Abstract

This retrospective cohort study identified factors associated with loss of follow-up and death due to tuberculosis (TB) in the homeless population (HP) in Brazil, estimating odds ratios (OR) and their 95% confidence intervals (95%CI) by multinomial logistic regression. A total of 3,831 TB cases in this population were analyzed, of which 57.0% had unfavorable outcomes. Loss of follow-up was associated with: history of abandonment (OR=2.38; 95%CI 2.05-2.77), unknown HIV serology (OR=1.79; 95%CI 1.38-2.32), HIV coinfection (OR=1.73; 95%CI 1.46-2.06), drug use (OR=1.54; 95%CI 1.31-1.80), age (OR=0.98; 95%CI 0.97-0.99), mixed clinical form (OR=0.64; 95%CI 0.42-0.97), extrapulmonary form (OR=0.46; 95%CI 0.29-0.73), government beneficiary (OR=0.64; 95%CI 0.50-0.81), and supervised treatment (OR=0.52; 95%CI 0.45-0.60). Regarding death, the following were associated: age (OR=1.03; 95%CI 1.01-1.05), unknown HIV serology (OR=2.39; 95%CI 1.48-3.86), alcohol consumption (OR=1.81; 95%CI 1.27-2.58), and supervised treatment (OR=0.70; 95%CI 0.51-0.96). Overlapping vulnerabilities in the health-disease process of homeless individuals with TB were observed, requiring comprehensive and cross-sectoral care practices.

Key words:
Homeless People; Health Inequities; Tuberculosis; Cohort Studies

Introduction

Despite the universal availability of treatment - within the scope of the Unified Health System (SUS) - and the possibility of a cure, tuberculosis (TB) still represents a social and public health problem worldwide¹1 Hino P, Yamamotto TT, Bastos SH, Beraldo AA, Figueiredo TMRM, Bertolozzi MR. Tuberculosis in the street population: a systematic review. Rev Esc Enferm USP 2021;55:e03688.. TB has a strong relationship with the processes of social production that generate and perpetuate situations of inequality and poverty²2 Gioseffi JR, Batista R, Brignol SM. Tuberculose, vulnerabilidades e HIV em moradores de rua: uma revisão sistemática. Rev Saude Publica 2022; 56:43.. Therefore, it is essential to boost knowledge and conduct debates on endemic TB, focused on achieving its elimination²2 Gioseffi JR, Batista R, Brignol SM. Tuberculose, vulnerabilidades e HIV em moradores de rua: uma revisão sistemática. Rev Saude Publica 2022; 56:43..

Brazil, a developing nation with great social disparities in its geopolitical territory, is ranked among the top thirty countries with the highest burden of the disease³3 Moreira ADSR, Kritski AL, Carvalho ACC. Social determinants of health and catastrophic costs associated with the diagnosis and treatment of tuberculosis. J Bras Pneumol 2020; 46(5):e20200015.. Therefore, a better understanding of the process of social determination linked to the occurrence and persistence of TB is of utmost importance, especially in the most vulnerable strata of society³3 Moreira ADSR, Kritski AL, Carvalho ACC. Social determinants of health and catastrophic costs associated with the diagnosis and treatment of tuberculosis. J Bras Pneumol 2020; 46(5):e20200015., of which victims of social, political, and economic inconsistencies likely suffer the most severe consequences of the disease³3 Moreira ADSR, Kritski AL, Carvalho ACC. Social determinants of health and catastrophic costs associated with the diagnosis and treatment of tuberculosis. J Bras Pneumol 2020; 46(5):e20200015..

Considering the susceptibility of populations that are most likely to become ill due to TB, especially due to social issues, responsible and committed action on the part of health professionals and managers is essential in an attempt to overcome obstacles to eliminating the disease⁴4 Magnabosco GT. Vulnerabilidades no contexto da saúde coletiva: contribuições, desafios e perspectivas da enfermagem. Cien Cuid Saude. 2023;22:e68409.^,55 Streit FV, Bartels C, Kuczius T, Cassier C, Gardemann J, Schaumburg F. Prevalence of latent tuberculosis in homeless persons: a single-centre cross-sectional study, Germany. PLoS One 2019; 14(3):e0214556.. Meanwhile, the homeless population (HP) stands out, as it presents a high risk of contracting the disease and having worse outcomes, especially in countries with a high burden of TB⁵5 Streit FV, Bartels C, Kuczius T, Cassier C, Gardemann J, Schaumburg F. Prevalence of latent tuberculosis in homeless persons: a single-centre cross-sectional study, Germany. PLoS One 2019; 14(3):e0214556., such as Brazil.

Homeless individuals face barriers to accessing education, work, food, hygiene, medical care, and health services, especially because they are targets of prejudice and stigma. They are therefore inserted in contexts of the overlapping of different vulnerabilities¹1 Hino P, Yamamotto TT, Bastos SH, Beraldo AA, Figueiredo TMRM, Bertolozzi MR. Tuberculosis in the street population: a systematic review. Rev Esc Enferm USP 2021;55:e03688.^,22 Gioseffi JR, Batista R, Brignol SM. Tuberculose, vulnerabilidades e HIV em moradores de rua: uma revisão sistemática. Rev Saude Publica 2022; 56:43.^,66 Hungaro AA, Gavioli A, Christóphoro R, Marangoni SR, Altrão RF, Rodrigues AL, Oliveira MLF. Homeless population: characterization and contextualization by census research. Rev Bras Enferm 2020; 73(5):e20190236.. This scenario represents a major challenge, especially for health services, since the persistence of TB has repercussions for both individuals and groups.

Previous studies have highlighted the close relationship between biological, social, economic, and programmatic factors, and the incidence of unfavorable outcomes from TB treatment in Brazil⁷7 Lima LV, Pavinati G, Palmieri IGS, Vieira JP, Blasque JC, Higarashi IH, Fernandes CAM, Magnabosco GT. Factors associated with loss to follow-up in tuberculosis treatment in Brazil: a retrospective cohort study. Rev Gaucha Enferm 2023; 44:e20230077.^,88 Saita NM, Andrade RLP, Bossonario PA, Bonfim RO, Hino P, Monroe AA. Factors associated with unfavorable outcome of tuberculosis treatment in people deprived of liberty: a systematic review. Rev Esc Enferm USP 2021; 55:e20200583.. However, the impact of living on the streets on such outcomes still requires further investigation, especially when considering the notable challenges in ensuring effective indicators of disease control in this vulnerable population, as has been observed in prior research⁹9 Pavinati G, Lima LV, Radovanovic CAT, Magnabosco GT. Geoprogrammatic disparities in the performance of tuberculosis indicators in the homeless population in Brazil: an ecological approach. Rev Bras Epidemiol 2023; 26:e230048..

From the perspective of the concept of vulnerability, this work adds a new interpretation to the contexts of the fragility of homeless individuals undergoing TB treatment, by highlighting how this situation can influence the loss of follow-up and death due to infection in this population group. Therefore, to collaborate with more assertive disease control strategies, this study aimed to identify the factors associated with the unfavorable outcomes of TB cases in the HP in Brazil.

Methods

This work is a non-concurrent, population-based cohort study, conducted using data from notifications of TB cases in the HP in Brazil, between 2015 and 2021. These records were obtained from the Notifiable Diseases Information System (Sistema de Informação de Agravos de Notificação - SINAN), which is fed throughout the treatment of TB patients, which lasts approximately six months¹⁰10 Brasil. Ministério da Saúde (MS). Secretaria de Vigilância em Saúde. Brasil livre da tuberculose - Plano nacional pelo fim da tuberculose como problema de saúde pública: estratégias para 2021-2025. Brasília: MS; 2021.. The information from this follow-up is included in the case monitoring and notification forms, serving as a retrospective source of data for SINAN¹⁰10 Brasil. Ministério da Saúde (MS). Secretaria de Vigilância em Saúde. Brasil livre da tuberculose - Plano nacional pelo fim da tuberculose como problema de saúde pública: estratégias para 2021-2025. Brasília: MS; 2021..

The time frame is justified by the inclusion of the variable “homeless population (yes; no; ignored)” in the notification form, which was applied in 2015¹¹11 Rocha MS, Bartholomay P, Cavalcante MV, Medeiros FC, Codenotti SB, Pelissari DM, Andrade KB, Silva GDM, Arakaki-Sanchez D, Pinheiro RS. Sistema de Informação de Agravos de Notificação Sinan: principais características da notificação e da análise de dados relacionada à tuberculose. Epidemiol Serv Saude 2020; 29(1):e2019017.. The data were accessed through the Brazilian National Health System Information Technology Department (Departamento de Informática do Sistema Único de Saúde - DATASUS) in November 2022, which is why the analysis ended in 2021. The files were downloaded in the “.dbc” format, read by the R® software, using the “read.dbc” package, and exported to the “.csv” format¹²12 Petruzalek D. Read.dbc - um pacote para importação de dados do Datasus na linguagem R. In: Anais do XV Congresso Brasileiro de Informática em Saúde. Goiânia; 2016..

TB cases were considered in people aged 18 to 59 years, in which the variable “homeless population” was marked as “yes”. For inclusion, only records were selected whose variable “entry type” was filled with the options: “new case” - person never subjected to treatment or treated for less than 30 days; “relapse” - person with reinfection or recurrence of the previous infection; and “re-entry after abandonment” - person who started new treatment after abandonment¹⁰10 Brasil. Ministério da Saúde (MS). Secretaria de Vigilância em Saúde. Brasil livre da tuberculose - Plano nacional pelo fim da tuberculose como problema de saúde pública: estratégias para 2021-2025. Brasília: MS; 2021..

In this study, the dependent variable was constructed by “closure situation”, considering the following: “cure” (treatment completed as recommended); “primary abandonment” (use of medication for less than 30 days and interruption for 30 consecutive days), and “abandonment” (use of medication for more than 30 days and interruption for 30 consecutive days) - called “loss of follow-up”; and “death” (with TB as the underlying cause)¹⁰10 Brasil. Ministério da Saúde (MS). Secretaria de Vigilância em Saúde. Brasil livre da tuberculose - Plano nacional pelo fim da tuberculose como problema de saúde pública: estratégias para 2021-2025. Brasília: MS; 2021.. The independent variables were:

a) sociodemographic: sex (male; female), age (in years), self-declared race/color (white; non-white - black/yellow/brown/indigenous), education (no study; up to 8 years of study; more than 8 years of study), and beneficiary of a government income transfer program (yes; no).

b) clinical-epidemiological: type of entry (reentry after abandonment; relapse; new case), diabetes (yes; no), use of drugs/alcohol/tobacco (yes; no), mental illness (yes; no), infection by human immunodeficiency virus - HIV (yes; no; unknown - not carried out/in progress), clinical form (pulmonary; extrapulmonary; mixed - pulmonary/extrapulmonary), and directly observed treatment - DOT (yes; no).

The complete case analysis approach was adopted, which consisted of excluding records with at least one ignored/blank variable. This decision was based on a previous exploration of the bank, observing a considerable amount of missing data. Given this finding, we opted for the aforementioned approach, since the presence of missing data could compromise the integrity and reliability of the analysis, preventing an appropriate interpretation of the results.

For the descriptive analysis of the data, the absolute and relative frequencies of the categorical variables, as well as the measures of central tendency and dispersion for the variable “age” (mean - me; median - md; standard deviation - sd), were calculated. To verify the associations between the dependent and independent variables, multinomial (multiclass)¹³13 Fernandes AAT, Figueiredo Filho DB, Rocha EC, Nascimento WS. Read this paper if you want to learn logistic regression. Rev Sociol Polit 2020; 28(74):006. logistic regression models were used, given the categorical nature with more than two categories in the response variable (cure; loss of follow-up; death).

Cure was considered to be the reference category for the dependent variable in order to model the factors that influenced the loss of follow-up and death due to TB in homeless individuals. Initially, bivariate models were constructed and associations were reported as odds ratios (OR) - not adjusted for any confounding factors (crude analysis) - and their 95% confidence intervals (95% CI). Variables with a p-value<0.20 were considered for multiple analysis.

Multicollinearity of variables was assessed and discarded before the multiple models. Afterwards, a selection was carried out using the backward stepwise method, in which all variables with a p-value<0.20 were initially included in the likelihood ratio test. They were then removed one by one, keeping only those with a p-value<0.05. From this, the adjusted ORs were presented and were considered to be significant when their 95%CI did not cross the null value (1,00).

To verify the robustness of the findings and point out possible association biases related to the exclusion of cases with missing information, a post hoc analysis was performed, considering the missing data as a subcategory of the independent variables present in the final model. All analyses were performed using the SPSS® software, version 21.0. Furthermore, it should be noted that the findings were discussed in light of the concept of vulnerability, proposed by Ayres et al.¹⁴14 Ayres JRCM, França Junior I, Calazans G, Salletti H. Vulnerabilidade e prevenção em tempos de Aids. In: Barbosa R, Parker R, organizadores. Sexualidade pelo avesso: direitos, identidades e poder. Rio de Janeiro: Relume Dumará; 1999. p. 50-71..

This study is derived from the Master’s thesis entitled Tuberculosis in homeless individuals in Brazil: evidence of geoprogrammatic disparities and predictors of unfavorable outcome, defended in the Programa de Pós-Graduação em Enfermagem of the Universidade Estadual de Maringá, in 2023. According to Resolutions no. 466/2012 and no. 674/2022, a favorable opinion was obtained from the Research Ethics Committee, logged under certificate of presentation for ethical assessment (CAAE) no. 63981922.6.0000.0104.

Results

In the period from 2015 to 2021, 13,839 TB cases were reported in homeless individuals in Brazil, considering those aged between 18 and 59 years and whose “closure status” on SINAN was filled in as “cure”, “primary abandonment/abandonment” or “death due to TB”. Of these, 10,008 (79.7%) cases were excluded due to a lack of information on a given variable; therefore, 3,831 (20.3%) cases remained and were included in the analysis of this cohort study.

Exclusions due to missing data for each specific variable are presented below, in descending order: beneficiary of government program (n=6,283; 45.4%), education (n=4,862; 35.1%), DOT (n=1,218; 8.8%), self-declared race/color (n=1,055; 7.6%), mental illness (n=1,005; 7.3%), diabetes (n=943; 6.8%), tobacco use (n=815; 5.9%), drug use (n=685; 4.9%) and alcohol consumption (n=663; 4.8%), HIV serology (n=67; 0.5%), and sex (n=1; 0.0%).

Regarding the outcome of the cases included, 42.8% (n=1,640) were concluded as cured, 52.6% (n=2,016) were loss of follow-up, and 4.6% (n=175) death due to TB. Among the notifications, it was observed that the majority referred to male individuals (n=2,973; 77.6%), of non-white race/color (n=2,863; 74.7%), with up to eight years of education (n=2,965; 77.4%), and who were not beneficiaries of income transfer programs (n=3,476; 90.7%) (Table 1).

Furthermore, the highest occurrence of TB cases was observed with the entry type filled as new case (n=2,051; 53.5%), pulmonary clinical form (n=3,605; 94.1%), DOT performed (n=1,930; 50.4%), and who consumed alcoholic beverages (n=2,129; 55.6%), tobacco (n=2,309; 60.3%), and/or drugs (n=2,600; 67.9%). Regarding comorbidities, it was observed that the absence of HIV, diabetes, and mental illness predominated among the included cases (Table 1).

In the bivariate analysis, significant associations were seen in the following variables: sex, race/color, age, education, government beneficiary, type of entry, clinical form, consumption of alcohol, diabetes, HIV, DOT, use of drugs, and use of tobacco (Table 2). Subsequently, eight variables remained associated in the final model, through multivariate analysis, namely: age, government beneficiary, type of entry, clinical form, alcohol consumption, HIV, DOT, and drug use.

Factors associated with loss to follow-up were: drug use (OR=1.54; 95%CI 1.31-1.80), reentry after dropout (OR=2.38; 95%CI 2.05-2.77), co-infection with HIV (OR=1.73; 95%CI 1.46-2.06) and unknown serology (OR=1.79; 95%CI 1.38-2.32), age (OR=0.98; 95%CI 0.97-0.99), mixed form (OR=0.64; 95%CI 0.42-0.97) and extrapulmonary form (OR=0.46; 95%CI 0.29-0.73), government beneficiary (OR=0.64; 95%CI 0.50-0.81), and completion of the DOT (OR=0.52; 95%CI 0.45-0.60) (Table 3).

Regarding death due to TB, the following factors were significantly associated in the multiple model: age (OR=1.03; 95%CI 1.01-1.05), lack of knowledge of HIV serology (OR=2.39; 95%CI 1.48-3.86), alcohol consumption (OR=1.81; 95%CI 1.27-2.58), and completion of DOT (OR=0.70; 95%CI 0.51-0.96) (Table 3). Sensitivity analysis revealed similar estimates for the final model variables in the complete case and missing indicator approaches (Table 4).

Discussion

This study made it possible to identify sociodemographic and clinical-epidemiological factors associated with unfavorable outcomes in TB cases among homeless individuals in Brazil. Despite advances in combating infection, it is commonly recognized that the HP represents a group that suffers the effects of the social determination of the health-disease process, and may find itself in a context that is less sensitive to TB prevention and management actions, even in developed countries¹⁵15 Nguipdop-Djomo P, Rodrigues LC, Smith PG, Abubakar I, Mangtani P. Drug misuse, tobacco smoking, alcohol and other social determinants of tuberculosis in UK-born adults in England: a community-based case-control study. Sci Rep 2020; 10(1):5639.^,1616 Self JL, Mcdaniel CJ, Morris SB, Silk BJ. Estimating and evaluating tuberculosis incidence rates among people experiencing homelessness, United States, 2007-2016. Med Care 2021; 59(Supl. 2):S175-S181..

It is well-known that homeless individuals are daily subject to challenges arising from the precarious and often inhumane living conditions in which they find themselves, requiring specific strategies to guarantee comprehensive and equitable care¹⁶16 Self JL, Mcdaniel CJ, Morris SB, Silk BJ. Estimating and evaluating tuberculosis incidence rates among people experiencing homelessness, United States, 2007-2016. Med Care 2021; 59(Supl. 2):S175-S181.^,1717 Carmo ME, Guizardi FL. O conceito de vulnerabilidade e seus sentidos para as políticas públicas de saúde e assistência social. Cad Saude Publica 2018; 34(3):e00101417.. Therefore, it is essential to employ the epidemiological approach associated with the concept of vulnerability for a more in-depth understanding of the contexts of street dwelling¹⁷17 Carmo ME, Guizardi FL. O conceito de vulnerabilidade e seus sentidos para as políticas públicas de saúde e assistência social. Cad Saude Publica 2018; 34(3):e00101417. and its relationship with TB.

According to Ayres et al. precepts¹⁴14 Ayres JRCM, França Junior I, Calazans G, Salletti H. Vulnerabilidade e prevenção em tempos de Aids. In: Barbosa R, Parker R, organizadores. Sexualidade pelo avesso: direitos, identidades e poder. Rio de Janeiro: Relume Dumará; 1999. p. 50-71.^,1818 Ayres JRCM, Paiva V, França JRI, Gravato N, Lacerda R, Della Negra M, Marques HHS, Galano E, Lecussan P, Segurado AC, Silva MH. Vulnerabilidade, direitos humanos e necessidades de atenção integral à saúde de jovens vivendo com HIV/aids. Am J Public Health 2006; 96(6):1001-1006., vulnerability can be understood in three interrelated dimensions, namely: individual - which involves access to information and the ability to incorporate it into attitudes, knowledge, beliefs, values, among others; social - related to social and cultural norms, encompassing stigma, prejudice, social support, among others; and programmatic - which refers to the actions and responses of social institutions (health, education, social assistance, public security, among others)¹⁴14 Ayres JRCM, França Junior I, Calazans G, Salletti H. Vulnerabilidade e prevenção em tempos de Aids. In: Barbosa R, Parker R, organizadores. Sexualidade pelo avesso: direitos, identidades e poder. Rio de Janeiro: Relume Dumará; 1999. p. 50-71.^,1818 Ayres JRCM, Paiva V, França JRI, Gravato N, Lacerda R, Della Negra M, Marques HHS, Galano E, Lecussan P, Segurado AC, Silva MH. Vulnerabilidade, direitos humanos e necessidades de atenção integral à saúde de jovens vivendo com HIV/aids. Am J Public Health 2006; 96(6):1001-1006..

In other words, in addition to individual and behavioral issues that influence illness, social and structural aspects related to access to health services are also considered¹⁸18 Ayres JRCM, Paiva V, França JRI, Gravato N, Lacerda R, Della Negra M, Marques HHS, Galano E, Lecussan P, Segurado AC, Silva MH. Vulnerabilidade, direitos humanos e necessidades de atenção integral à saúde de jovens vivendo com HIV/aids. Am J Public Health 2006; 96(6):1001-1006.. Therefore, to understand the social determination of TB, it is necessary to identify the specific needs of this social group - which certainly originate from vulnerable contexts - and consider the situations that discriminate, subordinate, and weaken these subjects.

In this retrospective cohort at a national level, it was observed that the following variables were positively associated with the chance of losing TB treatment follow-up, as compared to a cure, among people who are homeless: reentry after abandonment (OR=2.38; 95%CI 2.05-2.77), lack of knowledge of HIV serology (OR=1.79; 95%CI 1.38-2.32) and co-infection with HIV (OR=1.73; 95%CI 1.46-2.06), and drug use (OR=1.54; 95%CI 1.31-1.80).

One literature review, which did not investigate the loss of follow-up specifically in the HP, identified different factors related to treatment interruption and, among these, it was observed that the history of abandonment is an important predictor of a new episode of abandonment¹⁹19 Ferreira MRL, Bonfim RO, Siqueira TC, Orfão NH. Abandono do tratamento da tuberculose: uma revisão integrativa. Rev Enferm Contemp 2018; 7(1):63-71.. Considering the precarious contexts known to be faced by the HP²2 Gioseffi JR, Batista R, Brignol SM. Tuberculose, vulnerabilidades e HIV em moradores de rua: uma revisão sistemática. Rev Saude Publica 2022; 56:43.^,66 Hungaro AA, Gavioli A, Christóphoro R, Marangoni SR, Altrão RF, Rodrigues AL, Oliveira MLF. Homeless population: characterization and contextualization by census research. Rev Bras Enferm 2020; 73(5):e20190236., the high number of reentry after abandonment seen in this study raises an alert for professionals involved in providing health care.

One ecological study, which evaluated TB indicators in the HP between 2015 and 2019, pointed to the low HIV testing in the national territory, highlighting states, such as Mato Grosso and Pará, with percentages of lower than 75% in offering HIV tests to people with TB⁹9 Pavinati G, Lima LV, Radovanovic CAT, Magnabosco GT. Geoprogrammatic disparities in the performance of tuberculosis indicators in the homeless population in Brazil: an ecological approach. Rev Bras Epidemiol 2023; 26:e230048.. Although the strategy (double testing) is recommended for all people with TB or HIV²⁰20 Lima LV, Pavinati G, Oliveira RR, Couto RM, Alves KBA, Magnabosco GT. Temporal trend in the incidence of tuberculosis-HIV coinfection in Brazil, by macro-region, Federative Unit, sex and age group, 2010-2021. Epidemiol Serv Saude 2024; 33:e2023522., there are persistent challenges in implementing this approach among homeless individuals.

In Brazil, the high occurrence of TB-HIV co-infection in the HP is also a reality, corresponding to more than 30% of all cases in some states⁹9 Pavinati G, Lima LV, Radovanovic CAT, Magnabosco GT. Geoprogrammatic disparities in the performance of tuberculosis indicators in the homeless population in Brazil: an ecological approach. Rev Bras Epidemiol 2023; 26:e230048.. Considering that co-infection was recognized as a predictor of the loss to follow-up in other studies⁷7 Lima LV, Pavinati G, Palmieri IGS, Vieira JP, Blasque JC, Higarashi IH, Fernandes CAM, Magnabosco GT. Factors associated with loss to follow-up in tuberculosis treatment in Brazil: a retrospective cohort study. Rev Gaucha Enferm 2023; 44:e20230077.^,1919 Ferreira MRL, Bonfim RO, Siqueira TC, Orfão NH. Abandono do tratamento da tuberculose: uma revisão integrativa. Rev Enferm Contemp 2018; 7(1):63-71., it is highlighted that difficulties linked to understanding and accepting the double therapeutic regimen, side effects and lack of complacency on the part of health professionals when faced with complaints from people undergoing treatment can contribute to non-adherence²¹21 Magnabosco GT, Andrade RLP, Arakawa T, Monroe AA, Villa TCS. Desfechos dos casos de tuberculose em pessoas vivendo com HIV: subsídios para intervenção. Acta Paul Enferm 2019; 32(5):554-563..

The greatest risk of loss of TB treatment follow-up in the context of the lives of people who use legal/illicit drugs may be related to the overlapping and interrelationship of vulnerabilities, since stigma, discrimination, lack of financial resources, and risk behaviors tend to be high in these individuals²²22 Camargo PO, Oliveira MM, Raupp LM, Pereira GB, Ramos CI. Políticas públicas e sociais frente à vulnerabilidade social no território da Cracolândia. Saude Soc 2022; 31(1):e200969.^,2323 Apodaca MJFR, Castro-Granell V, Laguía A, Jaén Á, Cenoz S, Galindo MJ. Drug use and antiretroviral therapy (ART) interactions: a qualitative study to explore the knowledge, beliefs, adherence, and quality of life of people living with HIV taking ART and illicit drugs. AIDS Res Ther 2020; 17(1):24.. Especially in the HP with TB, these results are even more worrying, given that a large proportion of individuals consume alcohol and drugs.

It is also important to note that protective aspects were also identified in the multiple model. For each additional year of life, there was a 2% decrease in the chances of interrupting treatment. This may be associated, for example, with the greater susceptibility of young people to risk and negligence scenarios related to self-care or even as a result of the person not accepting the diagnosis¹⁹19 Ferreira MRL, Bonfim RO, Siqueira TC, Orfão NH. Abandono do tratamento da tuberculose: uma revisão integrativa. Rev Enferm Contemp 2018; 7(1):63-71., which would explain their greater abandonment of TB treatment follow-up.

Obtaining government benefits and undergoing DOT also diminished the chances of the loss of follow-up by 36% and 48%, respectively. Such strategies are crucial for the protection and guarantee of social rights, routinely neglected in the HP. One cohort study conducted with 274 homeless individuals developed in the United States detected a significant improvement in the chances of completing TB treatment, mainly attributed to the use of DOT²⁴24 Onwubiko U, Wall K, Sales RM, Holland DP. Using directly observed therapy (DOT) for latent tuberculosis treatment - a hit or a miss? A propensity score analysis of treatment completion among 274 homeless adults in Fulton County, GA. PLoS One 2019; 21(14):e0218373..

The mixed (OR=0.64; 95%CI 0.42-0.97) and extrapulmonary (OR=0.46; 95%CI 0.29-0.73) clinical forms of TB reduced the chances of the loss of follow-up. These results may be related to the logic of TB care in the care network, in which cases with more complex forms must also be treated on an outpatient and/or hospital basis. This could help, for example, to reduce the chances of abandonment, since people should be referred to services to continue their treatment.

Furthermore, it can be inferred that, by encompassing more teams and professionals involved in care, homeless individuals would be monitored at care levels in conjunction with primary health care (PHC)²⁵25 Pinto PFPS, Santos BPS, Teixeira CSS, Nery JS, Amorim LDAF, Sanchez MN, Barreto ML, Pescarini JM. Performance evaluation of tuberculosis control in Brazilian municipalities. Rev Saude Publica 2022; 56:53., which would thus enable greater opportunities to establish a tie and, consequently, adherence to TB treatment. These findings suggest the importance of coordinated action between different levels of care to guarantee better TB indicators among the HP.

Regarding death from TB, it was found that with each year of increasing age, there was a 3% increase in the chance of this outcome; the consumption of alcohol (OR=1.81; 95%CI 1.27-2.58) and the lack of knowledge of HIV serology (OR=2.39; 95%CI 1.48-3.86) were also associated with greater chances of death due to TB. This result is close to the findings of an integrative review that identified that individuals, aged 43 years and over, and the consumption of alcoholic beverages were predictors of death due to the disease in the prison system²⁶26 Camillo AJG, Ferreira MRL, Bossonario PA, Andrade RLP, Saita NM, Rezende CEM, Arcência RA, Monroe AA. Fatores associados ao óbito por tuberculose e HIV/aids em presídios: revisão integrativa. Acta Paul Enferm 2022; 35:eAPE01606..

Therefore, the potential influence of alcohol consumption on death due to TB can be assumed. The abuse of these substances can cause clinical and physiological repercussions, as it is commonly responsible for liver damage and nutritional and immunological deficiency²⁷27 Silva DR, Muñoz-Torrico M, Duarte R, Galvão T, Bonini EH, Arbex FF, Arbex MA, Augusto VM, Rabahi MF, Mello FCQ. Fatores de risco para tuberculose: diabetes, tabagismo, uso de álcool e uso de outras drogas. J Bras Pneumol 2018; 44(2):145-152., especially when there is prolonged exposure, primarily in adults of advanced age and/or diagnosed with alcoholism¹⁵15 Nguipdop-Djomo P, Rodrigues LC, Smith PG, Abubakar I, Mangtani P. Drug misuse, tobacco smoking, alcohol and other social determinants of tuberculosis in UK-born adults in England: a community-based case-control study. Sci Rep 2020; 10(1):5639.. Furthermore, it is assumed that high levels of consumption can, in general, interfere with living conditions and wellbeing, harmfully influencing people’s habits and daily lives.

Regarding the lack of knowledge about HIV serology, it must be considered that under-diagnosis, resulting from the failure to carry out tests - often resulting from the fragmentation of care in services - causes failures in HIV treatment, which can increase the risk of death due to TB, given the immunological compromise caused by the virus²⁸28 Mocellin LP, Winkler GB, Stella IM, Vieira PC, Beck C, Behar PRP, Kuchenbecker RS. Caracterização dos óbitos e dos itinerários terapêuticos investigados pelo comitê municipal de mortalidade por aids de Porto Alegre em 2015. Epidemiol Serv Saude 2020; 29(3):e2019355.. Therefore, it is urgent to promote dialogue between TB and HIV programs and other sectors so that universal testing can be offered to people at any given opportunity²⁰20 Lima LV, Pavinati G, Oliveira RR, Couto RM, Alves KBA, Magnabosco GT. Temporal trend in the incidence of tuberculosis-HIV coinfection in Brazil, by macro-region, Federative Unit, sex and age group, 2010-2021. Epidemiol Serv Saude 2024; 33:e2023522..

One possibility to integrate care is the creation and guidance of intersectoral and social protection strategies, such as the implementation of the singular therapeutic project (STP) for homeless individuals diagnosed with TB, in a humanized and comprehensive way²⁹29 Razente MS, Lima LV, Silva IGP, Monteiro LRS, Cecilio HPM, Gil NLM, Magnabosco GT. Perfil clínico-epidemiológico das pessoas acometidas por HIV/aids, tuberculose e hanseníase no Paraná, Brasil, 2010-2019. Cien Cuid Saude 2022; 21:e61725.. Still, from this perspective, it is important to highlight and reiterate DOT together with HP as an effective strategy to treat the disease, which, in this cohort study, was able to reduce the chance of death due to TB in this population by 30%.

Despite the existence of public policies, programs, guidelines, and actions that, in theory, guarantee health care for the HP, there are important gaps in their operationalization⁹9 Pavinati G, Lima LV, Radovanovic CAT, Magnabosco GT. Geoprogrammatic disparities in the performance of tuberculosis indicators in the homeless population in Brazil: an ecological approach. Rev Bras Epidemiol 2023; 26:e230048., making these individuals more exposed to situations of violence and deprivation of rights³3 Moreira ADSR, Kritski AL, Carvalho ACC. Social determinants of health and catastrophic costs associated with the diagnosis and treatment of tuberculosis. J Bras Pneumol 2020; 46(5):e20200015.. In this study, this scenario of disparities was viewed by the high proportion of homeless individuals with unfavorable outcomes from TB cases, together with the contexts that exacerbate them.

It is essential to mitigate obstacles that affect the reality of homeless people living with the disease. In this sense, the Federal Government recently ratified Decree No. 11,908/2024, establishing the Healthy Brazil Program (Programa Brasil Saudável), coordinated by the Interministerial Committee for the Elimination of Tuberculosis and Other Socially Determined Diseases (Comitê Interministerial para a Eliminação da Tuberculose e de Outras Doenças Determinadas Socialmente - CIEDDS)³⁰30 Brasil. Decreto nº 11.908, de 6 de fevereiro de 2024. Institui o Programa Brasil Saudável - Unir para Cuidar, e altera o Decreto nº 11.494, de 17 de abril de 2023, para dispor sobre o Comitê Interministerial para a Eliminação da Tuberculose e de Outras Doenças Determinadas Socialmente - CIEDDS. Diário Oficial da União 2024; 7 fev.. This initiative seeks to promote intersectoral actions that eliminate such problems as public health issues by the end of this decade³⁰30 Brasil. Decreto nº 11.908, de 6 de fevereiro de 2024. Institui o Programa Brasil Saudável - Unir para Cuidar, e altera o Decreto nº 11.494, de 17 de abril de 2023, para dispor sobre o Comitê Interministerial para a Eliminação da Tuberculose e de Outras Doenças Determinadas Socialmente - CIEDDS. Diário Oficial da União 2024; 7 fev..

The implementation of strategies that involve different sectors to combat TB, such as CIEDDS, is essential for the elimination of TB by 2030³¹31 Almeida CPB, Trajman A, Mitnick CD, Kritski AL, Santos-Filho ET. Beyond goodwill and promises-urgent needs and opportunities to accelerate the fight against tuberculosis in Brazil. Lancet Reg Health Am 2024; 30:100663.. This goal cannot be achieved without measures to reduce social inequality rooted in the country³¹31 Almeida CPB, Trajman A, Mitnick CD, Kritski AL, Santos-Filho ET. Beyond goodwill and promises-urgent needs and opportunities to accelerate the fight against tuberculosis in Brazil. Lancet Reg Health Am 2024; 30:100663.. Therefore, it is urgent that the management spheres and services involved in the care of people with TB act in an integrated manner so that it will be possible to implement this audacious program, with the potential to eliminate socially determined diseases and to promote intersectoral care throughout Brazil.

The abyss of persistent inequality in our society demands social and health policies that fully address the needs of a population as vulnerable as the HP³²32 Aldridge RW, Story A, Hwang SW, Nordentoft M, Luchenski AS, Hartwell G, Tweed EJ, Lewer D, Katikireddi SV, Hayward AC. Morbidity and mortality in homeless individuals, prisoners, sex workers, and individuals with substance use disorders in high-income countries: a systematic review and meta-analysis. Lancet 2018; 391(10117):241-250.^,3333 Macedo LR, Maciel ELN, Struchiner CJ. Vulnerable populations and tuberculosis treatment outcomes in Brazil. Cien Saude Colet 2021; 26(10):4749-4759.. Therefore, the urgency for the integration and articulation of the various spheres of SUS and other governmental and non-governmental sectors should be reinforced, seeking continuous and longitudinal care through health services and actions to prevent and control the disease.

As limitations of this study, the use of secondary data is assumed, which may be subject to incorrect completion and the underreporting of cases, in turn falsifying the true epidemiological scenario of the disease, especially in the HP. Furthermore, the high quantity of missing data stands out, which can contribute to selection, information, and analysis biases; in other words, this may lead to uncertain conclusions about the relationship between the variables and reduce the power to generalize the results.

However, even when considering that the large number of exclusions may limit the interpretation of the findings, the post hoc analysis revealed similar estimates for the variables in the final model when comparing the complete cases and missing indicator approaches. Thus, the inclusion of ignored information as a subcategory of the independent variables appears to have maintained the direction and strength of the association for most of the analyses presented in this study.

Furthermore, given the inclusion of the variable “homeless population” in the notification form only in 2015, another limitation arises related to the possible incompleteness of the records on the part of health professionals, who may not be used to filling out the variable properly. Accurate information about each case, especially those related to the HP, is essential for the organization and sustainability of a comprehensive, integrated, and effective health network.

Finally, it is worth noting that the COVID-19 pandemic may have influenced the analyses, given the inclusion of data from 2020 and 2021, generating incorrect measurements of unfavorable outcomes that need to be considered in future studies. Changes in the patterns of seeking health services, social isolation, prioritization of resources, changes in people’s behavior, and adaptations of assistance and surveillance services may have influenced access to care, the quality of treatment, and the surveillance of TB cases³⁴34 Hino P, Yamamoto TT, Magnabosco GT, Bertolozzi MR, Taminato M, Fornari LF. Impacto da covid-19 no controle e reorganização da atenção à tuberculose. Acta Paul Enferm 2021; 34:eAPE002115.^,3535 Garcia MT, Resende MR, Santos NMS, Ferreira AT, Silva MDFN, Mendes EDT, Angerami RN, Agrela MVR, Martins AC, Hofling CC. Impacto da pandemia de covid-19 no desfecho de tratamento da tuberculose em hospital terciário. Braz J Infect Dis 2022; 26:102478..

Conclusion

The loss of follow-up of TB cases in the HP was associated with younger age, reentry after abandonment, co-infection with HIV and the lack of knowledge of serology, and drug use. Receiving a government grant, undergoing DOT, and having mixed and extrapulmonary clinical forms diminished the chances. Death from TB was linked to advancing age, the lack of knowledge of HIV serology, and the consumption of alcohol, while performing the DOT reduced the chances of death.

These findings highlight the contexts that make people vulnerable and have an impact on the health-disease process of homeless individuals, making them more susceptible to the unfavorable outcomes of TB. Thus, the importance of analyzing the relationship between sociodemographic and clinical-epidemiological aspects, as well as other elements that weaken these people, is clear, seeking to qualify coping strategies to be sensitive to this well-known marginalized group.

Furthermore, the potential of surveillance is highlighted in relation to the planning, control, and management of TB, which leads to the need for a constant qualification of services and their tools. Finally, the demand for studies with different approaches is highlighted, which aims to elucidate the contexts experienced by this social group in order to broaden the understanding of the bases that support this reality and to enable the implementation of strategies that modify their structuring conditions.

References

Publication Dates

History