Bayesian hierarchical model for mapping positive patient COVID-19 in Surabaya, Indonesia

Abstract

Covid-19 is a disease caused by the coronavirus and has infected the world population not only in developing countries like Indonesia but also in developed countries like the United States of America. The spread of the disease originating from Wuhan in China is evenly distributed throughout the world. Surabaya, as one of the major cities in Indonesia was also affected by Covid-19. The aim of this study was to map the relative risk of the spread of positive Covid-19 patients in Surabaya using the Bayesian Hierarchical Model with spatial analysis to deal with regional dependencies. The object used was spatial object as many as 154 villages in Surabaya. Meanwhile, a number of positive patients Covid-19 as of May 10th, 2020 was as many as 708 patients. The method used to estimate the relative risk was the Bayesian method with the Integrated Nested Laplace Approximation (INLA) approach. It was used after several studies have shown that the INLA approach is more accurate in providing estimated values compared to the Maximum-Likelihood estimation. The mapping results showed that there is a spatial dependence on the spread of Covid-19 disease in Surabaya.