Background. In the Americas, endemic countries for Aedes-borne diseases such as dengue, chikungunya, and Zika face great challenges particularly since the recent outbreaks of CHIKV and ZIKV, all transmitted by the same insect vector Aedes aegypti and Ae. albopictus. The Special Program for Research and Training in Tropical Diseases (TDR- WHO) has developed together with partners an early warning and Response System (EWARS) for dengue outbreaks based on a variety of alarm signals with a high sensitivity and positive predictive value (PPV). The question is if this tool can also be used for the prediction of Zika and chikungunya outbreaks.
Methodology. We conducted in nine districts of Mexico and one large city in Colombia a retrospective analysis of epidemiological data (for the outbreak definition) and of climate and entomological data (as potential alarm indicators) produced by the national surveillance systems for dengue, chikungunya and Zika outbreak prediction covering the following outbreak years: for dengue 2012-2016, for Zika 2015-2017, for chikungunya 2014-2016. This period was divided into a “run in period” (to establish the “historical” pattern of the disease) and an “analysis period” (to identify sensitivity and PPV of outbreak prediction).
Results. In Mexico, the sensitivity of alarm signals for correctly predicting an outbreak was 92% for dengue, and 97% for Zika (chikungunya data could not be obtained in Mexico); the PPV was 68% for dengue and 100% for Zika. The time period between alarm and start of the outbreak (i.e. the time available for early response activities) was for dengue 6-8 weeks and for Zika 3-5 weeks. In Colombia the sensitivity of the outbreak prediction was 92% for dengue, 93% for chikungunya and 100% for Zika; the PPV was 68% for dengue, 92% for chikungunya and 54% for Zika; the prediction distance was for dengue 3-5 weeks, for chikungunya 10-13 weeks and for Zika 6-10 weeks.
Conclusion. The implementation of an early warning and response system (EWARS) could predict outbreaks of three Aedes borne diseases with a high sensitivity and positive predictive value and with a lag time long enough for preparing an adequate outbreak response in order to reduce the magnitude or avert the occurrence of outbreaks with their elevated social and economic tolls.