Integer time series models for tuberculosis in Africa.

Area of Science:

• Epidemiology
• Biostatistics
• Public Health

Background:

• Tuberculosis (TB) is a leading airborne infectious disease globally.
• The African region faces a high burden of TB cases.
• Accurate forecasting of TB incidence is crucial for effective control.

Purpose of the Study:

• To model and predict new tuberculosis cases in African countries.
• To identify the most suitable statistical models for TB time series data.
• To provide forecasts for TB incidence from 2022 to 2031.

Main Methods:

• Integer time series analysis was applied to TB case data from 2000-2021.
• Twenty distinct statistical models, primarily Poisson and negative binomial distributions, were fitted.
• Model selection was based on the best data fit for each African country.

Main Results:

• The study identified the optimal time series models for TB incidence in various African nations.
• Fitted models incorporated linear or quadratic trends over time.
• These selected models serve as the basis for future TB case predictions.

Conclusions:

• The chosen models provide a robust framework for understanding and forecasting TB dynamics in Africa.
• These predictions can inform targeted public health interventions and resource allocation.
• Continued monitoring and modeling are essential for TB control in the region.