@article{10.11648/j.ajtas.20261503.12,
  author = {Samuel Okinyi and Joseph Eyang’an Esekon and Martin Mutwiri Kithinji},
  title = {Modelling Malaria Incidence Using Poisson Generalized Linear Model: A Case Study Of Mbita Sub-County, Kenya
},
  journal = {American Journal of Theoretical and Applied Statistics},
  volume = {15},
  number = {3},
  pages = {88-103},
  doi = {10.11648/j.ajtas.20261503.12},
  url = {https://doi.org/10.11648/j.ajtas.20261503.12},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajtas.20261503.12},
  abstract = {Malaria continues to pose a major threat to public health across the world, particularly in African countries where infection rate s remain high. Nearly half of the global population is exposed to the risk of contracting the disease. Malaria is caused by parasites belonging to the Plasmodium family and affects both humans and other warm-blooded animals. Over the years, researchers have explored different causes of malaria transmission using techniques such as spatial analysis, time series methods and regression models. Although these methods are useful, they are less suitable when the data set involved are categorical or count variables. This study used Poisson Generalized Linear Model to investigate factors associated with malaria incidence in Mbita Sub-county. The variables considered included treated mosquito bed net use, age group, educational level of household heads and access to healthcare services. The Poisson Generalized Linear Model was fitted having estimated its parameters. The findings showed that treated mosquito net usage, age group, and access to healthcare facilities were statistically significant. The educational background of the household head was not significant. The association between the exploratory variables and the response variable was determined by use of the Chi-square test.The results indicated that there was an association between mosquito bed net use and malaria, age group and malaria, education level of family head and malaria and finally healthcare access and malaria. The goodness of fit was conducted by the use of deviance statistic. A comparison between the null model and the full model was done and this resulted into a p-value of 0.001871, which was below the 0.05 significance threshold. As a result, the null hypothesis was rejected, indicating that additional exploratory variable improve the model. This suggests that the full model together with additional parameters significantly improves the fit of the model to the data. The study’s findings reinforce existing evidence that the use of treated mosquito bed net use plays an important role in lowering malaria infections. organizations that handle matters in relation to health and environment such as World Health Organization and United Nations may apply the outcome to aid in developing mechanisms to lower the spread of malaria within Mbita Sub-county and other parts of the world with similar settings.},
 year = {2026}
}

Copy | Download

TY  - JOUR
T1  - Modelling Malaria Incidence Using Poisson Generalized Linear Model: A Case Study Of Mbita Sub-County, Kenya

AU  - Samuel Okinyi
AU  - Joseph Eyang’an Esekon
AU  - Martin Mutwiri Kithinji
Y1  - 2026/06/12
PY  - 2026
N1  - https://doi.org/10.11648/j.ajtas.20261503.12
DO  - 10.11648/j.ajtas.20261503.12
T2  - American Journal of Theoretical and Applied Statistics
JF  - American Journal of Theoretical and Applied Statistics
JO  - American Journal of Theoretical and Applied Statistics
SP  - 88
EP  - 103
PB  - Science Publishing Group
SN  - 2326-9006
UR  - https://doi.org/10.11648/j.ajtas.20261503.12
AB  - Malaria continues to pose a major threat to public health across the world, particularly in African countries where infection rate s remain high. Nearly half of the global population is exposed to the risk of contracting the disease. Malaria is caused by parasites belonging to the Plasmodium family and affects both humans and other warm-blooded animals. Over the years, researchers have explored different causes of malaria transmission using techniques such as spatial analysis, time series methods and regression models. Although these methods are useful, they are less suitable when the data set involved are categorical or count variables. This study used Poisson Generalized Linear Model to investigate factors associated with malaria incidence in Mbita Sub-county. The variables considered included treated mosquito bed net use, age group, educational level of household heads and access to healthcare services. The Poisson Generalized Linear Model was fitted having estimated its parameters. The findings showed that treated mosquito net usage, age group, and access to healthcare facilities were statistically significant. The educational background of the household head was not significant. The association between the exploratory variables and the response variable was determined by use of the Chi-square test.The results indicated that there was an association between mosquito bed net use and malaria, age group and malaria, education level of family head and malaria and finally healthcare access and malaria. The goodness of fit was conducted by the use of deviance statistic. A comparison between the null model and the full model was done and this resulted into a p-value of 0.001871, which was below the 0.05 significance threshold. As a result, the null hypothesis was rejected, indicating that additional exploratory variable improve the model. This suggests that the full model together with additional parameters significantly improves the fit of the model to the data. The study’s findings reinforce existing evidence that the use of treated mosquito bed net use plays an important role in lowering malaria infections. organizations that handle matters in relation to health and environment such as World Health Organization and United Nations may apply the outcome to aid in developing mechanisms to lower the spread of malaria within Mbita Sub-county and other parts of the world with similar settings.
VL  - 15
IS  - 3
ER  - 

Copy | Download