Evidence analysisNovember 19, 2026Evidence

Loss to follow-up during tuberculosis treatment and 5-year mortality in Brazil: A target trial emulation

FreedomWitness Evidence Engine: This is an AI-assisted original analysis built from the linked source and available public source context. It separates evidence from interpretation and may challenge conventional framing only where the evidence supports it. It is not a copied article.

Loss to follow-up during tuberculosis treatment and 5-year mortality in Brazil: evidence dossier

What is established

Tuberculosis (TB) remains a leading cause of infectious disease mortality worldwide, with over a million deaths annually. Effective drug regimens exist and have been available for decades, but maintaining patient adherence during the standard six-month treatment remains challenging. Loss to follow-up (LTFU) — when patients interrupt or discontinue treatment — is linked to poor outcomes including drug resistance, ongoing transmission, and increased mortality. Certain demographic and clinical factors such as younger age, male sex, HIV co-infection, and substance use influence the likelihood of LTFU. However, quantifying exactly how much LTFU contributes to excess mortality is complex because those who disengage differ from those who remain in care, and the timing and reasons for disengagement vary.

What the source claims

A recent study published in PLOS Medicine analyzed a large cohort of 171,048 people aged 15 years or older initiating a first TB treatment course in São Paulo State, Brazil, between 2013 and 2023. Using linked health surveillance and mortality registries combined with advanced statistical methods emulating a target trial, the study compared patients who disengaged during the first six months of treatment to those who completed treatment.

Key findings include:

  • 12.2% (20,830) of patients experienced LTFU, defined as ≥30 consecutive days without treatment.
  • Patients who disengaged from care showed an increased 5-year mortality rate by approximately 2.5 percentage points compared to those who completed treatment.
  • About 61% of this excess mortality was directly attributed to tuberculosis as the cause of death.
  • The relative increase in mortality was largest in younger, stably housed individuals, though the largest absolute excess risk occurred among people living with HIV (around 5 percentage points).
  • The timing of disengagement mattered; excess mortality was similar for loss occurring anytime during the first four months of treatment and diminished afterward.
  • Most patients who discontinued treatment had received directly observed therapy (DOT), indicating that DOT alone may not be sufficient to ensure treatment completion.

The study also adjusted for multiple potential confounders, including demographics, comorbid conditions, and clinical presentation, enhancing the reliability of the findings under the stated assumptions.

What may be missing from the usual framing

While the study robustly quantifies excess mortality linked to LTFU, some aspects are not fully addressed due to data limitations:

  • Markers of TB severity—such as radiographic extent or bacterial load—were not available in the registry data, leaving residual confounding possible.
  • Dynamic changes during treatment, such as improvements or worsening health status, were not incorporated, potentially affecting risk estimates.
  • Patient-level social determinants beyond basic demographics (e.g., socioeconomic status, mental health, stigma) that influence adherence and mortality might be incompletely captured.
  • Intervention types and quality beyond DOT, including psychosocial support or community health initiatives, were not evaluated but may modulate outcomes.

These gaps suggest the usual emphasis on DOT as a solitary adherence strategy may overlook broader systemic and individual factors critical to treatment success.

Competing evidence or uncertainty

The study’s findings align broadly with existing knowledge that treatment interruption increases mortality risk in TB patients. However, some uncertainties remain:

  • Because observational data and emulation techniques were used, although adjustments were thorough, the possibility of unmeasured confounding affecting causal interpretation cannot be excluded.
  • The generalizability outside São Paulo State or Brazil might be limited due to differences in health systems, TB epidemiology, and socioeconomic conditions.
  • Other research suggests varied effectiveness of DOT, especially when not coupled with comprehensive patient support, highlighting the need for integrated approaches.
  • Some studies point to the importance of early diagnosis severity and resistance profiles in mortality, factors not adjusted for here due to data constraints.

Overall, while this evidence strengthens the causal link between LTFU and mortality, interpretation nuances and contextual differences call for cautious extrapolation.

Why it matters

Understanding the magnitude and drivers of excess mortality linked to TB treatment interruption is vital for public health policies and program design. This study highlights:

  • LTFU during TB treatment significantly increases the risk of death up to five years later, mainly due to TB itself.
  • High coverage of DOT, a widely promoted intervention, does not fully prevent loss to follow-up or its lethal consequences.
  • Specific vulnerable populations, including younger people and those living with HIV, face disproportionate mortality risks if they disengage.
  • The timing of treatment disengagement is critical, especially during the first four months of therapy, suggesting a window for intensified retention efforts.

These insights underscore the need for comprehensive, patient-centered strategies that extend beyond medication supervision to address social, clinical, and structural barriers to sustained treatment adherence and survival.

What to verify next

Future investigations and verifications should consider:

  • Independent replication of these findings in diverse geographic settings with different healthcare infrastructures.
  • Inclusion of clinical severity markers, drug resistance data, and more detailed social determinants of health to refine risk attribution.
  • Evaluation of intervention packages combining DOT with psychosocial, economic, and community-based supports for reducing LTFU and mortality.
  • Longitudinal monitoring of patients lost to follow-up to understand reasons for disengagement and patterns of re-engagement or mortality causes.
  • Impact assessment of tailored strategies for high-risk subgroups such as people living with HIV or those with substance use disorders.

Readers and stakeholders should consult the original PLOS Medicine article to review the detailed methodology, statistical analyses, and contextual discussion supporting these conclusions.

Source & verification

Primary source: PLOS Medicine · November 18, 2026

Read the original source

FreedomWitness links to the original source so readers can inspect the underlying material and judge the context for themselves.

Continue reading

Disagreement doesn’t have to be toxic

Evidence Dossier: The Nature and Practice of Disagreement What is established Disagreement is a common and unavoidable part of social life, extending beyond politics…