Analysis of the utilization and effectiveness of tuberculosis digital adherence technology in southeastern China

Abstract

Background: In response to the World Health Organization’s advocacy for patient-centered care, Fujian Province, China, initiated the promotion of Digital Adherence Technology (DAT) in 2024 as an optional strategy to enhance treatment adherence—a critical determinant of patient outcomes. This study aimed to evaluate the real-world implementation and effectiveness of this program. Methods: We conducted a multi-site study in four prefecture-level cities where DAT had been deployed to over 100 patients. Data were extracted from the national Tuberculosis Information Management System and linked with DAT platform records. Patients with DAT records were assigned to the DAT group, others to the conventional care group. We analyzed factors associated with DAT non-adoption. Adherence was compared between groups using three indicators: sputum smear examination rate at month two, smear conversion rate at month two (among bacteriologically confirmed patients), and loss to follow-up rate. Propensity score matching (PSM) was employed to balance baseline characteristics, followed by subgroup analyses. Results: The overall DAT utilization rate was 50.90%. Protective factors against DAT non-use included female sex (aOR = 0.829, 95% CI: 0.716–0.960), non-Han ethnicity (aOR = 0.555, 95% CI: 0.343–0.898), use of fixed-dose combination therapy (aOR = 0.806, 95% CI: 0.666–0.975), and concurrent extrapulmonary TB (aOR = 0.761, 95% CI: 0.642–0.902). Compared to Fuzhou, patients in Putian, Quanzhou, and Ningde had significantly lower odds of non-use (aORs: 0.024, 0.076, 0.030). Migrant status (aOR = 2.621, 95% CI: 2.144–3.203) and symptomatic presentation at diagnosis (aOR = 1.604, 95% CI: 1.401–1.836) were risk factors. The DAT group demonstrated a significantly higher month-two smear examination rate than the conventional group in both the full and PSM-matched cohorts (p < 0.001). No significant inter-group differences were found in month-two smear conversion rates or loss to follow-up. Conclusion: DAT utilization in Fujian remains moderate and is influenced by specific demographic and clinical factors. While DAT can improve process adherence indicators, its impact on definitive treatment outcomes requires further investigation. Optimization of patient management should include formal integration of DAT, workflow streamlining, and personalized intervention strategies. Future research must develop and evaluate effective interventions for DAT-identified non-adherent patients to translate monitoring data into improved clinical outcomes.