Original Research

A scoring scheme prediction model for dengue outbreaks using weather factors in Ho Chi Minh city, Vietnam


Background: The dengue infection cases are increasing in Ho Chi Minh city (HCMC), Vietnam. Previous studies have demonstrated the correlation between dengue cases and weather factors, which then are used to built prediction models for dengue outbreaks. However, the association between dengue and weather varies greatly between regions and locations. In HCMC, a tropical climate city in Vietnam, there is no such a weather-based prediction model for dengue outbreaks.

Objectives: This study aims to determine the correlation between weather factors and a weekly number of dengue cases and to develop a scoring scheme prediction model for dengue outbreaks using weather factors in HCMC, Vietnam. 

Methods: An ecological study was conducted on the evaluation of weekly time-series data from 1999 to 2017. A Poisson regression model coupled with Distributed Lag Non-Linear Model (DLNM) was constructed to evaluate the effects of weather factors (i.e., temperature, relative humidity, cumulative rainfall, wind speed) and the weekly dengue cases in HCMC with lag 1-12 weeks.

Results: The predictive model was based on the following weather factors: wind speed at lag 5-8 and 9-12 weeks; temperature amplitude and humidity at lag 5-8 weeks; rainfall at lag 1-4, 5-8, and 9-12 weeks. The predictive model using climate predictors explained about 80% of the variance in dengue cases with a small value of the mean absolute percentage error (MAPE= 0.17). The scoring scheme was then developed from the predictive model; it had a good prediction power – with the accuracy rate = 81%, sensitivity = 1, and specificity = 0.80. In summary, our study indicated that weather factors significantly influence and are predictors for the variation of dengue cases in Ho Chi Minh city, Vietnam. We recommend applying this model to improve the prevention of dengue outbreak.

Please insert your valid email address and click Free download

Recommended Article