Do electronic medication monitors improve tuberculosis treatment outcomes? Programmatic experience from China.

Area of Science:

• Public Health
• Infectious Diseases
• Health Technology Assessment

Background:

• An indigenously developed electronic medication monitor (EMM) was utilized in China to indirectly measure tuberculosis treatment adherence.
• Previous studies indicated satisfactory EMM uptake but frequent missing adherence data and infrequent shifts to directly observed therapy (DOT) for non-adherent patients.

Purpose of the Study:

• To assess the relative differences in unfavorable outcomes and deaths among tuberculosis patients using EMM-supported self-administered therapy (SAT) at baseline compared to SAT alone.
• To evaluate the effectiveness of EMM in improving tuberculosis treatment adherence and patient outcomes in a programmatic setting.

Main Methods:

• A cohort study utilizing secondary data from 30 counties in China (July-December 2018).
• Intention-to-treat analysis was employed, with modified Poisson regression used to assess associations between EMM use and outcomes.
• Eligible participants received EMM-supported SAT, with comparisons made between those starting EMM at baseline versus those not using EMM initially.

Main Results:

• No significant differences in unfavorable outcomes were observed between patients who started EMM at baseline (6.3%) and those who did not (6.7%).
• Fewer deaths were noted in the EMM at baseline group (2.5%) compared to the non-EMM group (3.5%), though this difference was not statistically significant (p=0.191).
• The lack of association persisted after adjusting for confounders such as occupation, TB classification, and TB category.

Conclusions:

• In programmatic settings, electronic medication monitors (EMMs) did not demonstrate a significant impact on tuberculosis treatment outcomes.
• Optimization of EMM implementation is necessary, including timely shifts to directly observed therapy (DOT) when indicated, addressing missing data issues, and ensuring continuous device usage.