Indonesia, as a tropical country, continues to face mosquito-borne diseases, particularly Dengue Fever (DF), transmitted by the Aedes aegypti mosquito. This disease remains a significant public health issue, with fluctuating but generally high incidence rates across various regions. MARS (Multivariate Adaptive Regression Splines) is a non-parametric regression method that is adaptive in modeling non-linear relationships between dependent and independent variables and is capable of capturing interactions among independent variables. The best MARS model is one that has the lowest Generalized Cross Validation (GCV) and Mean Square Error (MSE) values. This study employs a quantitative approach using secondary data obtained from the 2023 Indonesia Health Survey Report. The MARS model for Dengue Fever prevalence in Indonesia is as follows: Y = 0.522 + 0.157 × BF1 – 0.046 × BF3 + 0.272 × BF8 + 0.038 × BF10 – 0.018 × BF12. The proportion of households without trash bins is the most influential factor affecting Dengue Fever prevalence in Indonesia, with an importance level of 100%. This is followed by the proportion of households not implementing mosquito control efforts, with an importance level of 37.811%, and the proportion of households without handwashing facilities, with an importance level of 35.669%. By inserting the Basis Function values into the model equation, it was concluded that the Dengue Fever prevalence in East Java Province (Y = 0.3881) is lower compared to the national prevalence in Indonesia (Y = 0.522) because variables X3 and X5 have lower values.
Copyrights © 2026