Small Area Estimation (SAE) is the process of using statistical models to link survey outcome variables to a set of predictor variables known for small domains, in order to predict domain-level estimates. The need for detailed statistics on small area is constantly increasing. Small area estimation is becoming important in survey sampling due to a growing demand for reliable small area statistics from both public and private sectors. Bayesian hierarchical models provide a convenient framework for disease mapping and geographical correlation studies. Computation may be carried out using the freely-available WinBUGS software. Two approaches prediction to estimate total patient in small area i will be presented. For the purpose of this paper, the small area estimation in this context use data of Indnesia's population based on the 2000 census for the population of Jakarta and data of patient diarrhea from District Health Service of Jakarta. We interest to predict total patient of diarrhea as variable of interest and data population as auxiliary data from unsample for each small area.