TITLE:
Statistical Modeling of the Spread of Mpox Disease in Burundi
AUTHORS:
Menedore Karimumuryango, Florence Munezero, Olivier Nijimbere, Léonce Nzisabira, Jean Bosco Girukwisha, Kokou Nouwame Alinon, Anatole Nkeshimana, Stany Nduwimana, Issa Diallo, Xavier Crespin
KEYWORDS:
Statistical Modeling, Mpox, Health District, Burundi
JOURNAL NAME:
Open Journal of Statistics,
Vol.16 No.5,
October
26,
2026
ABSTRACT: Mpox has become a significant public health problem and Burundi is one of the countries that has been affected by Mpox outbreak. This study aimed to explore the factors that can influence the variation of the Mpox incidence rates across 49 health districts by using aggregated data. This was a descriptive and analytical cross-sectional study using three statistical models: Poisson Regression, Mixed-Effect Poisson Regression and Bayesian Hierarchical Model. For the Poisson Regression, the number of health facilities (IRR = 1.1), laboratory workers (IRR = 1.01), community health workers group (IRR = 0.88), population density (IRR = 0.99), number of people outside the districts using first antenatal consultation services (IRR = 0.92) and number of children full vaccinated from outside the districts (IRR = 1.11) were significant. For the Mixed-Effect Poisson Regression Model, the random effects between groups are almost nil. For the Bayesian Hierarchical Model, the number of laboratory workers had the plot that showed a good convergence. Based on the aggregated data, the results show that in order to improve the reporting of new Mpox case, it is necessary to act on availability of different categories of resources for health promotion, diagnosis and case management. A study based on individual data could complete the findings of this study.