International Journal of Physical Sciences Vol. 2(4), pp. 061-068, April 2014 ISSN 2331-1827 ©2014 Academe Research Journals

Full Length Research Paper

Modeling the prevalence of malaria in Niger State: An application of Poisson regression and negative binomial regression models

Evans, O. Patience and Adenomon, M. Osagie*

Department of Mathematics and Statistics, The Federal Polytechnic, Bida, P.M.B. 55, Niger State, Nigeria, West Africa.

*Corresponding author. E-mail: admonsagie@yahoo.com.

Accepted 24 April, 2014

Abstract

In statistics, Generalized Linear Models (GLM) are an extension of the linear modeling process that allow model to be fit to data that follow probability distributions other than the normal distribution. Poisson regression model is a special case of a generalized linear model (GLM) with a log link - this is why the Poisson regression may also be called Log-Linear Model. The Poisson distribution is often used to model rare events while the Poisson regression model is very suitable for modeling count data. One of the problems of Poisson regression is that it is affected by overspreading. In that case, it is advisable to employ the Negative binomial regression. In this work, we study the trend of malaria prevalence in Minna, Niger State using monthly malaria outpatient data collected from the General Hospital, Minna. The Poisson regression and the Negative binomial regression models were used in the analysis. The results from the Poisson regression and the Negative binomial regression models revealed an increase of 0.053 and 0.054 per month respectively. In addition, the incidence rate ratios (IRR) revealed that the prevalence of Malaria in Minna, Niger State increased by approximately 6% every month. Our work therefore recommended that more effective measures by the Nigeria government and NGOs should be geared towards reducing the prevalence and danger posed by malaria in Nigeria.

Key words: Poisson, modeling, malaria, prevalence, negative binomial, regression, models.