부산대학교 도서관

Intro: Severe outbreaks in Taiwan during 1987-2015 had always started from failure in detecting early dengue cases. However, the extreme low numbers of indigenous dengue cases in southern Taiwan during 2016-2022 have been succeeded with early efforts through integrated surveillance systems, including, free charge NS1 screening of virological surveillance for patients with dengue-like illness, fever surveillance at the airport and entomological surveillance looking for mosquito breeding sites. Recently, we used machine learning models to assist in finding lab.-confirmed dengue cases and predicting severe dengue cases in the elderly. Methods: We investigated factors that were associated with severe dengue outbreaks in southern Taiwan and used geographical information system (GIS) to plot spatial distribution of laboratory-confirmed dengue cases. Findings: Six epidemiological findings can direct prevention and control policies. (1) The onset of epidemics begins with imported dengue cases and appropriate weather conditions (high temperature, low humidity) enlightening to educate the symptoms/signs of dengue for travelers is important. (2) DENV-2, had resulted in most severe epidemics. (3) Once DENV- 2/DENV-3 circulates too long or severe cases occur, fatal dengue cases will increase alongside the prolongation of epidemic wave. (4) In areas with high population densities, either longer epidemic wave or higher transmission intensity can result in greater percentage of severe dengue cases. However, in areas with low population densities, both conditions can lead to greater percentage of severe dengue cases Therefore., interrupting DENV transmission efficiently can reduce severe/fatal cases. (5) Severe dengue cases frequently occur in tempo-spatial dengue clusters implying the importance to find out the sources of infection. The overlapping areas between high mosquito indices and dengue clusters facilitate transmission persistence. (6) Source reduction is more effective than insecticide-spraying. Conclusion: integrated surveillance systems, immediate interrupting transmission, and identifying mosquito breeding sites all together can reduce epidemic severity and thus promoting global health.