Modeling of Prediction Tuberculosis Prevalence Based Geographically Weighted Poisson Regression

Suprajitno, Suprajitno, Mugianti, Sri, Martiningsih, Wiwin and Wagiyo, Wagiyo (2016) Modeling of Prediction Tuberculosis Prevalence Based Geographically Weighted Poisson Regression. International Journal of Science and Research (IJSR). pp. 1477-1482. ISSN 2319-7064

[img]
Preview
Text
ART2016488.pdf

Download (351kB) | Preview

Abstract

Tuberculosis is an infectious disease was cause of death. Tuberculosis patient were continues each year. In epidemiology, the environment and the human were two factors of tuberculosis prevalence, particularly individuals at risk of infection. Basically, tuberculosis prevalence able to predicted with statistically to determine of the two main factors in epidemiology. The purpose of this research was to produce a formula prediction model based GWPR (Geographically Weighted Poisson Regression). Methods: Research design was descriptive. The research sample as much as 111 Tb patients were selected by simple random sampling, which all pat ients were recorded in Dinas Kesehatan Kota and Kabupaten Blitar on 2015, January – May. Inclusion criteria were Non MDR (multi drug resistance) and not being hospitalized. Environmental factors were predictors is spacious house, spacious living room, spacio us bedrooms, number of bedroom windows, spacious bedroom window, living room temperature, humidity living room, and the amount o f sunlight entering the homes of people Tb. The human factor was a patient body weight. Result:, where X1 = weight, X2 = area of homes of people, X3 = spacious living room patients, X4 = spacious bedrooms patients, X5 = number of bedroom window sufferers, X6 = spacious bedroom window sufferers, X7 = the temperature of the patient's living room, X8 = humidity living room patients, and X9 = the amount of light entering the living room. Analysis: Formula produce a proportion of 0.07% and a predictor effect by 27%. Discuss: Before using a formula to predict tuberculosis prevalence needed to measurements variables and population that can be predicted precisely match the desired time.

Item Type: Article
Subjects: ?? R1 ??
Divisions: ?? ijsr ??
Depositing User: FRISCO TALISTI
Date Deposited: 22 Feb 2018 07:39
Last Modified: 22 Feb 2018 07:39
URI: http://repository.phb.ac.id/id/eprint/57

Actions (login required)

View Item View Item