Statistical modeling for identifying chikungunya high-risk areas of two large-scale outbreaks in Thailand's southernmost provinces

Abstract

Abstract Chikungunya fever (CHIKF) has re-emerged in the southernmost Thailand and presents a significant threat to public health. The problem areas can be identified using appropriate statistical models. This study aimed to determine the geographic epidemic patterns and high-risk locations. Data on CHIKF’s case characteristics, including age, gender, and residence sub-district, were obtained from the Office of Disease Prevention and Control of Thailand from 2008 to 2020. A logistic model was applied to detect illness occurrences. After removing records with no cases, a log-linear regression model was used to determine the incidence rate. The results revealed that two large-scale infections occurred in the southernmost provinces of Thailand between 2008 and 2010, and again between 2018 and 2020, indicating a 10-year epidemic cycle. The CHIKF occurrence in the first and second outbreaks was 28.4% and 15.5%, respectively. In both outbreaks of occurrence CHIKF, adolescents and working-age groups were the most infected groups but the high incidence rate of CHIKF was elderly groups. The first outbreak had a high occurrence and incidence rate in 39 sub-districts, the majority of which were in Narathiwat province, whilst the second outbreak was identified in 15 sub-districts, the majority of which were in Pattani province. In conclusion, the CHIKF outbreak areas can be identified and addressed by combining logistic and log-linear models in a two-step process. The findings of this study can serve as a guide for developing a surveillance strategy or an earlier plan to manage or prevent the CHIKF outbreak.