PEMODELAN KASUS BARU HIV DI JAWA BARAT MENGGUNAKAN POISSON INVERSE GAUSSIAN REGRESSION

Abstract

West Java is ranked as the province with the highest number of HIV cases in Indonesia in 2022 with 8,680 cases and is still lagging behind in achieving one of the government's targets to reach 95% of People Living with HIV (ODHIV) knowing their status that they have been infected with HIV, which is currently still at 81%. This is the basis of this study which aims to determine what factors have a significant effect on the number of HIV cases in West Java Province. In Poisson Regression analysis, the mean and variance assumptions are often not met because there can be overdispersion or underdispersion which can be overcome using PIG Regression. PIG Regression has flexible properties that can overcome overdispersion and underdispersion cases well, compared to Negative Binomial Regression which is only able to overcome overdispersion problems. The results of this study indicate that the characteristics of areas with the highest number of new HIV cases are influenced by high population mobility, low HIV counseling, and the availability of adequate health facilities and personnel for case detection. Conversely, areas with the lowest number of cases have high health screening coverage. PIG regression modeling found that the factors that have a significant effect on the number of new HIV cases in West Java in 2022 are the percentage of adolescents who received HIV counseling, the percentage of the population 15 years and over who are not HIV positive.

Keywords:  HIV Cases, Overdispersion, PIG Regression, Poisson Regression, Socio-economic.