A spatiotemporal analysis of the spread of African swine fever in Vietnam in 2019
Source: PubMed Central Open Access, NCBI / U.S. National Library of Medicine
Introduction African swine fever (ASF) was first detected in Vietnam in early February 2019 and spread rapidly nationwide. By December 2019, ASF had been reported in 8,509 of 11,055 communes (77%) across all 63 provinces and municipalities. Methods Descriptive analyses were performed to summarize temporal and spatial patterns of ASF-affected communes (outbreaks). Outbreak data were aggregated weekly (week 1–48) based on onset date, and weekly counts, cumulative numbers, and proportions of affected communes were calculated. To quantify the early spread of ASF, we analyzed commune-level outbreak data from 2019. Temporal progression was assessed using linear and negative binomial regression models, whereas spatiotemporal clusters were identified using a space–time permutation model. Results The linear regression analysis indicated an approximately 2.5 percentage-point weekly increase in the cumulative proportion of affected communes (= 0.94) from 1 February to 14 July 2019. The negative binomial regression model estimated an approximately 7.6% weekly increase in affected commune counts (= 0.073) from 1 February to 14 July 2019, corresponding to a doubling time of approximately 9.5 weeks. Consequently, more than half of the communes were affected within the first 24 weeks of the first detected case. A total of 18 statistically significant spatiotemporal clusters were identified. The earliest cluster was detected in the North, where the first outbreak was confirmed, and the most n
Abstract
Introduction African swine fever (ASF) was first detected in Vietnam in early February 2019 and spread rapidly nationwide. By December 2019, ASF had been reported in 8,509 of 11,055 communes (77%) across all 63 provinces and municipalities. Methods Descriptive analyses were performed to summarize temporal and spatial patterns of ASF-affected communes (outbreaks). Outbreak data were aggregated weekly (week 1–48) based on onset date, and weekly counts, cumulative numbers, and proportions of affected communes were calculated. To quantify the early spread of ASF, we analyzed commune-level outbreak data from 2019. Temporal progression was assessed using linear and negative binomial regression models, whereas spatiotemporal clusters were identified using a space–time permutation model. Results The linear regression analysis indicated an approximately 2.5 percentage-point weekly increase in the cumulative proportion of affected communes (= 0.94) from 1 February to 14 July 2019. The negative binomial regression model estimated an approximately 7.6% weekly increase in affected commune counts (= 0.073) from 1 February to 14 July 2019, corresponding to a doubling time of approximately 9.5 weeks. Consequently, more than half of the communes were affected within the first 24 weeks of the first detected case. A total of 18 statistically significant spatiotemporal clusters were identified. The earliest cluster was detected in the North, where the first outbreak was confirmed, and the most numerous clusters were in the Central region. Many cluster centers were located in areas of relatively high pig density, particularly in the Red River Delta and Mekong River Delta. Conclusion These findings highlight the extensive geographic scale of ASF transmission in Vietnam in the first year of the epidemic. The rapid and extensive spread observed suggests that a stamping-out strategy, as typically implemented, may face substantial challenges under such epidemic conditions. Instead, the results emphasize the importance of impact-mitigation strategies and long-term management approaches, including enhanced biosecurity, farmer support, and measures to maintain a stable pork supply. These findings also provide relevant insights for countries facing similar endemic conditions or at risk of ASF introduction.