---
title: "Impact of loss to follow-up on 5-year mortality after tuberculosis treatment in São Paulo, Brazil"
id: "plos-medicine-0-loss-to-follow-up-during-tuberculosis-treatment-and-5-year-mortality-in-brazil"
canonical_url: "https://medichelpline.com/clinical-feed/plos-medicine-0-loss-to-follow-up-during-tuberculosis-treatment-and-5-year-mortality-in-brazil"
content_type: "clinical_feed_article"
specialty: "Infectious Disease"
source_name: "PLOS Medicine"
source_url: "https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303"
published_at: "2026-11-18T14:00:00.000Z"
evidence_level: "Journal Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Impact of loss to follow-up on 5-year mortality after tuberculosis treatment in São Paulo, Brazil
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/plos-medicine-0-loss-to-follow-up-during-tuberculosis-treatment-and-5-year-mortality-in-brazil
- **Specialty:** [Infectious Disease](https://medichelpline.com/clinical-feed/infectious-disease.md)
- **Primary Source:** PLOS Medicine
- **Source URL:** [Original Journal Publication](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303)
- **Published At:** 2026-11-18T14:00:00.000Z
- **Evidence Rating:** Journal Feed
## Executive GIST (TL;DR)
- Cohort study of 171,048 people aged ≥15 who started first-line **tuberculosis** treatment in São Paulo State (2013–2023), using linked surveillance and mortality registries. - **Loss to follow-up** (LTFU) was defined as interruption of treatment for ≥30 consecutive days; 20,830 (12.2%) met this definition. - Investigators used a clone-censor-weight target trial emulation to compare two strategies over 60 months from treatment start: disengage during the first 6 months versus remain in care. - Analyses adjusted for age, sex, race, education, **HIV** status, diabetes, alcohol use, drug use, incarceration, homelessness, hospitalization at diagnosis, clinical form, **directly observed therapy** (DOT), and municipality. - LTFU increased standardized 5-year mortality by 2.49 percentage points (95% CI 1.94–3.17); tuberculosis was the attributed cause for 61% of that excess. - Relative effects were larger in younger, stably housed people because baseline mortality was lower; the largest absolute excess was among people living with **HIV** (5.16 percentage points, 95% CI 3.76–6.60). - Among those who became LTFU, 65% had received DOT before disengagement. - A rolling landmark analysis found similar mortality increases for disengagement occurring during any of the first 4 months of treatment, with attenuation thereafter. - The authors noted limitations: registry lacked markers of disease severity (radiographic extent, bacillary load), residual confounding is possible, and some patients were excluded for incomplete records or transfers. - The study team made analysis code and a minimal deidentified dataset publicly available; underlying identifiable records require formal approval for access.
## Clinical Analysis & Structured Key Points
[ Skip to main content ](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#main-content) Advertisement * [plos.org](https://plos.org/) * [Create account](https://community.plos.org/registration/new) * [Sign in](https://journals.plos.org/user/secure/login?page=%2Fplosmedicine%2Farticle%3Fid%3D10.1371%2Fjournal.pmed.1005303) * * About * Browse * Publish * [](https://journals.plos.org/plosmedicine/ "PLOS Medicine") * Search [advanced search](https://journals.plos.org/plosmedicine/search) * 0 [Save](https://journals.plos.org/plosmedicine/article/metrics?id=10.1371/journal.pmed.1005303#savedHeader) [Total Mendeley and Citeulike bookmarks.](https://journals.plos.org/plosmedicine/article/metrics?id=10.1371/journal.pmed.1005303#savedHeader) * 0 [Citation](https://journals.plos.org/plosmedicine/article/metrics?id=10.1371/journal.pmed.1005303#citedHeader) [Paper's citation count computed by Dimensions.](https://journals.plos.org/plosmedicine/article/metrics?id=10.1371/journal.pmed.1005303#citedHeader) * 44 [View](https://journals.plos.org/plosmedicine/article/metrics?id=10.1371/journal.pmed.1005303#viewedHeader) [PLOS views and downloads.](https://journals.plos.org/plosmedicine/article/metrics?id=10.1371/journal.pmed.1005303#viewedHeader) * 0 [Share](https://journals.plos.org/plosmedicine/article/metrics?id=10.1371/journal.pmed.1005303#discussedHeader) [Sum of Facebook, Twitter, Reddit and Wikipedia activity.](https://journals.plos.org/plosmedicine/article/metrics?id=10.1371/journal.pmed.1005303#discussedHeader) Open Access Peer-reviewed Research Article # Loss to follow-up during tuberculosis treatment and 5-year mortality in Brazil: A target trial emulation * Evelyn Lepka de Lima , Roles Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Software, Writing – original draft * E-mail: evelepka@bu.edu Affiliations Department of Epidemiology, Boston University School of Public Health, Boston, Massachusetts, United States of America, Tuberculosis Control Program, Municipal Health Secretariat of Guarujá, Guarujá, São Paulo, Brazil [ ![ORCID logo](https://journals.plos.org/resource/img/orcid_16x16.png) https://orcid.org/0000-0003-4236-8295 ](https://orcid.org/0000-0003-4236-8295 "ORCID Registry") ⨯ * Ana Angélica Lindoso, Roles Conceptualization, Data curation Affiliations Emilio Ribas Institute of Infectious Diseases, Epidemiology Surveillance, São Paulo, Brazil, São Paulo State Department of Health, Epidemiologic Surveillance Center, São Paulo, Brazil [ ![ORCID logo](https://journals.plos.org/resource/img/orcid_16x16.png) https://orcid.org/0000-0001-7030-6134 ](https://orcid.org/0000-0001-7030-6134 "ORCID Registry") ⨯ * Giovanna Mariah Orlandi, Roles Data curation Affiliation São Paulo State Department of Health, Epidemiologic Surveillance Center, São Paulo, Brazil [ ![ORCID logo](https://journals.plos.org/resource/img/orcid_16x16.png) https://orcid.org/0000-0002-1451-7495 ](https://orcid.org/0000-0002-1451-7495 "ORCID Registry") ⨯ * Suely Fukasava, Roles Data curation Affiliation São Paulo State Department of Health, Epidemiologic Surveillance Center, São Paulo, Brazil [ ![ORCID logo](https://journals.plos.org/resource/img/orcid_16x16.png) https://orcid.org/0009-0003-4601-8329 ](https://orcid.org/0009-0003-4601-8329 "ORCID Registry") ⨯ * Catia Martinez, Roles Data curation Affiliation São Paulo State Department of Health, Center for Disease Control, São Paulo, Brazil ⨯ * Julio Croda, Roles Visualization, Writing – review & editing Affiliations School of Medicine, Federal University of Mato Grosso do Sul, Campo Grande, Mato Grosso do Sul, Brazil, Oswaldo Cruz Foundation, Campo Grande, Mato Grosso do Sul, Brazil, Department of Epidemiology of Microbial Diseases, Yale University School of Public Health, New Haven, Connecticut, United States of America [ ![ORCID logo](https://journals.plos.org/resource/img/orcid_16x16.png) https://orcid.org/0000-0002-6665-6825 ](https://orcid.org/0000-0002-6665-6825 "ORCID Registry") ⨯ * C. Robert Horsburgh Jr, Roles Formal analysis, Methodology, Supervision, Validation, Writing – review & editing Affiliations Department of Epidemiology, Boston University School of Public Health, Boston, Massachusetts, United States of America, Department of Global Health, Boston University School of Public Health, Boston, Massachusetts, United States of America [ ![ORCID logo](https://journals.plos.org/resource/img/orcid_16x16.png) https://orcid.org/0000-0001-6838-7895 ](https://orcid.org/0000-0001-6838-7895 "ORCID Registry") ⨯ * Otavio Ranzani, Roles Formal analysis, Methodology, Writing – review & editing Affiliation DataHealth Lab, Institut de Recerca Sant Pau (IR SANT PAU), Barcelona, Spain [ ![ORCID logo](https://journals.plos.org/resource/img/orcid_16x16.png) https://orcid.org/0000-0002-4677-6862 ](https://orcid.org/0000-0002-4677-6862 "ORCID Registry") ⨯ * Meredith B. Brooks, Roles Formal analysis, Methodology, Supervision, Visualization, Writing – review & editing Affiliation Department of Global Health, Boston University School of Public Health, Boston, Massachusetts, United States of America [ ![ORCID logo](https://journals.plos.org/resource/img/orcid_16x16.png) https://orcid.org/0000-0001-8431-5517 ](https://orcid.org/0000-0001-8431-5517 "ORCID Registry") ⨯ * Jason R. Andrews Roles Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Software, Supervision, Writing – original draft, Writing – review & editing Affiliation Division of Infectious Diseases and Geographic Medicine, Stanford University School of Medicine, Stanford, California, United States of America [ ![ORCID logo](https://journals.plos.org/resource/img/orcid_16x16.png) https://orcid.org/0000-0002-5967-251X ](https://orcid.org/0000-0002-5967-251X "ORCID Registry") ⨯ # Loss to follow-up during tuberculosis treatment and 5-year mortality in Brazil: A target trial emulation * Evelyn Lepka de Lima, * Ana Angélica Lindoso, * Giovanna Mariah Orlandi, * Suely Fukasava, * Catia Martinez, * Julio Croda, … * C. Robert Horsburgh Jr, * Otavio Ranzani, * Meredith B. Brooks, * Jason R. Andrews ![PLOS](https://journals.plos.org/resource/img/logo-plos-full-color.svg) x * Published: November 18, 2026 * * [Article](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303) * [Authors](https://journals.plos.org/plosmedicine/article/authors?id=10.1371/journal.pmed.1005303) * [Metrics](https://journals.plos.org/plosmedicine/article/metrics?id=10.1371/journal.pmed.1005303) * [Comments](https://journals.plos.org/plosmedicine/article/comments?id=10.1371/journal.pmed.1005303) * [Media Coverage](http://plos.altmetric.com/details/doi/10.1371/journal.pmed.1005303) * [Peer Review](https://journals.plos.org/plosmedicine/article/peerReview?id=10.1371/journal.pmed.1005303) * [Abstract](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#abstract0) * [Author summary](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#abstract1) * [Introduction](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#sec004) * [Methods](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#sec005) * [Results](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#sec019) * [Discussion](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#sec025) * [Supporting information](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#sec026) * [Acknowledgments](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#ack) * [References](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#references) * [Reader Comments](https://journals.plos.org/plosmedicine/article/comments?id=10.1371/journal.pmed.1005303) * [Figures](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303) [?](https://journals.plos.org/plosmedicine/s/accepted-manuscripts#loc-early-version) ## This is an uncorrected proof. ## Abstract ### Background Loss to follow-up (LTFU) during tuberculosis treatment is a major programmatic gap, but how much excess mortality it causes has been difficult to estimate, because individuals who disengage differ from those who do not, and they must survive long enough to be recorded as having disengaged. Quantifying this excess and identifying which populations are at greatest risk are critical to informing program design and resource allocation. ### Methods and findings We followed 171,048 people aged 15 years or older starting a first course of tuberculosis treatment in São Paulo State, Brazil, between 2013 and 2023, linking surveillance and mortality registries; 20,830 (12.2%) were LTFU, defined as interruption of treatment for at least 30 consecutive days. Among those LTFU, 65% had received directly observed therapy (DOT). Using a clone-censor-weight target trial emulation, we contrasted two strategies over 60 months from the first day of treatment: disengagement at any point during the first 6 months and remaining in care throughout. Analyses were adjusted for age, sex, race, education, HIV status, diabetes, alcohol use, drug use, incarceration, homelessness, hospitalization at diagnosis, clinical form of tuberculosis, DOT, and municipality. A rolling landmark analysis addressed the timing of disengagement and cause-specific mortality excess. LTFU increased standardized 5-year mortality by 2.49 percentage points (95% CI [1.94, 3.17]). In a rolling landmark analysis, tuberculosis accounted for 61% of the 5-year excess mortality. Relative effects were largest in younger, stably housed individuals, whose baseline mortality was low; the largest absolute excess fell on people living with HIV (5.16 percentage points, 95% CI [3.76, 6.60]). The excess was similar for disengagement at any point in the first 4 months of treatment and attenuated thereafter. Limitations include the absence from the registry of severity markers such as radiographic extent and bacillary load, so residual confounding cannot be excluded. ### Conclusions Despite high DOT coverage, disengagement during the first 6 months of tuberculosis treatment was common and increased 5-year mortality under the stated identification assumptions, with the excess concentrated in deaths attributed to tuberculosis. Interventions that provide support throughout treatment should be tested for their effectiveness in sustaining engagement and reducing mortality. ## Author summary ### Why was this study done? * People who interrupt tuberculosis treatment have a higher risk of death, but it remains unclear how much of this risk is attributable to treatment interruption itself. * Estimating this impact is challenging because people who interrupt treatment differ from those who remain in care, and the way these groups are compared can lead to misleading results. * Understanding how treatment interruption affects deaths from tuberculosis and other causes, including years after treatment begins, is important for improving tuberculosis care. ### What did the researchers do and find? * We studied 171,048 people starting tuberculosis treatment in São Paulo State, Brazil, between 2013 and 2023, using health records to examine mortality over 5 years. * We estimated that interrupting treatment increased the 5-year risk of death by ~25 additional deaths per 1,000 people, with tuberculosis itself the leading cause of these additional deaths. * The increase in mortality was similar when treatment was interrupted during any of the first 4 months after starting treatment, and people living with HIV experienced the largest absolute increase in risk. ### What do these findings mean? * Interrupting tuberculosis treatment may contribute to deaths occurring years after treatment begins, making sustained engagement in care important for long-term survival. * Although most people who interrupted treatment had received directly observed therapy, our findings suggest that tuberculosis programs need approaches that go beyond observing medication intake to help patients complete treatment. * Because the study used routinely collected health records, we could not fully account for differences in disease severity or changes in patients’ health during treatment, and some patients were excluded because of incomplete records or transfers. ## Figures ![Fig 6](https://journals.plos.org/plosmedicine/article/figure/image?size=inline&id=10.1371/journal.pmed.1005303.g006) ![Fig 1](https://journals.plos.org/plosmedicine/article/figure/image?size=inline&id=10.1371/journal.pmed.1005303.g001) ![Fig 2](https://journals.plos.org/plosmedicine/article/figure/image?size=inline&id=10.1371/journal.pmed.1005303.g002) ![Table 1](https://journals.plos.org/plosmedicine/article/figure/image?size=inline&id=10.1371/journal.pmed.1005303.t001) ![Fig 3](https://journals.plos.org/plosmedicine/article/figure/image?size=inline&id=10.1371/journal.pmed.1005303.g003) ![Fig 4](https://journals.plos.org/plosmedicine/article/figure/image?size=inline&id=10.1371/journal.pmed.1005303.g004) ![Fig 5](https://journals.plos.org/plosmedicine/article/figure/image?size=inline&id=10.1371/journal.pmed.1005303.g005) ![Fig 6](https://journals.plos.org/plosmedicine/article/figure/image?size=inline&id=10.1371/journal.pmed.1005303.g006) ![Fig 1](https://journals.plos.org/plosmedicine/article/figure/image?size=inline&id=10.1371/journal.pmed.1005303.g001) ![Fig 2](https://journals.plos.org/plosmedicine/article/figure/image?size=inline&id=10.1371/journal.pmed.1005303.g002) ![Table 1](https://journals.plos.org/plosmedicine/article/figure/image?size=inline&id=10.1371/journal.pmed.1005303.t001) **Citation:** Lepka de Lima E, Lindoso AA, Orlandi GM, Fukasava S, Martinez C, Croda J, et al. (2026) Loss to follow-up during tuberculosis treatment and 5-year mortality in Brazil: A target trial emulation. PLoS Med 23(11): e1005303. https://doi.org/10.1371/journal.pmed.1005303 **Academic Editor:** Alice Zwerling, University of Ottawa, CANADA **Received:** June 24, 2026; **Accepted:** November 11, 2026; **Published:** November 18, 2026 **Copyright:** © 2026 Lepka de Lima et al. This is an open access article distributed under the terms of the [Creative Commons Attribution License](http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. **Data Availability:** The minimal de-identified dataset underlying the findings of this study is publicly available from Zenodo at (and from the source repository at ). It contains the 171,048 individuals and 32 analysis variables used in the analyses, with the personal identifier replaced by a sequential anonymous code, all exact dates removed, ICD-10 cause-of-death codes reduced to a high-level classification, and no geographic identifier retained. The underlying individual-level surveillance and mortality records cannot be redistributed, because they contain sensitive personal health information and the authorization granted for their secondary use by the São Paulo State Health Department does not permit it. Researchers who meet the criteria for access to the confidential records may request them, on the basis of a research protocol, from the Tuberculosis Division of the Centro de Vigilância Epidemiológica “Prof. Alexandre Vranjac” (Epidemiological Surveillance Center “Prof. Alexandre Vranjac”), São Paulo State Health Department (dvtbc@saude.sp.gov.br), subject to approval by the Research Ethics Committee of the Instituto de Infectologia Emílio Ribas (comiteetica@emilioribas.sp.gov.br), which is independent of the authors. The authors did not have any special access privileges that other researchers would not have. The analysis code is publicly available at and archived at . **Funding:** The author(s) received no specific funding for this work. **Competing interests:** The authors have declared that no competing interests exist. **Abbreviations:** DOT, directly observed therapy; DST, Drug-susceptibility testing; IQR, interquartile range; LTFU, Loss to follow-up. ## Introduction Tuberculosis remains a major cause of infectious disease mortality globally, accounting for an estimated 1.23 million deaths in 2024 [[1](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#pmed.1005303.ref001)]. Achieving progress in reducing this burden depends critically on maintaining high rates of completion of curative therapy. Despite the availability of effective drug regimens for decades, many high-burden countries have struggled to translate the high cure rates observed in clinical trials and controlled settings into everyday practice [[2](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#pmed.1005303.ref002)]. Mortality trends in WHO Region of the Americas further highlight this challenge, with limited progress overall and increasing mortality in several countries, including Brazil, since 2015 [[1](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#pmed.1005303.ref001)]. Loss to follow-up (LTFU) over the standard 6 months of tuberculosis treatment represents an important programmatic gap in tuberculosis care and a major contributor to adverse outcomes. When LTFU occurs, consequences include _Mycobacterium tuberculosis_ drug resistance, ongoing transmission, and increased mortality [[1](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#pmed.1005303.ref001)]. However, quantifying these effects presents several challenges. LTFU is not a random event, and a range of demographic and clinical characteristics have been associated with increased risk, including male sex, younger age, lower educational attainment, HIV co-infection, prior tuberculosis, and substance use [[1](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#pmed.1005303.ref001),[2](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#pmed.1005303.ref002)]. Estimating the impact of LTFU on mortality is therefore complicated by confounding, as factors influencing disengagement from care also affect subsequent outcomes [[3](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#pmed.1005303.ref003)]. These effects can act in multiple directions: individuals with greater social vulnerabilities may be more likely to interrupt treatment and have poor outcomes, while individuals with more severe disease, who are at greatest risk of death, might be less likely to discontinue therapy than those with less advanced disease and few symptoms. Comparisons between individuals who interrupt treatment and those who complete it are also subject to several distinct forms of immortal-time bias [[4](https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005303#pmed.1005303.ref004)]. When exposure is assigned at treatment initiation according to how treatment later ended, individuals are classified as LTFU on the basis of an event that occurs only after part of their treatment has been survived: anyone who dies before disengaging is recorded as an on-treatment death and can only appear in the comparison group. The time from treatment initiation to disengagement, which may occur at any point during treatment, is therefore immortal-time for the LTFU group, and counting it as expo
## Related Clinical Research

- [AI detection of pulmonary tuberculosis on chest X‑ray: meta-analysis of clinical cohort accuracy](https://medichelpline.com/clinical-feed/medrxiv-12-artificial-intelligence-for-pulmonary-tuberculosis-detection-on-chest-x-ray.md)
- [Phage therapy funding 1999–2023: rapid growth from a narrow funder base](https://medichelpline.com/clinical-feed/medrxiv-2-twenty-five-years-of-phage-therapy-funding-rapid-growth-narrow-base.md)
- [Self-reported tuberculosis prevalence and vulnerability factors in six rural Indian districts](https://medichelpline.com/clinical-feed/plos-one-16-self-reported-prevalence-of-tuberculosis-and-selected-associated-vulnerability.md)
- [Temporal patterns of tuberculosis-associated hyperglycaemia and effects on treatment outcomes in n](https://medichelpline.com/clinical-feed/plos-one-0-temporal-patterns-of-tuberculosis-associated-hyperglycaemia-and-its-effect-on.md)
- [Parenchymal Neuro-Behçet Disease Case Report: Heterogeneous Clinical Presentations](https://medichelpline.com/clinical-feed/frontiers-in-immunology-16-parenchymal-neuro-beh-et-disease-with-heterogeneous-clinical-presentations-a.md)

## Navigation
- [← Back to Infectious Disease Feed](https://medichelpline.com/clinical-feed/infectious-disease.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.