Evaluating the impact of decentralized testing for tuberculosis in Ghana: A simulation model

Area of Science:

• Public Health
• Epidemiology
• Health Systems Research

Background:

• Geographical inaccessibility to tuberculosis (TB) diagnostic services in Ghana hinders timely diagnosis.
• Decentralizing molecular testing for TB can improve case notification and treatment initiation, especially in underserved areas.

Purpose of the Study:

• To project the epidemiological impact of decentralized TB testing at the country level in Ghana.
• To evaluate the effects of expanding Xpert MTB/RIF testing to district-level facilities.

Main Methods:

• An individual-based simulation of Ghana's TB diagnostic system and treatment cascade was developed.
• The model was calibrated using District Health Information Management System (DHIMSII) data from 2019-2022.
• Two decentralization scenarios were assessed against the current diagnostic network.

Main Results:

• Under the baseline, 26% of samples were untested, with over 30% of diagnoses relying on clinical judgment and 35% false positives.
• Decentralizing Xpert to district hospitals reduced untested samples to 20% and false positives to 32%, potentially saving 43 lives per million by 2030.
• Full decentralization to all facilities could eliminate untested samples, reduce false positives to 25%, and save 180 lives per million by 2030.

Conclusions:

• Decentralizing Xpert MTB/RIF testing significantly improves diagnostic accuracy and TB treatment outcomes.
• Wide-scale implementation is crucial to achieve the full benefits of decentralized TB diagnostics in Ghana.
• This strategy enhances TB case finding and reduces unnecessary treatments, ultimately saving lives.