Subnational burden estimates to find missing people with tuberculosis: wrong but useful?

Area of Science:

• Epidemiology
• Public Health
• Biostatistics

Background:

• Reliable subnational data is crucial for tuberculosis (TB) control efforts, but estimates at lower disaggregation levels (district, community, ward) are scarce.
• Existing approaches for generating subnational TB burden estimates face challenges in accuracy validation and programmatic integration.

Purpose of the Study:

• To review subnational TB models, assessing their accuracy, purpose, and utility in identifying individuals with TB.
• To explore the role of these models in spatial targeting and programmatic learning for TB interventions.

Main Methods:

• A review of various subnational TB models was conducted.
• The analysis focused on model accuracy, intended use, and practical application in TB program planning.

Main Results:

• Subnational TB models, despite accuracy limitations, can effectively guide spatial targeting of interventions.
• These models contribute to understanding intervention effectiveness and facilitate programmatic learning.

Conclusions:

• Subnational TB models are valuable evidence-based tools for guiding the search for missing individuals with TB.
• Increased utilization and systematic comparison with programmatic data can enhance model accuracy and usefulness over time.