model. We validated the model by forecasting dengue cases for week 1 of 2011 up to week 16 of 2012 using weather data alone. Model selection and validation were based on Akaike's Information Criterion, standardized Root Mean Square Error, and residuals diagnoses. A Receiver Operating Characteristics curve was used to analyze the sensitivity of the forecast of epidemics. The optimal period for dengue forecast was 16 weeks. Our model forecasted correctly with errors of 0.3 and 0.32 of the standard deviation of reported cases during the model training and validation periods, respectively. It was sensitive enough to distinguish between outbreak and non-outbreak to a 96% (CI = 93–98%) in 2004–2010 and 98% (CI = 95%–100%) in 2011. The model predicted the outbreak in 2011 accurately with less than 3% possibility of false alarm. We have developed a weather-based dengue forecasting model that allows warning 16 weeks in advance of dengue epidemics with high sensitivity and specificity. We demonstrate that models using temperature and rainfall could be simple, precise, and low cost tools for dengue forecasting which could be used to enhance decision making on the timing, scale of vector control operations, and utilization of limited resources. Without effective drugs or a vaccine, vector control remains the only method of controlling dengue fever outbreaks in Singapore. Based on our previous findings on the effects of weather on dengue cases and optimal timing for issuing dengue early warning in Singapore, the purpose of this study was to develop a dengue forecasting model that would provide early warning of a dengue outbreak several months in advance to allow sufficient time for effective control to be implemented. We constructed a statistical model using weekly mean temperature and rainfall. This involved 1) identifying the optimal lag period for forecasting dengue cases; 2) developing the model that described past dengue distribution patterns; 3) performing sensitivity tests to analyze whether the selected model could detect actual outbreaks. Finally, we used the selected model to forecast dengue cases from 2011–2012 week16 using weather data alone. Our model forecasted for a period of 16 weeks with high sensitivity in distinguishing between an outbreak and a non-outbreak. We conclude that weather can be an important factor for providing early warning of dengue epidemics, long term sustainability of forecast precision is challenging considering the complex dynamics of disease transmission.

Climate variability and increase in incidence and magnitude of dengue incidence in Singapore

Hand, foot, and mouth disease (HFMD) outbreaks leading to clinical and fatal complications have increased since late 1990s; especially in the Asia Pacific Region. Outbreaks of HFMD peaks in the warmer season of the year, but the underlying factors for this annual pattern and the reasons to the recent upsurge trend have not yet been established. This study analyzed the effect of short-term changes in weather on the incidence of HFMD in Singapore. The relative risks between weekly HFMD cases and temperature and rainfall were estimated for the period 2001–2008 using time series Poisson regression models allowing for over-dispersion. Smoothing was used to allow non-linear relationship between weather and weekly HFMD cases, and to adjust for seasonality and long-term time trend. Additionally, autocorrelation was controlled and weather was allowed to have a lagged effect on HFMD incidence up to 2 weeks. Weekly temperature and rainfall showed statistically significant association with HFMD incidence at time lag of 1–2 weeks. Every 1°C increases in maximum temperature above 32°C elevated the risk of HFMD incidence by 36% (95% CI = 1.341–1.389). Simultaneously, one mm increase of weekly cumulative rainfall below 75 mm increased the risk of HFMD by 0.3% (CI = 1.002–1.003). While above 75 mm the effect was opposite and each mm increases of rainfall decreased the incidence by 0.5% (CI = 0.995–0.996). We also found that a difference between minimum and maximum temperature greater than 7°C elevated the risk of HFMD by 41% (CI = 1.388–1.439). Our findings suggest a strong association between HFMD and weather. However, the exact reason for the association is yet to be studied. Information on maximum temperature above 32°C and moderate rainfall precede HFMD incidence could help to control and curb the up-surging trend of HFMD.