Tuberculosis recurrence risk remains high years after treatment

Tuberculosis recurrence—when the disease returns after a seemingly successful course of treatment—poses a stubborn challenge to efforts to eliminate TB, especially in places where the disease is common. To better understand how often and how soon recurrence happens, and who is most at risk, Evelyn Lepka de Lima and colleagues looked at more than a decade of routine records from São Paulo state, Brazil. The team focused on people who completed treatment for a first episode of tuberculosis between 2013 and 2024, following them after treatment to see who developed a new, confirmed episode. By examining this large group of TB survivors they aimed to map the timing of recurrences and identify the demographic, clinical, behavioral, and social factors present at the start of the first treatment that predicted later relapse. The goal was to provide clear evidence to help design post-treatment care policies and programs that can catch and prevent recurrence, rather than assuming cure at the end of standard therapy.

This was a retrospective cohort study using routine surveillance data. Researchers drew their records from TBweb, the state tuberculosis surveillance system, and linked those records with the national mortality database to account for deaths. They used competing risks survival analysis to estimate recurrence incidence and to test risk factors, measuring covariates at treatment initiation for the first tuberculosis episode. The cohort included 154,579 people who completed treatment; over a median follow-up of 5.4 years (IQR 2.5–8.2) the team identified 9,464 first recurrences. The overall recurrence incidence was 1,127.2 per 100,000 person-years. Recurrence rates peaked between 3–12 months after treatment completion but remained above 1,000 per 100,000 person-years during years 2–5. Models stratified by age (<15 vs. ≥15 years) found strong independent risk factors in adults: hospitalization (subdistribution hazard ratio [SHR] 2.53, 95%CI 2.36–2.71), incarceration during the initial episode (SHR 2.30, 95%CI 2.13–2.48), and an initial diagnosis of pulmonary tuberculosis (SHR 2.10, 95%CI 1.87–2.35). In children, HIV carried the highest risk (SHR 6.38, 95%CI 2.64–15.40).

The findings make clear that finishing treatment is not the same as being free from risk. The high incidence of recurrence shortly after treatment—and the persistence of elevated risk for several years—suggest that one-time confirmation of cure should be followed by planned post-treatment care. The results point to groups who would benefit most from closer follow-up: adults who were hospitalized or incarcerated during their first episode and people whose initial disease was pulmonary, plus children living with HIV. Because the study used routine surveillance data from a large population over many years, it offers strong evidence that health systems should adopt risk-based follow-up strategies rather than a uniform, brief check at treatment end. In short, to reduce repeat TB and advance elimination efforts, programs need to monitor survivors over time and prioritize those with the highest demonstrated risk.