Sero-Prevalence and Predictive Statistical Model for the Hepatitis B Virus Exposure Among People in Community, Yasothon, Thailand

Abstract

This study investigated the prevalence of Hepatitis B virus (HBV) and identified a predictive statistical model for the HBV exposure among people in the community, Yasothon, Thailand. A cross-sectional study was performed on participants over 26 years old and living in Muang district, Yasothon province, Thailand. The research was conducted from July to August 2019. All 1,258 participants were verbally screened. Four hundred and fifty nine people were the risk group and tested for HBsAg, and 18 cases were positive for HBsAg (3.9% [95%CI 3.5-4.4]). For the predictive model, the HBV exposure connected with sex, marital status, alcohol, smoking, and knowledge. The area under the receiver operating characteristics (ROC) curve was 61.8 % (95%CI, 58.6 to 65.0). At cut-off-point -0.66, the sensitivity, specificity and accuracy were 72.6%, 42.4 % and 53.4%, respectively. HBV infection was a serious health problem, it can cause cirrhosis and liver cancer in the future. The predictive model of five variables can predict risk exposure of HBV which may had other relevant factors. Verbal screening by questionnaire to classify HBsAg risk group can lower the implement cost.