Abstract
BACKGROUND: TB preventive treatment (TPT) is crucial for preventing the progression from TB infection (TBI) to active disease, particularly among vulnerable populations such as indigenous people. Although shorter regimens have improved adherence, evidence from real-world programmatic settings is limited. This study compared TPT completion rates between indigenous and non-indigenous populations in the Brazilian Amazon. METHODS: We conducted a retrospective cohort study using data from the IL-TB system (2019–2023) in Amazonas, Brazil. Individuals who initiated TPT were stratified by race. The primary outcome was TPT completion. Multivariate logistic regression identified factors associated with completion, adjusting for sociodemographic, clinical, and operational variables. RESULTS: Of the 4,887 individuals (non-indigenous = 4,666; indigenous = 221), 79.6% completed TPT. Completion was higher among indigenous individuals (90.9%) than among non-indigenous individuals (79.1%) (p < 0.001). In the adjusted analysis, indigenous race (OR: 2.4; 95% CI: 1.49–4.06), shortened TPT regimens (OR: 2.36; 95% CI: 2.04–2.73), TBI confirmed by TST/IGRA (OR: 1.57; 95% CI: 1.06–1.65), indication of TPT for PLHIV (OR: 1.31; 95% CI: 1.04–2.73) and receiving TPT in health facilities outside the capital (OR: 2.03; 95% CI: 1.49–2.82) were positively associated with completion of TPT. Foreign nationality (OR: 0.28; 95% CI: 0.18–0.44) and receiving TPT in referral hospitals/specialist services—not exclusive to PCH (OR: 0.79; 95% CI: 0.66–0.94)—were negatively associated with such outcomes. Among non-indigenous individuals, associations mirrored those in the overall population. Among indigenous individuals, only the shortened regimen was significantly associated with completion of TPT (OR: 5.75; 95% CI: 1.78–23.4), although this estimate should be interpreted with caution, given the small subgroup size and wide confidence interval. CONCLUSIONS: In this Amazon cohort, TPT completion was greater among indigenous individuals and was associated with shortened regimens and facilities outside the state capital. These patterns suggest that expanding shortened regimens and PHC-based access beyond capital may support adherence in high-burden, inequity-affected settings; however, the programmatic impact should be confirmed in prospective or quasi-experimental evaluations. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12889-025-25177-8.