Modeling the bacterial meningitis death cases and case fatality rates over Nigeria

Abstract

Meningitis a communicable disease caused by infection and inflammation of the meninges leading to a substantial morbidity and mortality in the whole world. Meningitis epidemic occurs frequently across the African meningitis belts. This article presents analysis of bacterial meningitis incidence and mortality cases over Nigeria. A negative binomial regression model was fitted for number of bacterial meningitis death cases based on reported cases, laboratory confirmed cases and their interaction. Also, a dynamic regression model was fitted for the fatality rates of bacterial meningitis under the assumption that the residuals of linear regression model are auto-correlated. Results from this study show that there were significantly increasing trends in the number of bacterial meningitis reported cases, laboratory confirmed cases and death cases. The number of death cases is significantly affected by number of reported cases, number of laboratory confirmed cases and interaction between them. The decrease in case fatality rate may depend on number of laboratory confirmed individuals and total reported cases. The Ljung–Box test shows that the residuals of the fitted models are not auto-correlated. Observed data are compared with the fitted data from the optimal models. Results show that the optimal models fit the data well. Fold change is estimated based on crude case-fatality risk to investigate whether there is massive underreporting and under-testing of bacterial meningitis cases in Nigeria. There is an evidence of massive underreporting and under-testing of bacterial meningitis cases in Nigeria.

Keywords:

• Meningitis
• massive under-reporting
• negative Binomial model
• case fatality
• auto-correlated residuals
• Nigeria

Disclosure statement

The authors declare that they have no conflicting interest.