Spatial scan statistic for aids hotspots detection at regencies and municipalities in Java

Abstract

Scan statistic is a statistical method with many potential applications. It is designed to detect a local excess of events and to test if such an excess can reasonably have occurred by chance. It can be applied on wide area of interest, such as on disease clustering. In this research spatial scan statistic was applied important to detect the locations of the significant hotspot clusters of severe and dangerous disease, AIDS. This research applied the probability distribution generating events of poisson under the null hypothesis, supported by likelihood ratio test analysis to get the statistic test value and a Monte Carlo simulation to obtain the significant hotspot cluster from calculation of those statistic test values in computational process algorithm. Analysis with 19 percents of the population at risk of maximum spatial cluster size and the cilcular spatial window shape, spatial scan statistic presented the significant hotspot clusters of AIDS cases in Java. The places considered as hotspot are the municipalities of Center Jakarta and North Jakarta, Bandung, Malang, Surahaya, Semarang, Yogyakarta, and Salatiga. Based on the fact, government has to provide AIDS eradication and surveillance program in the hotspot area while the program of prevention must be implemented in the surrounding area (non hotspot area).