Space-time scanning statistics in the prediction and evaluation of dengue epidemic clusters

Objectives: To detect clusters of dengue hemorrhagic fever in an urbanized district of Hai Phong City, Vietnam using Poisson space-time retrospective and prospective analysis. Methods: A cross-sectional and retrospective study analyzed dengue surveillance data in the period from January 01, 2018, to December 31, 2022. Spatial-temporal scanning statistics were performed using the free software SatScan v10.1.2. Results: A total of 519 cases were recorded. The cumulative incidence per 100,000 inhabitants was 3.37, 127.36, 10.96, 0, and 296.04 in 2018, 2019, 2020, 2021, and 2022, respectively. By retrospective Poisson model-based analysis, seven clusters were detected. Six of these seven detected outbreaks occurred in November and December 2022. The largest cluster had a relative risk (RR) of 1539.5 ( P < 0.00001). The smallest cluster has a RR of 316.1 ( P = 0.006). Prospective analysis using the Poisson model significantly detected four active case clusters at the time of the study. The largest cluster of cases with RR was 47.7 ( P < 0.00001) and the smallest cluster with RR was 18.2 ( P < 0.00001). Conclusions: This study provides a basis for improving the effectiveness of interventions and conducting further investigations into risk factors in the study area, as well as in other urban and suburban areas nationwide.