Pneumonia is a major respiratory infection that significantly affects toddlers. Based on Riskesdas 2021, East Java recorded the highest number of pneumonia cases in Indonesia. This study aims to model the number of toddler pneumonia cases in East Java using the Generalized Additive Model (GAM) to address the issue of overdispersion, comparing the performance of Poisson and Negative Binomial distributions. This study utilized secondary data sourced from the East Java Health Profile 2024, with a total of 38 districts/cities as the unit of observation. The analysis technique employed GAM with smoothing functions to capture non-linear relationships between the response and predictor variables, which include nutritional status, breastfeeding, immunization, vitamin A, and household air quality, followed by an overdispersion test. The results indicate that the GAM with a Negative Binomial distribution is superior to the Poisson model in handling overdispersion, as evidenced by a significantly lower deviance of 507.7 versus 8932.6. In conclusion, the Negative Binomial GAM provides a more accurate model for this case. Furthermore, improving public health awareness and upgrading health facilities are crucial steps to more effectively managing toddler pneumonia cases.
Copyrights © 2026