Record linkage in tuberculosis epidemiological surveillance in Brazil: systematic methodological literature review

ABSTRACT

Objective:  To describe the literature on the methodological characteristics and use of record linkage techniques applied to tuberculosis surveillance in Brazil.

Methods:  A systematic methodological review (PROSPERO: CRD42024623969) was conducted following the 2020 guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). Searches were performed on the following databases: PubMed, Embase, Scopus, Virtual Health Library (VHL), Scientific Electronic Library Online (Scielo), and Latin American and Caribbean Literature on Health Sciences (LILACS) on February 5, 2025. Scientific articles produced in Brazil and whose authors applied record linkage between health information systems for epidemiological surveillance purposes were included. Grey literature, studies without a focus on tuberculosis, studies without application of linkage, and studies conducted outside Brazil were excluded. Descriptive synthesis was used, reporting frequencies, proportions, and measures of central tendency. Risk of bias or methodological quality were not assessed due to the heterogeneity of study designs and objectives, and because this review aimed to describe the methodological characteristics of the studies rather than synthesize effect estimates or epidemiological associations.

Results:  The search retrieved 1,549 records; 49 studies were included. Most studies used the tuberculosis module of the Notifiable Diseases Information System (SINAN-TB, 98.0%) and the Mortality Information System (SIM, 67.3%); the most frequently used software was OpenRecLink (57.1%). The most common key variables were "name," "date of birth," and "mother's name." The linkage percentage ranged from 0.1 to 97.3%. The most frequent objectives were the description of epidemiological scenarios (24.5%), analysis of associated factors (20.4%), and investigation of underreporting (16.3%). Cohort (49.0%) and descriptive (34.7%) study designs predominated.

Conclusion:  Record linkage integrates complementary databases and may strengthen tuberculosis surveillance. The findings suggest the potential usefulness of record linkage to support epidemiological analyses, improve records, and inform surveillance actions.

Keywords:
Record linkage; Health Information Systems; Tuberculosis; Public Health Surveillance; Systematic Review

RESUMO

Objetivo:  Descrever a literatura sobre as características metodológicas e o uso de técnicas de vinculação de dados aplicadas à vigilância da tuberculose no Brasil.

Métodos:  Revisão metodológica sistemática (PROSPERO: CRD42024623969) conduzida segundo as diretrizes Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020. As buscas foram realizadas nas bases de dados PubMed, Embase, Scopus, Biblioteca Virtual em Saúde (BVS), Scientific Electronic Library Online (SciELO) e Literatura Latino-Americana e do Caribe em Ciências da Saúde (LILACS) em 5 de fevereiro de 2025. Incluíram-se artigos científicos conduzidos no Brasil que aplicaram vinculação de dados entre sistemas de informação em saúde para fins de vigilância epidemiológica. Excluíram-se literatura cinzenta e estudos sem foco em tuberculose, sem aplicação de vinculação e realizados fora do Brasil. A síntese foi descritiva, reportando frequências, proporções e medidas de tendência central. Não foi realizada avaliação de risco de viés ou qualidade metodológica, devido à heterogeneidade dos delineamentos e objetivos, também porque o objetivo desta revisão foi descrever características metodológicas dos estudos e não sintetizar estimativas de efeito ou associações epidemiológicas.

Resultados:  A busca retornou 1.549 registros; 49 estudos foram incluídos. A maioria utilizou o Sistema de Informação de Agravos de Notificação (Sinan-TB) (98%) e Sistema de Informação sobre Mortalidade (SIM) (67,3%); o software mais empregado foi o OpenRecLink (57,1%). As variáveis-chave mais comuns foram "nome", "data de nascimento" e "nome da mãe". O percentual de vinculação variou de 0,1% a 97,3%. Os objetivos mais frequentes foram a descrição de cenários epidemiológicos (24,5%), análise de fatores associados (20,4%) e investigação de subnotificação (16,3%). Predominaram delineamentos de coorte (49%) e descritivos (34,7%).

Conclusões:  A vinculação de dados integra bases complementares e pode fortalecer a vigilância da tuberculose. Os achados sugerem potencial utilidade da vinculação de dados para apoiar análises epidemiológicas, qualificar registros e subsidiar ações de vigilância.

Palavras-chave:
Record linkage; Sistemas de Informação em Saúde; Tuberculose; Vigilância em Saúde Pública; Revisão Sistemática

INTRODUCTION

Tuberculosis is one of the main infectious diseases in the world and in Brazil, with a high incidence especially in regions with severe socioeconomic inequalities¹. Brazil is among the nations with the highest tuberculosis burden in the world¹, with incidence rates above 37 cases every 100 thousand inhabitants and over 80 thousand new cases of tuberculosis per year since the new coronavirus (COVID-19) pandemic².

To eliminate tuberculosis, Brazil has adopted a series of measures provided for in the EndTB Strategy, which aims to eliminate the disease in the country by 2035. This strategy is aligned with the Sustainable Development Goals (SDGs), which, in its Target 3.3, aims at ending the epidemics of infectious and parasitic diseases, including tuberculosis, by 2030¹,³,⁴.

Among the various issues that influence the process of eliminating tuberculosis in Brazil, the challenges of using Brazilian Health Information Systems (HIS) can be highlighted, such as the Notifiable Diseases Information System (Sistema de Informação de Agravos de Notificação – SINAN) and the Mortality Information System (Sistema de Informação sobre Mortalidade – SIM), which do not have interoperability, making it difficult, for instance, to jointly monitor cases of tuberculosis and deaths⁵–⁷.

To overcome this technological limitation, record linkage techniques are used for relating the HIS. This process consists of identifying and combining records that correspond to the same individual in different databases, even when there is no common unique identifier, which could be the Taxpayer Registration Number (Cadastro de Pessoa Física – CPF) or the National Health Card (SUS [Brazilian Unified Health System] Card)⁷.

Within the context of tuberculosis epidemiological surveillance, linkage allows a more complete view of the clinical history of patients and the outcomes⁸,⁹. In addition, these techniques help detect and eliminate duplicate records (deduplication technique) by consolidating non-repeated notifications¹⁰,¹¹. Overall, the process involves previous steps of data preparation, such as cleaning, standardization and, in some studies, blocking, with the aim of making linkage more feasible and efficient¹².

Considering that the use of record linkage techniques is widespread, especially in the context of tuberculosis surveillance, in this systematic review, we aim to describe the methodological characteristics and the use of linkage techniques applied to studies on tuberculosis in Brazil—and, based on these findings, to discuss how these techniques are used and can be used for supporting the epidemiological surveillance of tuberculosis.

METHODS

This systematic methodological review was registered on the International Prospective Register of Systematic Reviews (PROSPERO) platform under identifier CRD42024623969 and developed according to the 2020 guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).

Context

The study is inserted in the Brazilian context according to the following research question: "How have record linkage techniques been used in epidemiological studies on tuberculosis in Brazil?", mainly focusing on the application of record linkage techniques from HIS used by epidemiological surveillance of tuberculosis in the country.

Study design

This is a systematic methodological literature review seeking to describe methodological characteristics of studies that applied record linkage techniques for tuberculosis surveillance in Brazil. Moreover, we sought to evaluate the contexts for the application of these techniques, the challenges faced, and the opportunities resulting from their use in the HIS.

Search strategies and databases

The final search was carried out on February 5, 2025 in six electronic databases: PubMed, Embase, Scopus, Virtual Health Library (VHL), Scientific Electronic Library Online (Scielo), and Latin American and Caribbean Literature on Health Sciences (LILACS). The strategies were adapted for each database and controlled descriptors and free terms related to record linkage, tuberculosis, and Brazil were combined with explicit use of Boolean operators. In Table 1, the terms used and their combination structure are presented according to database.

Table 1
Search strategies used.

Database	Strategy
PubMed	(("linkage" OR "linked" OR "Medical Record linkage "[Mesh] OR "Medical Record linkage " OR "Health Record linkage " OR "Record linkage " OR "Data linkage " OR "Centre for Data and Knowledge Integration for Health" OR "CIDACS") AND (Brazil OR Brazilian) AND (Tuberculosis))
Embase	‘tuberculosis’ AND [embase]/lim AND (‘brazil’ OR ‘brazilian’) AND [embase]/lim AND (linkage OR ‘linked’ OR ‘medical Record linkage /exp OR ‘medical Record linkage OR ‘health Record linkage OR ‘Record linkage OR ‘data linkage OR ‘centre for data and knowledge integration for health’ OR ‘cidacs’) AND [embase]/lim AND ‘article’/it
Scopus	ALL("linkage " OR "linked" OR "medical Record linkage " OR "health Record linkage " OR "Record linkage " OR "data linkage " OR "centre for data and knowledge integration for health" OR "CIDACS") AND TITLE-ABS-KEY(tuberculosis) AND AFFILCOUNTRY("Brazil") AND DOCTYPE("ar") AND NOT (SUBJAREA("IMMU") OR SUBJAREA("BIOC") OR SUBJAREA("AGRI"))
VHL	("vinculação" OR "vinculado" OR "vinculação de registros médicos" OR "vinculação de dados" OR "ligação de registros médicos" OR "CIDACS" OR "vinculación" OR "vinculado" OR "vinculación de registros médicos" OR "vinculación de datos" OR "enlace de registros médicos" OR " linkage " OR "linked" OR "medical Record linkage " OR "health Record linkage " OR "Record linkage " OR "data linkage " OR "Centre for Data and Knowledge Integration for Health") AND ("Brasil" OR "brasileiro" OR "brasileira" OR "brasileiros" OR "brasileiras" OR "Brasil" OR "brasileño" OR "brasileña" OR "brasileños" OR "brasileñas" OR "Brazil" OR "Brazilian") AND ("tuberculose" OR "tuberculosis")
SciELO	("vinculação" OR "vinculado" OR "vinculação de registros médicos" OR "vinculação de dados" OR "ligação de registros médicos" OR "CIDACS" OR "vinculación" OR "vinculado" OR "vinculación de registros médicos" OR "vinculación de datos" OR "enlace de registros médicos" OR " linkage " OR "linked" OR "medical Record linkage " OR "health Record linkage " OR "Record linkage " OR "data linkage " OR "Centre for Data and Knowledge Integration for Health") AND ("Brasil" OR "brasileiro" OR "brasileira" OR "brasileiros" OR "brasileiras" OR "Brasil" OR "brasileño" OR "brasileña" OR "brasileños" OR "brasileñas" OR "Brazil" OR "Brazilian") AND ("tuberculose" OR "tuberculosis" OR "tuberculose" OR "tuberculose")
LILACS	("vinculação" OR "vinculado" OR "vinculação de registros médicos" OR "vinculação de dados" OR "ligação de registros médicos" OR "CIDACS" OR "vinculación" OR "vinculado" OR "vinculación de registros médicos" OR "vinculación de datos" OR "enlace de registros médicos" OR " linkage " OR "linked" OR "medical Record linkage " OR "health Record linkage " OR "Record linkage " OR "data linkage " OR "Centre for Data and Knowledge Integration for Health") AND ("Brasil" OR "brasileiro" OR "brasileira" OR "brasileiros" OR "brasileiras" OR "Brasil" OR "brasileño" OR "brasileña" OR "brasileños" OR "brasileñas" OR "Brazil" OR "Brazilian") AND ("tuberculose" OR "tuberculosis")

Study location and population

The study focuses on research carried out in the Brazilian context whose authors employed record linkage techniques applied to the epidemiological surveillance of tuberculosis, without restrictions concerning time, location, or population.

Inclusion and exclusion criteria

Scientific articles addressing the use of record linkage techniques in epidemiological investigations on tuberculosis in Brazil were included. Studies that used data from Brazilian health information systems and implemented record linkage procedures were considered eligible.

In turn, studies whose main focus was not tuberculosis; carried out outside Brazil; without direct application of record linkage for epidemiological surveillance purposes; or studies in a format other than scientific articles—such as conference abstracts, technical reports, and institutional documents—were excluded.

Selection process

The references of the six databases were exported to Rayyan, consolidated and deduplicated. Two researchers screened titles and abstracts according to predefined criteria. Potentially eligible records were read in full, with reapplication of the criteria and exclusion of remaining duplicates. The selection was independent between the two researchers.

In addition to the electronic search, a manual search was performed in the bibliography of the included studies, with the objective of identifying potentially eligible articles not retrieved in the initial strategies. The identified studies were evaluated according to the same eligibility criteria and, where relevant, incorporated into the final set.

Data extraction and analysis

Data extraction was performed independently by two researchers to ensure consistency and quality. The extracted variables were: title, main objective, year and place of publication, language, study period, type of linkage (probabilistic or deterministic), use of blocking, key variables of record linkage, involved HIS, and software used. For data tabulation and analysis, the Microsoft Excel software was used, while Python language¹³ was used for producing graphs and visual representations of the results. There was no formal evaluation of methodological quality or risk of bias due to the heterogeneity of the study designs, analytical objectives, and forms of reporting of the linkage procedures in the included studies.

Data Availability Statement:

All data used in this review were extracted from published scientific articles and are described in the methods section and Supplementary Table 1. No individual-level, confidential, or unpublished data were used.

RESULTS

The initial search resulted in 1,549 articles, of which 678 were identified as duplicates and excluded. After reading titles and abstracts, 777 records were excluded, with 94 remaining for reading in full. After evaluating the texts in full, 47 articles were included. Subsequently, two additional studies identified by manual search in the bibliography were incorporated, totaling 49 included studies (Figure 1).

Figure 1
Flowchart of the selection process of the articles included in the review.

The articles were published from 2004 to February 2025, the average of publications was 2.2 articles per year, and in the years 2006, 2008, and 2013 no eligible publications were identified. The journal with the highest number of publications on the subject was Cadernos de Saúde Pública [Reports in Public Health] (n=10; 20.4%), followed by Revista Brasileira de Epidemiologia [Brazilian Journal of Epidemiology] (n=4; 8.2%) and Revista de Saúde Pública [Journal of Public Health] (n=4; 8.2%) (Figure 2A). Most articles were published in national journals (n=29; 59.2%) (Figure 2B), but mostly in English (n=28; 57.1%) (Figure 2C).

Figure 2
Main characteristics of the included articles.

Of the 49 articles, 24 were cohort studies (49%), 17 were descriptive studies (34.7%), and five were cross-sectional studies (10.2%) (Figure 2D). Of the 49 articles, the most frequent objective was to describe epidemiological scenarios (n=12; 24.5%), followed by analyzing factors associated with tuberculosis-related outcomes (n=10; 20.4%), and analyzing the underreporting of tuberculosis in HIS (n=8; 16.3%) (Figure 2E). Furthermore, linkage was also used to analyze scenarios of co-infections and impacts of public programs and policies on the incidence of tuberculosis (Figure 2E).

The most used software was OpenRecLink (version 3) (n=28; 57.1%), followed by R (n=4; 8.2%), and Cidacs-RL, Linkplus, and Microsoft Excel (three articles for each software; 6.1%), while in eight articles (16.3%) the used software was not reported (Figure 2F). We observed that the use of RecLink remained constant over the years and that, as of 2019, there was a greater diversity of software, including R and the Cidacs-RL algorithms (Figure 2G).

The key variables most frequently used were the combination of "patient's name" (n=39 articles; 79.6%), "date of birth" (n=39 articles; 79.6%), and "mother's name" (n=34 articles; 69.4%), while in ten articles (20.4%) the key linkage variables used were not informed. Other articles, less frequently, also used "sex" and "address" information as key variables.

In addition, 21 articles (42.9%) had Brazil as a whole as the study location, and the second most frequent study location was the city of Rio de Janeiro, with 11 articles (22.4%). The analyzed study periods varied, presenting a median of six years (minimum: one year; maximum: 15 years) of study in each analysis (Figure 3).

Figure 3
Information on locations* and periods of the included studies.

Of the 49 studies, the use of the tuberculosis module of SINAN (SINAN-TB) comprised the majority, being used in 48 articles (98%). The second most used HIS was SIM, present in 33 articles (67.3%). Other databases were also greatly used, but on a smaller degree, such as the Single Registry for Federal Government Social Programs (Cadastro Único para Programas Sociais do Governo Federal – CadÚnico), the Benefit Management System of the Bolsa Família Program (conditional cash transfer program), the Information System of Special Treatments for Tuberculosis (Sistema de Informação de Tratamentos Especiais da Tuberculose – Site-TB), the Laboratory Environment Management System (Sistema Gerenciador de Ambiente Laboratorial – GAL), the Laboratory Test Control System of CD4/CD8 and Viral Load (Sistema de Controle de Exames Laboratoriais de CD4/CD8 e Carga Viral – Siscel), the Logistic Medicines Control System (Sistema de Controle Logístico de Medicamentos – Siclom), the AIDS module of SINAN (SINAN-AIDS), among others (Supplementary Table 1).

Eight articles (16.3%) did not report the results of record linkage, that is, how many pairs were found. Among those who reported such information with transparency (n=42; 85.7%), the amount of data used varied, ranging from 411 tuberculosis cases to 1,090,375 notifications. Other databases, such as CadÚnico, have used more than 130 million records in the linkage process. The number of records used in the studies ranged from 25 to 168,804, with a median of 1,008.5 (Q1 321.8; Q3 4,447.0) and a mean of 13,551.4 (Standard deviation [SD] 37,549.2); whereas the percentage of linked pairs ranged between 0.1 and 97.3%, with a median of 31% (Q1 12.5%; Q3 60.8%) and mean of 38.5% (SD 29.8). Manual review of the pairs was used in 31 articles (63.3%) (Supplementary Table 1).

DISCUSSION

In this review, we identified 49 articles that used record linkage techniques in studies on tuberculosis in Brazil. We predominantly found publications in national journals; studies of national scope or carried out in Rio de Janeiro; and analyses using SINAN-TB, SIM, and OpenRecLink. Together, the findings indicate that record linkage has been mainly used to describe epidemiological scenarios, investigate associated factors, and identify underreporting.

The application of record linkage in Brazil faces challenges due to the quality of HIS data (filling errors, lack of standardization, and incomplete fields)⁷. The approaches are divided into deterministic and probabilistic. Deterministic approaches require exact correspondence (for example, full name, date of birth, or CPF number) and are fast, with a low rate of false positives, but not flexible in the face of inconsistencies, frequent in Brazilian health systems¹⁴. In turn, the probabilistic approaches use similarity algorithms to combine attributes with small discrepancies, better dealing with variations in spelling and partial fields and increasing the sensitivity of record linkage; yet, the accuracy of both approaches can be compromised by low-quality data¹⁴,¹⁵.

Technical and operational challenges persist in the implementation of record linkage, especially in contexts with extensive databases, incomplete fields, and the need for manual review. These factors may influence the operational feasibility of record linkage and the consistency of the analyses carried out¹⁰,¹⁶,¹⁷.

The increase in specificity and sensitivity of record linkage processes allows health systems to identify tuberculosis transmission patterns and evaluate the effectiveness of public policies for disease control⁸,¹⁸–²⁰. The use of computational tools for record linkage, such as RecLink²¹, and programming languages, such as R²²,²³ and Python¹³,²⁴, requires technical training that is not always available in a homogeneous way among public health professionals.

RecLink was the most widely used software in this review, especially for its accessibility and user-friendly interface without programming experience. R- and Python-based solutions have also been mentioned in the included studies, especially in recent years, suggesting diversification of the employed tools²⁵,²⁶. Considering that, in this review, we did not formally compare software performance, these findings should be interpreted as a description of the observed panorama, and not as evidence of superiority of one tool over another.

Regarding the geographic scope, authors of many studies used national data, but no specific studies were identified for six states in the North, six in the Northeast, two in the Midwest, in addition to the states of Minas Gerais (Southeast), Rio Grande do Sul and Santa Catarina (South). Although record linkage has the potential to guide local decisions, its use remains concentrated at the national level or in the capitals¹⁰,¹¹,¹⁸–²⁰,²⁷–⁴¹.

The percentage of data linked between databases widely varied, from 0.1⁴² to 97.3%⁴³. Ten articles did not objectively inform this indicator¹⁷–¹⁹,³⁰,³⁴,³⁵,³⁸,⁴⁴–⁴⁶, which limits the evaluation of reproducibility and transparency of the employed procedures. Low percentages may reflect very extensive databases, more restrictive linkage criteria, or specific analytical objectives, while higher percentages tend to be found on databases with greater proximity of purpose or greater compatibility between identifiers⁹,²⁰,²⁹,³¹,⁴²,⁴⁷–⁴⁹. In theory, all deaths from tuberculosis in the SIM should correspond to records in SINAN; in practice, this is not confirmed, motivating the active search for deaths from tuberculosis not registered in SINAN³⁹,⁵⁰–⁵².

Furthermore, among the three states with the highest occurrence of new cases of tuberculosis between 2021 and 2024 (São Paulo, Rio de Janeiro, and Pernambuco), all presented studies with record linkage—this may be indicative that tuberculosis surveillance in these states has integration with academia and stimulates the production of scientific knowledge that contributes to its control. However, it is worth highlighting that for other states with high incidence, such as Pará and Rio Grande do Sul, we did not identified studies, which deserves attention.

Currently, there is the proposal to implement the e-SUS Linha da Vida [Life Line] project—which seeks interoperability between the different HIS and allows the monitoring of users of the Brazilian Unified Health System (SUS) from birth to death⁵³. In addition, there are positive perspectives for the improvement and facilitation of record linkage processes. As of 2025, the Ministry of Health started issuing the National Health Card (SUS Card) based on the CPF number, which may favor the standardization of identifiers between databases and, consequently, support implementing linkage routines⁵⁴.

As long as SUS does not implement both proposals, the need for and dependence on record linkage techniques for integrated monitoring will continue to be challenges. Without significant structural and technological advances, monitoring hospitalizations, deaths, cases of tuberculosis, and co-infections with other diseases will continue to depend on these techniques and have limited impact on the epidemiological surveillance of tuberculosis.

With regard to the contribution to tuberculosis surveillance in Brazil, the reviewed articles highlighted the potential of record linkage to qualify the notification process, identify vulnerable populations—such as persons deprived of liberty—and evaluate the effectiveness of intersectoral actions⁸,¹¹,³³,⁴³,⁵⁵. Authors of national studies have shown that the use of record linkage can be feasible on a large scale, contributing to structural analysis and evidence-based public policies⁸,¹⁹.

According to the results, record linkage has been used for various purposes, such as identifying underreporting³⁹,⁵⁰,⁵²,⁵⁶–⁵⁸, analyzing data quality¹⁰,¹¹,¹⁶,¹⁷,³²,³⁶,⁵⁹, describing epidemiological scenarios³¹,³³,⁴⁷,⁶⁰–⁶³, investigating co-infections and comorbidities⁹,⁴⁵,⁴⁸,⁴⁹,⁶⁴, assessing public policies—such as social programs or health care strategies—⁸,¹⁸–²⁰,³⁰,³⁴,³⁵ and analyzing factors associated with the disease²⁷,²⁹,⁴¹,⁴³,⁴⁴,⁶⁵–⁶⁸.

The most frequently employed HIS were SINAN-TB and SIM, whose linkage enables monitoring critical outcomes (such as deaths) and clinical events related to tuberculosis⁴⁰,⁴¹,⁴⁶,⁶¹,⁶⁹. Moreover, we evidenced the integration of data from SINAN-TB, SINAN-AIDS, SIM, and GAL, allowing for the detection of underreported cases of tuberculosis and deaths not properly registered and providing more accurate estimates of the burden of the disease in the country.

This integration of databases showed significant gaps in mortality records, exposing failures in the surveillance of deaths from tuberculosis⁴⁶,⁶⁹. In addition, the use of these techniques allowed identifying inconsistencies in SINAN-TB records¹⁶,⁵⁹, expanding the analytical potential of epidemiological information and strengthening the formulation of public policies for disease control.

The studies highlighted limitations of coverage and quality of HIS, especially due to the absence of critical information. Secondary databases undergo updating of variables, which constitutes intrinsic limitation in the context of tuberculosis. Nevertheless, they offer important advantages, such as wide spatial and temporal coverage and availability for large-scale analyses. Considering the diversity of tools identified, the findings suggest that the field has incorporated different record linkage solutions over time. Authors of future investigations can explore, in a comparative way, the performance, usability, and operational requirements of these tools in different contexts.

The concentration of studies at the national level or in few subnational scenarios indicates a gap in the literature, especially for Federative Units with high TB burden and lower representation in the reviewed scientific production.

Another relevant finding was the variability in the way of reporting the linkage procedures and their results, which made it difficult to obtain standardized information—such as number of pairs found and manual review criteria—especially when a flowchart summarizing the linkage process was presented.

Altogether, as per the included studies, record linkage has played a relevant role in the qualification of records, in the identification of underreporting, and in the expansion of the analytical possibilities of tuberculosis surveillance in Brazil.

The findings of this review suggest that record linkage remains a useful and necessary strategy to integrate information from different systems and support epidemiological analyses on tuberculosis. However, its application still depends on the quality of the records, the availability of identifiers, and the transparency with which the studies report their procedures.

All in all, according to the literature review, record linkage has been widely used for various purposes in the surveillance of tuberculosis in Brazil. At the same time, challenges remain related to the quality of information systems, the methodological heterogeneity of studies, and the concentration of production in certain geographical and analytical contexts.

This review has some limitations. There was difficulty in collecting data on the number of observations used in the linkage processes and the records effectively paired, due to the lack of objectivity of some articles when reporting this information. Many articles presented confusing writing and inconsistent sample calculations, directly interfering in the collection of information. Furthermore, some record linkage processes had their retrieval percentage reduced after manual reviews or application of epidemiological criteria related to data quality such as date inconsistency or duplicate records.

In addition, no formal evaluation of methodological quality or risk of bias of the included studies was performed, as the objective of the review was to describe methodological characteristics of the studies, and not to synthesize effect estimates or epidemiological associations; moreover, we verified heterogeneity of the study designs, analytical objectives, and forms of reporting the linkage procedures in the included studies.

• FUNDING:
  This study was carried out with the support from the National Council for Scientific and Technological Development (CNPq) – Process No. 445755/2023-7, Call for Submission 29/2023.

Supplementary material

Supplementary Table 1

Supplementary Table 1

REFERENCES

• ¹ World Health Organization. Global Tuberculosis Report 2024. Geneva: WHO; 2024 [cited on Oct. 20, 2025]. Available at: https://iris.who.int/bitstream/handle/10665/379339/9789240101531-eng.pdf?sequence=1
  » https://iris.who.int/bitstream/handle/10665/379339/9789240101531-eng.pdf?sequence=1
• ² Brasil. Ministério da Saúde. Boletim Epidemiológico de Tuberculose 2024. Ministério da Saúde; 2024 [cited on Oct. 20, 2025]. Available at: https://www.gov.br/aids/pt-br/central-de-conteudo/boletins-epidemiologicos/2024/boletim-epidemiologico-tuberculose-2024/view
  » https://www.gov.br/aids/pt-br/central-de-conteudo/boletins-epidemiologicos/2024/boletim-epidemiologico-tuberculose-2024/view
• ³ Addisu A, Adriaensen W, Balew A, Asfaw M, Diro E, Garba Djirmay A, et al. Neglected tropical diseases and the sustainable development goals: an urgent call for action from the front line. BMJ Glob Health 2019; 4(1): e001334. https://doi.org/10.1136/bmjgh-2018-001334
  » https://doi.org/10.1136/bmjgh-2018-001334
• ⁴ United Nations. Sustainable Development Goals Report 2022. New York: United Nations; 2022 [cited on Oct. 20, 2025]. Available at: https://unstats.un.org/sdgs/report/2022/
  » https://unstats.un.org/sdgs/report/2022/
• ⁵ Rocha MS, Bartholomay P, Cavalcante MV, Medeiros FC de, Codenotti SB, Pelissari DM, et al. Sistema de Informação de Agravos de Notificação (Sinan): principais características da notificação e da análise de dados relacionada à tuberculose. Epidemiol Serv Saúde 2020; 29(1): e2019017. https://doi.org/10.5123/S1679-49742020000100009
  » https://doi.org/10.5123/S1679-49742020000100009
• ⁶ Coelho Neto GC, Chioro A. Afinal, quantos Sistemas de Informação em Saúde de base nacional existem no Brasil? Cad Saúde Pública 2021;37(7): e00182119. https://doi.org/10.1590/0102-311X00182119
  » https://doi.org/10.1590/0102-311X00182119
• ⁷ Garcia KKS, de Miranda CB, de Sousa FNEF. Procedures for health data linkage: applications in health surveillance. Epidemiol Serv Saúde 2022; 31(3): e20211272. https://doi.org/10.1590/s2237-96222022000300004
  » https://doi.org/10.1590/s2237-96222022000300004
• ⁸ Durovni B, Saraceni V, Puppin MS, Tassinari W, Cruz OG, Cavalcante S, et al. The impact of the Brazilian Family Health Strategy and the conditional cash transfer on tuberculosis treatment outcomes in Rio de Janeiro: an individual-level analysis of secondary data. J Public Health (Oxf) 2018; 40(3): e359-66. https://doi.org/10.1093/pubmed/fdx132
  » https://doi.org/10.1093/pubmed/fdx132
• ⁹ Abreu RG de, Sousa AIA de, Oliveira MRF de, Sanchez MN. Tuberculose e diabetes: relacionamento probabilístico de bases de dados para o estudo da associação entre ambas doenças. Epidemiol Serv Saúde 2017; 26(2): 359-68. https://doi.org/10.5123/S1679-49742017000200013
  » https://doi.org/10.5123/S1679-49742017000200013
• ¹⁰ Bierrenbach AL, Stevens AP, Gomes ABF, Noronha EF, Glatt R, Carvalho CN, et al. Efeito da remoção de notificações repetidas sobre a incidência da tuberculose no Brasil. Rev Saúde Pública 2007; 41(Suppl 1): 67-76. https://doi.org/10.1590/S0034-89102007000800010
  » https://doi.org/10.1590/S0034-89102007000800010
• ¹¹ Bierrenbach AL, de Oliveira GP, Codenotti S, Gomes ABF, Stevens AP. Duplicates and misclassification of tuberculosis notification records in Brazil, 2001-2007. Int J Tuberc Lung Dis 2010; 14(5): 593-9.
• ¹² Oliveira GP de, Bierrenbach ALS, Camargo Júnior KR de, Coeli CM, Pinheiro RS. Accuracy of probabilistic and deterministic record linkage: the case of tuberculosis. Rev Saúde Pública 2016; 50: 49. https://doi.org/10.1590/s1518-8787.2016050006327
  » https://doi.org/10.1590/s1518-8787.2016050006327
• ¹³ Chapman BE, Irwin J. Python as a First Programming Language for Biomedical Scientists. In Proceedings of the 14th Python in Science Conference (SciPy 2015); 2015; Austin. Austin: SciPy; 2015. p. 12-7 [cited on Oct. 20, 2025]. Available at: https://scispace.com/pdf/python-as-a-first-programming-language-for-biomedical-4pra7ewfge.pdf
  » https://scispace.com/pdf/python-as-a-first-programming-language-for-biomedical-4pra7ewfge.pdf
• ¹⁴ Zhu Y, Matsuyama Y, Ohashi Y, Setoguchi S. When to conduct probabilistic linkage vs. deterministic linkage? A simulation study. J Biomed Inform 2015; 56: 80-6. https://doi.org/10.1016/j.jbi.2015.05.012
  » https://doi.org/10.1016/j.jbi.2015.05.012
• ¹⁵ Dusetzina SB, Tyree S, Meyer AM, Meyer A, Green L, Carpenter WR. Linking Data for Health Services Research: A Framework and Instructional Guide. Rockville: Agency for Healthcare Research and Quality; 2014.
• ¹⁶ Santos DT, Arroyo LH, Alves YM, Alves LS, Berra TZ, Crispim JA, et al. Survival time among patients who were diagnosed with tuberculosis, the precocious deaths and associated factors in southern Brazil. Trop Med Health 2021; 49(1): 31. https://doi.org/10.1186/s41182-021-00320-4
  » https://doi.org/10.1186/s41182-021-00320-4
• ¹⁷ Oliveira SP de, Silveira JTP da, Beraldi-Magalhães F, Oliveira RR de, Andrade L de, Cardoso RF. Early death by tuberculosis as the underlying cause in a state of Southern Brazil: Profile, comorbidities and associated vulnerabilities. Int J Infect Dis. 2019; 80: S50–7. https://doi.org/10.1016/j.ijid.2019.02.043
  » https://doi.org/10.1016/j.ijid.2019.02.043
• ¹⁸ Selig L, Belo M, Jose Ledo Alves Cunha A DA, Teixeira EG, Brito R, Lucia Luna A, et al. Óbitos atribuídos à tuberculose no Estado do Rio de Janeiro* Deaths attributed to tuberculosis in the state of Rio de Janeiro. J Bras Pneumol 2004; 30(4): 417-24. https://doi.org/10.1590/S1806-37132004000400006
  » https://doi.org/10.1590/S1806-37132004000400006
• ¹⁹ Rocha MS, Oliveira GP de, Aguiar FP, Saraceni V, Pinheiro RS. Do que morrem os pacientes com tuberculose: causas múltiplas de morte de uma coorte de casos notificados e uma proposta de investigação de causas presumíveis. Cad Saúde Pública 2015; 31(4): 709-21. https://doi.org/10.1590/0102-311X00101214
  » https://doi.org/10.1590/0102-311X00101214
• ²⁰ Viana PVS, Paiva NS, Villela DAM, Bastos LS, Bierrenbach ALS, Basta PC. Factors associated with death in patients with tuberculosis in Brazil: Competing risks analysis. PLoS One 2020; 15(10): e0240090. https://doi.org/10.1371/journal.pone.0240090
  » https://doi.org/10.1371/journal.pone.0240090
• ²¹ Bartholomay P, De Oliveira GP, Pinheiro RS, Vasconcelos AMN. Improved quality of tuberculosis data using record linkage. Cad Saúde Pública 2014; 30(11): 2459-69. https://doi.org/10.1590/0102-311x00116313
  » https://doi.org/10.1590/0102-311x00116313
• ²² Rocha MS, de Oliveira GP, Guillen LCT, Coeli CM, Saraceni V, Pinheiro RS. Use of database linkage and scripting rules to upgrade variables in the Sinan-TB database. Cad Saúde Pública 2019; 35(12): e00074318. https://doi.org/10.1590/0102-311x00074318
  » https://doi.org/10.1590/0102-311x00074318
• ²³ Torrens AW, Rasella D, Boccia D, Maciel ELN, Nery JS, Olson ZD, et al. Effectiveness of a conditional cash transfer programme on TB cure rate: A retrospective cohort study in Brazil. Trans R Soc Trop Med Hyg 2015; 110(3): 199-206. https://doi.org/10.1093/trstmh/trw011
  » https://doi.org/10.1093/trstmh/trw011
• ²⁴ Jesus GS, Gestal PFPS, Silva AF, Cavalcanti DM, Lua I, Ichihara MY, et al. Effects of conditional cash transfers on tuberculosis incidence and mortality according to race, ethnicity and socioeconomic factors in the 100 Million Brazilian Cohort. Nat Med 2025; 31(2): 653-62. https://doi.org/10.1038/s41591-024-03381-0
  » https://doi.org/10.1038/s41591-024-03381-0
• ²⁵ Carter DJ, Daniel R, Torrens AW, Sanchez MN, Maciel ELN, Bartholomay P, et al. The impact of a cash transfer programme on tuberculosis treatment success rate: A quasi-experimental study in Brazil. BMJ Glob Health 2019; 4(1): e001029. https://doi.org/10.1136/bmjgh-2018-001029
  » https://doi.org/10.1136/bmjgh-2018-001029
• ²⁶ Camargo Jr. KR, Coeli CM. Reclink: aplicativo para o relacionamento de bases de dados, implementando o método probabilistic record linkage. Cad Saúde Pública 2000; 16(2): 439-47. https://doi.org/10.1590/S0102-311X2000000200014
  » https://doi.org/10.1590/S0102-311X2000000200014
• ²⁷ R Core Team. R: A Language and environment for statistical computing. Version 4.1. Vienna: R Foundation for Statistical Computing; (2021).
• ²⁸ Asher J, Resnick D, Brite J, Brackbill R, Cone J. An Introduction to Probabilistic Record Linkage with a Focus on Linkage Processing for WTC Registries. Int J Environ Res Public Health. 2020; 17(18): 6937. https://doi.org/10.3390/ijerph17186937
  » https://doi.org/10.3390/ijerph17186937
• ²⁹ Miletic M, Sariyar M. Implementing Informative-Based Active Learning in Biomedical Record Linkage for the Splink Package in Python. Stud Health Technol Inform 2023; 305: 509-12. https://doi.org/10.3233/shti230545
  » https://doi.org/10.3233/shti230545
• ³⁰ Ali MS, Ichihara MY, Lopes LC, Barbosa GCG, Pita R, Carreiro RP, et al. Administrative Data Linkage in Brazil: Potentials for Health Technology Assessment. Front Pharmacol 2019; 10: 984. https://doi.org/10.3389/fphar.2019.00984
  » https://doi.org/10.3389/fphar.2019.00984
• ³¹ Barbosa GCG, Ali MS, Araujo B, Reis S, Sena S, Ichihara MYT, et al. CIDACS-RL: a novel indexing search and scoring-based record linkage system for huge datasets with high accuracy and scalability. BMC Med Inform Decis Mak 2020; 20(1): 289. https://doi.org/10.1186/s12911-020-01285-w
  » https://doi.org/10.1186/s12911-020-01285-w
• ³² Aridja UM, Rocha MS, Bartholomay P, Pelissari DM, Silva DA da, Poças KC, et al. Fatores associados à notificação pós-óbito de casos de tuberculose no Brasil, 2014. Cad Saúde Pública 2023; 39(6): e00301521. https://doi.org/10.1590/0102-311XPT301521
  » https://doi.org/10.1590/0102-311XPT301521
• ³³ Pinto PFPS, Teixeira CSS, Ichihara MY, Rasella D, Nery JS, Sena SOL, et al. Incidence and risk factors of tuberculosis among 420 854 household contacts of patients with tuberculosis in the 100 Million Brazilian Cohort (2004–18): a cohort study. Lancet Infect Dis 2024; 24(1): 46-56. https://doi.org/10.1016/s1473-3099(23)00371-7
  » https://doi.org/10.1016/s1473-3099(23)00371-7
• ³⁴ Pelissari DM, Lima LV de, Pavinati G, Magnabosco GT, Silva Júnior JNB, Bartholomay P, et al. Time until tuberculosis recurrence and associated factors in Brazil: a populationbased retrospective cohort study using a linked database. Rev Bras Epidemiol 2024; 27: e240016. https://doi.org/10.1590/1980-549720240016
  » https://doi.org/10.1590/1980-549720240016
• ³⁵ Torrens A, Bartholomay P, Silva S, Khogali M, Verdonck K, Bissell K, et al. HIV testing, antiretroviral therapy, and treatment outcomes in new cases of tuberculosis in Brazil, 2011. Rev Panam Salud Publica 2016; 39(1): 26-31.
• ³⁶ Saraceni V, Benzaken AS, Pereira GFM, Andrade BK, Oliveira PB, Arakaki-Sanchez D, et al. Tuberculosis burden on AIDS in Brazil: A study using linked databases. PLoS One 2018; 13(11): e0207859. https://doi.org/10.1371/journal.pone.0207859
  » https://doi.org/10.1371/journal.pone.0207859
• ³⁷ Bartholomay P, Pinheiro RS, Johansen FDC, De Oliveira SB, Rocha MS, Pelissari DM, et al. Gaps in drug-resistant tuberculosis surveillance: Probabilistic database linkage in Brazil. Cad Saúde Pública 2020; 36(5): e00082219. https://doi.org/10.1590/0102-311x00082219
  » https://doi.org/10.1590/0102-311x00082219
• ³⁸ Aridja UM, Gallo LG, Oliveira AFM, Silva AWM, Duarte EC. Casos de tuberculose com notificação após o óbito no Brasil, 2014: um estudo descritivo com base nos dados de vigilância. Epidemiol Serv Saúde 2020; 29(5): e2020060. https://doi.org/10.1590/S1679-49742020000500014
  » https://doi.org/10.1590/S1679-49742020000500014
• ³⁹ Ferreira AJF, Pescarini J, Sanchez M, Flores-Ortiz RJ, Teixeira CS, Fiaccone R, et al. Evaluating the health effect of a Social Housing programme, Minha Casa Minha Vida, using the 100 million Brazilian Cohort: A natural experiment study protocol. BMJ Open 2021; 11(3): 041722. https://doi.org/10.1136/bmjopen-2020-041722
  » https://doi.org/10.1136/bmjopen-2020-041722
• ⁴⁰ Jesus GS, Pescarini JM, Silva AF, Torrens A, Carvalho WM, Pinto Junior EP, et al. The effect of primary health care on tuberculosis in a nationwide cohort of 7·3 million Brazilian people: a quasi-experimental study. Lancet Glob Health 2022; 10(3): e390-7. https://doi.org/10.1016/s2214-109x(21)00550-7
  » https://doi.org/10.1016/s2214-109x(21)00550-7
• ⁴¹ Oliveira GP, Pinheiro RS, Coeli CM, Codenotti SB, Barreira D. linkage entre SIM e SINAN para a melhoria da qualidade dos dados do Sistema de Informação da tuberculose: a experiência nacional. Cad Saúde Colet 2010; 18(1): 107-11.
• ⁴² Carvalho CN, Dourado I, Bierrenbach AL. Subnotificação da comorbidade tuberculose e aids: uma aplicação do método de linkage. Rev Saúde Pública 2011; 45(3): 548-55. https://doi.org/10.1590/S0034-89102011005000021
  » https://doi.org/10.1590/S0034-89102011005000021
• ⁴³ Bierrenbach AL, Gomes ABF, Noronha EF, Souza MFM de. Incidência de tuberculose e taxa de cura, Brasil, 2000 a 2004. Rev Saúde Pública 2007; 41(Suppl 1): 24-33. https://doi.org/10.1590/S0034-89102007000800005
  » https://doi.org/10.1590/S0034-89102007000800005
• ⁴⁴ Oliveira GP de, Pinheiro RS, Coeli CM, Barreira D, Codenotti SB. Uso do sistema de informação sobre mortalidade para identificar subnotificação de casos de tuberculose no Brasil. Rev Bras Epidemiol 2012; 15(3): 468-77. https://doi.org/10.1590/S1415-790X2012000300003
  » https://doi.org/10.1590/S1415-790X2012000300003
• ⁴⁵ Rodrigues LC, Pereira SM, Cunha SS, Genser B, Ichihara MY, de Brito SC, et al. Effect of BCG revaccination on incidence of tuberculosis in school-aged children in Brazil: the BCG-REVAC cluster-randomised trial. Lancet. 2005; 366 (9493): 1290-5. https://doi.org/10.1016/s0140-6736(05)67145-0
  » https://doi.org/10.1016/s0140-6736(05)67145-0
• ⁴⁶ Pelissari DM, Saita NM, Monroe AA, Diaz-Quijano FA. Environmental factors associated with the time to tuberculosis diagnosis in prisoners in São Paulo, Brazil. Am J Infect Control 2022; 50(11): 1246-52.
• ⁴⁷ Aquino DS, Moura LCRV, Maruza M, Da Silva AP, Ximenes RAA, Lacerda HR, et al. Fatores associados ao tratamento da tuberculose latente em pessoas vivendo com HIV/AIDS. Cad Saúde Pública 2015; 31(12): 2505-13. https://doi.org/10.1590/0102-311X00154614
  » https://doi.org/10.1590/0102-311X00154614
• ⁴⁸ Tancredi MV, Sakabe S, Waldman EA. Mortality and survival of tuberculosis coinfected patients living with AIDS in São Paulo, Brazil: a 12-year cohort study. BMC Infect Dis 2022; 22(1): 223. https://doi.org/10.1186/s12879-022-07232-6
  » https://doi.org/10.1186/s12879-022-07232-6
• ⁴⁹ Grassi MFR, dos Santos NP, Lírio M, Kritski AL, Almeida MCC, Santana LP, et al. Tuberculosis incidence in a cohort of individuals infected with human T-lymphotropic virus type 1 (HTLV-1) in Salvador, Brazil. BMC Infect Dis 2016; 16: 491. https://doi.org/10.1186/s12879-016-1428-z
  » https://doi.org/10.1186/s12879-016-1428-z
• ⁵⁰ Soares KKS, Hisatugu WH, de Souza FM, Maciel ELN, Prado TN do. Analysis of the Vulnerability Profile of tuberculosis coinfection in people living with HIV. J Hum Growth Dev 2024; 34(2): 278-85. https://doi.org/10.36311/jhgd.v34.15778
  » https://doi.org/10.36311/jhgd.v34.15778
• ⁵¹ Miranda AE, Golub JE, Lucena FF, Maciel EN, Gurgel MF, Dietze R. Tuberculosis and AIDS Co-Morbidity in Brazil: Linkage of the Tuberculosis and AIDS Databases. Braz J Infect Dis 2009; 13(2): 137-41. https://doi.org/10.1590/s1413-86702009000200013
  » https://doi.org/10.1590/s1413-86702009000200013
• ⁵² Sousa LMO, Pinheiro RS. Óbitos e internações por tuberculose não notificados no município do Rio de Janeiro. Rev Saúde Pública 2011; 45(1): 31-9. https://doi.org/10.1590/S0034-89102011000100004
  » https://doi.org/10.1590/S0034-89102011000100004
• ⁵³ Silva MLB, Durovini P, Mota P, Kritski AL. Factors associated with underreporting of cases of multidrug-resistant tuberculosis in the state of Rio de Janeiro, Brazil: Probabilistic database linkage. Cad Saúde Pública 2021; 37(10) :e00293920. https://doi.org/10.1590/0102-311x00293920
  » https://doi.org/10.1590/0102-311x00293920
• ⁵⁴ Pinheiro RS, Andrade VL, Oliveira GP de. Subnotificação da tuberculose no Sistema de Informação de Agravos de Notificação (SINAN): abandono primário de bacilíferos e captação de casos em outras fontes de informação usando linkage probabilístico. Cad Saúde Pública 2012; 28(8): 1559-68. https://doi.org/10.1590/S0102-311X2012000800014
  » https://doi.org/10.1590/S0102-311X2012000800014
• ⁵⁵ Miranda CB de, Garcia KKS. Occupational Health Surveillance Perspectives on e-SUS Linha da Vida Program. Epidemiol Serv Saúde 2023; 32(4): e20231171. https://doi.org/10.1590/S2237-962220230004000014.en
  » https://doi.org/10.1590/S2237-962220230004000014.en
• ⁵⁶ Brasil. Ministério da Saúde começa a emitir Cartão SUS com base no CPF [Internet]. Ministério da Saúde; 2025 [cited on Oct. 20, 2025]. Available at: https://www.gov.br/saude/pt-br/assuntos/noticias/2025/setembro/ministerio-da-saude-comeca-a-emitir-cartao-sus-com-base-no-cpf
  » https://www.gov.br/saude/pt-br/assuntos/noticias/2025/setembro/ministerio-da-saude-comeca-a-emitir-cartao-sus-com-base-no-cpf
• ⁵⁷ Mabud TS, Alves MLD, Ko AI, Basu S, Walter KS, Cohen T, et al. Evaluating strategies for control of tuberculosis in prisons and prevention of spillover into communities: An observational and modeling study from Brazil. PLoS Med 2019; 16(1): e1002737. https://doi.org/10.1371/journal.pmed.1002737
  » https://doi.org/10.1371/journal.pmed.1002737
• ⁵⁸ Paula AA, Chequer P, Pires DRF, de Lemos KRV, Barone LG, Veloso VG, et al. Assessing the underreporting of deaths among people living with HIV in Rio de Janeiro, Brazil, from 2014 to 2019. Cad Saúde Pública 2022; 38(1): e00081821. https://doi.org/10.1590/0102-311X00081821
  » https://doi.org/10.1590/0102-311X00081821
• ⁵⁹ Magno ES, Saraceni V, Souza AB de, Magno RS, Saraiva MGG, Bührer-Sékula S. Fatores associados à coinfecção tuberculose e HIV: o que apontam os dados de notificação do Estado do Amazonas, Brasil, 2001-2012. Cad Saúde Pública 2017; 33(5): e00019315. https://doi.org/10.1590/0102-311X00019315
  » https://doi.org/10.1590/0102-311X00019315
• ⁶⁰ Santos ML, Coeli CM, Batista JDL, Braga MC, Albuquerque MFPM de. Factors associated with underreporting of tuberculosis based on data from Sinan Aids and Sinan TB. Rev Bras Epidemiol 2018; 21: e180019. https://doi.org/10.1590/1980-549720180019
  » https://doi.org/10.1590/1980-549720180019
• ⁶¹ Lima SVMA, Cruz LZ, Araújo DC, Santos AD dos, Queiroz AAFLN, Araújo KCGM, et al. Quality of tuberculosis information systems after record linkage. Rev Bras Enferm 2020; 73(Suppl 5): e20200536. https://doi.org/10.1590/0034-7167-2020-0536
  » https://doi.org/10.1590/0034-7167-2020-0536
• ⁶² Alves Filho P, Pelegrini Filho A, Ribeiro PT, de Toledo LM, Romão AR, Novaes LCM. Desigualdades socioespaciais relacionadas à tuberculose no município de Itaboraí, Rio de Janeiro. Rev Bras Epidemiol 2017; 20(4): 559-72. https://doi.org/10.1590/1980-5497201700040001
  » https://doi.org/10.1590/1980-5497201700040001
• ⁶³ Prado TN, Caus AL, Marques M, Maciel EL, Golub JE, Miranda AE. Perfil epidemiológico de pacientes adultos com tuberculose e AIDS no estado do Espírito Santo, Brasil: Relacionamento dos bancos de dados de tuberculose e AIDS. J Bras Pneumol 2011; 37(1): 93-9. https://doi.org/10.1590/S1806-37132011000100014
  » https://doi.org/10.1590/S1806-37132011000100014
• ⁶⁴ Saraceni V, Cohn S, Cavalcante SC, Pacheco AGF, Moulton LH, Chaisson RE, et al. Prevalent tuberculosis at HIV diagnosis in Rio de Janeiro, Brazil: The TB/HIV in Rio (THRio) cohort. J Acquir Immune Defic Syndr 2014; 67(1): 98-101. https://doi.org/10.1097/qai.0000000000000247
  » https://doi.org/10.1097/qai.0000000000000247
• ⁶⁵ Miranda AE, Dietze R, Maciel EL, Prado TN, Caus AL, Silva MM, et al. Tuberculosis and AIDS co-morbidity in children: Linkage of databases from espírito Santo State, Brazil. J Trop Pediatr 2011; 57(4): 296-8. https://doi.org/10.1093/tropej/fmq087
  » https://doi.org/10.1093/tropej/fmq087
• ⁶⁶ Naves EF, Andrade RLP, Santos GP, Ferreira MRL, Ballestero JGA, Alencar V, et al. Fatores associados ao óbito pela coinfecção tuberculose/HIV no sistema prisional. Acta Paul Enferm 2024; 37: eAPE02545. https://doi.org/10.37689/acta-ape/2024AO00002545
  » https://doi.org/10.37689/acta-ape/2024AO00002545
• ⁶⁷ Lopes VS, Couto RM, Luz AP da, Silva PRS, Lima JC. Fatores associados aos óbitos por tuberculose no estado de Mato Grosso: estudo de coorte retrospectiva, 2011-2020. Epidemiol Serv Saúde 2024; 33: e20231402. https://doi.org/10.1590/S2237-96222024V33E20231402.PT
  » https://doi.org/10.1590/S2237-96222024V33E20231402.PT
• ⁶⁸ Rocha MS, Oliveira GP, Saraceni V, Aguiar FP, Coeli CM, Pinheiro RS. Effect of inpatient and outpatient care on treatment outcome in tuberculosis: A cohort study. Rev Panam Salud Publica 2018; 42: e112. https://doi.org/10.26633/rpsp.2018.112
  » https://doi.org/10.26633/rpsp.2018.112
• ⁶⁹ Albuquerque MDFPM, Coimbra I, Batista JDA, Maruza M, Ximenes RAA, Lacerda HR, et al. Empirical treatment for TB in HIV: Lessons from a cohort study of people living with HIV treated in Recife, Brazil. BMC Public Health 2014; 14(1): 289. https://doi.org/10.1186/1471-2458-14-289
  » https://doi.org/10.1186/1471-2458-14-289

Edited by

• ASSOCIATE EDITOR:
  Maria Rita Donalísio https://orcid.org/0000-0003-4457-9897

• SCIENTIFIC EDITOR:
  Antonio Fernando Boing https://orcid.org/0000-0001-9331-1550

Data availability

All data used in this review were extracted from published scientific articles and are described in the methods section and Supplementary Table 1. No individual-level, confidential, or unpublished data were used.

Publication Dates

• Publication in this collection
  24 Aug 2026
• Date of issue
  2026

History

• Received
  28 Jan 2026
• Reviewed
  09 June 2026
• Accepted
  11 June 2026