PREVALENCE AND MODEL PREDICTIONS OF MYCOBACTERIUM TUBERCULOSIS IN NIGER STATE, NIGERIA

Abstract

Nigeria is among the hotspots of tuberculosis (TB) globally, with subnational variations in the disease dynamics. Niger State contributes substantially to the national TB notifications, yet current incidence patterns and short-term forecasts remain poorly characterised. Reliable incidence data and predictive models are essential for targeted interventions and resource planning to meet End TB Strategy milestones. This study determined the incidence and predicted future trends of Mycobacterium tuberculosis in Niger State. We performed a retrospective review of all bacteriologically and clinically confirmed TB cases reported to the Niger State Tuberculosis and Leprosy Control Program from January 2014 to December 2024. Annual incidence rates per 100,000 population were calculated using National Population Commission projections. An Auto-Regressive Integrated Moving Average (ARIMA) time series model was fitted to historical data to forecast incidence for the year 2030. Model performance was evaluated using Mean Absolute Percentage Error (MAPE). There was a massive increase in new cases by 991%, from 1,044 cases in 2014 to 10,341 cases in 2024, which represents an average annual increase of approximately 927 cases. This massive increase is likely due to improved case detection capacity, enhanced surveillance systems, and stronger diagnostic infrastructure rather than a real epidemic peak. Significant spatial clustering occurred in urban regions including Minna, Suleja, and Bida. ARIMA (1,1,1) provided optimal fit with MAPE 8.2%. The model predicts incidence will increase to 68.4/100,000 by 2027 under current conditions. The infection incidence in Niger State shows a consistent upward trajectory with urban clustering. Without enhanced case finding and control measures, the burden will increase through 2027. Findings support prioritising active case detection in hotspots, expanding molecular diagnostic coverage, and using predictive models for state-level TB program planning.