<![CDATA[Childhood tuberculosis (TB) remains a major public health problem in Indonesia, particularly in East Java Province. This study aimed to identify factors associated with childhood TB cases to support effective disease prevention and control. Since the number of TB cases represents count data that may exhibit overdispersion, Generalized Poisson Regression (GPR) was employed as an appropriate analytical approach. This ecological study used secondary data from 38 districts and municipalities in East Java Province, Indonesia, in 2024, obtained from the East Java Provincial Health Office. Multicollinearity was assessed using the Variance Inflation Factor (VIF). Because overdispersion was detected in the Poisson regression model, the analysis was subsequently performed using Generalized Poisson Regression. The final GPR model included the percentage of the population reporting health complaints, the average number of cigarettes smoked per week, the percentage of households with access to improved sanitation, the percentage of households with access to safe drinking water, the number of healthcare workers, and the number of healthcare facilities as explanatory variables. Model evaluation yielded an Akaike Information Criterion (AIC) value of 457.46 and an Adjusted R² value of 0.9813, indicating that the model explained 98.13% of the variation in childhood TB cases. These findings demonstrate that Generalized Poisson Regression is an appropriate method for modeling overdispersed childhood TB count data and provides useful evidence to strengthen tuberculosis surveillance and support targeted public health interventions for childhood TB control in East Java Province.]]>
