Abstract
Background
Globally tuberculosis disease (TB) poses a significant public health challenge. People deprived of liberty (PDL) are one of the most affected. This study aims to evaluate the positivity of interferon gamma release assay (IGRA) and molecular sputum testing for TB through a multicenter survey of Brazilian prisons and evaluate factors associated with IGRA positivity.
Methods
We performed a cross-sectional study among PDL in six male and two female prisons within 6 cities in Brazil in 2025. Eligible participants included consenting adults over 18 years. We randomized 130 individuals per site using prison census for evaluation with a sociodemographic questionnaire, blood testing with QuantiFERON-TB Gold Plus® (IGRA) and sputum testing with GeneXpert Ultra® (Xpert). We included individuals with complete data and valid IGRA results for analysis. Multivariable logistical regression was performed to evaluate factors associated with IGRA positivity stratified by gender.
Results
Among 1040 PDL randomized, 884 were included. The overall IGRA positivity was 53.6% (95% confidence interval [CI] 50.3–56.9) varying between 29.2% (21.6–38.2) and 84.0% (75.3–90.1) among prisons sites, male and female individuals had a positivity of 61.9% (58.1–65.5) and 30.3% (24.7–36.5), respectively. In the multivariable analysis, IGRA positivity was associated with lower education (adjusted odds ratio [aOR] 1.46, 95% CI 1.03–2.07, p = 0.034), previous incarceration (aOR 2.81, 1.85–4.25, p < 0.001), previous TB history (aOR 2.99, 1.39–6.42, p < 0.001) and incarceration time (< 1 year as reference, 3–4 years: aOR 2.15, 1.18–3.92, p = 0.012; 5 years or above: aOR 2.85, 1.66–4.90, p < 0.001) for males. In females, IGRA positivity was associated with previous TB history (aOR 5.93, 1.97–20.2, p = 0.002). One prison could not collect sputum samples; at the remaining sites we detected 20 individuals (2.5%, 20/770) with positive Xpert.
Conclusion
Despite variations among prisons and gender, IGRA and Xpert positivity were elevated across sites evaluated in Brazil. These findings underscore the urgent need for systematic and comprehensive tuberculosis screening, prevention, and control strategies in prisons, including the potential evaluation of new preventative technologies such as new TB vaccines.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12889-026-27643-3.
Keywords: Tuberculosis prevalence, Mycobacterium tuberculosis infection, people deprived of liberty, Brazil
Background
Tuberculosis (TB) remains as one of the greatest public health challenges globally, affecting over 10 million people each year [1]. With increasing efforts being made to curb TB incidence and mortality with the End TB goals, a better understanding of the current epidemiological situation is needed to plan healthcare responses and guide future research strategies [2]. Targeted evaluations of the prevalence of Mycobacterium tuberculosis (MTB) infection and active TB disease performed through active case finding or surveys in high burden groups can provide vital information for policymakers, researchers and public health authorities [3].
People deprived of liberty (PDL) are one of the groups with the highest burden of TB disease worldwide. Previous estimates found that in South America, the TB incidence rate is 27-fold higher in prisons than in the overall population [4]. In Brazil, despite only 0.4% of the population being incarcerated, 9.1% of the new TB cases occurred in PDL in 2023 [5]. These disparities result in carceral settings becoming reservoirs of disease to the surrounding communities and a large excess burden of infection in the national population [6, 7].
Recommendations from the World Health Organization and the Brazilian Ministry of Health suggest that active TB screening should be performed periodically in prisons [8, 9]. However, there is no current systematic screening program in Brazilian prisons. Several groups have performed active case finding studies independently with different methodologies [10–15] but no prevalence estimate was performed consistently across different regions and current data corresponds mostly to passive diagnosis, which may lead to gaps in case detection rates [16].
New surveys can provide updated information about the active TB cases and MTB infection for public health planning, infection control and dynamic infectious diseases modelling. Despite the current understanding that tests evaluating immunoreactivity to MTB are not a perfect measure of the true underlying infection proportion, those results could provide an essential panorama for planning new technologies for disease control [17]. Currently there are several new TB vaccines at the development pipeline, the participation of people deprived of liberty in these trials has been discussed in the context of possible benefits for disease prevention in this vulnerable group and increased efficiency of trials due to the frequency of disease endpoints [18]. Previous studies have assessed that these trials proposals are well accepted by the prison population [19, 20].
Analyzing the data generated from a multicenter cross-sectional survey in prisons across all five Brazilian regions, this paper aims to understand the prevalence of IGRA positivity and active TB disease, evaluating associated sociodemographic and clinic characteristics associated with IGRA positivity.
Methods
Study setting and design
We performed a cross-sectional study in six male prisons (Belo Horizonte, Campo Grande, Manaus, Porto Velho, Salvador and Venâncio Aires) and two female prisons (Campo Grande and Manaus) located in six cities in Brazil, covering all five country regions. Individuals incarcerated in those units were all adults (over 18 years old) sentenced to a closed system regimen where individuals do not leave the prison during incarceration. The prisons had variable architecture, where most individuals are held in cell blocks constituted of closed cells organized around open wards. Individuals spend less than 4 h a day outside of the cells, with smaller cell blocks dedicated to specific populations such as workers or solitary confinement. Movement between cells and cell blocks is frequent and is determined by prison staff, however we did not had access to reliable measures of individual movements in each prison. These sites were selected for the study because they already had research teams experienced with prison health research.
In 2025, male prison population varied between 606 individuals (Venâncio Aires) and 1593 individuals (Campo Grande) with a mean occupancy rate of 158%, varying between 105% (Salvador) and 360% (Campo Grande). The female prisons had a population of 220 individuals with 111% of occupancy (Manaus) and 311 individuals with 137% occupancy (Campo Grande) [21]. Only two units, both in Campo Grande, currently perform some screening for active TB disease in their population with frequent mass screening campaigns in both prisons with Xpert and chest X-ray performed for all individuals and an ongoing cohort in the male prison evaluating a proportion of the population every four months with chest X-ray and sputum analysis. The remaining units rely on passive diagnosis alone, where individuals with TB symptoms seek care in the prison clinic and are evaluated. Data collection and study procedures occurred between April 2025 and October 2025.
Data collection
Prison staff provided study workers with a complete unit census with unique prison identification numbers in the week of study procedures that was used to randomize our sample through simple random sampling. We did not perform any stratification accounting for cell block size, crowding or other site characteristics. After randomization, we obtained name, cell block and cell number of the randomized individuals to be approached for interview. Study procedures occurred inside prison grounds, within the prison wards, so it was necessary for prison staff to know who would be approached to open cell blocks for contact with individuals. Prison staff were not present during interview, neither were they informed about willingness to participate. There were no incentives nor reprimands for individuals to engage in the study procedures.
We estimated from previous studies [22] that a minimum sample size of 428 individuals would be required to detect a 50% proportion of IGRA positivity with a precision of ± 5%, accounting for 10% of losses or refusals to participate. We opted for a sample frame where each site would be equally sampled plus 10 extra individuals per site to account for losses and refusals. Thus, we opted for an equal sampling of 120 PDL for each unit, randomizing 130 individuals from each prison unit to account for losses or refusals. We excluded individuals that did not want to participate and those who were illiterate and would not be able to read the informed consent form. Security measures in each unit only allowed a determined number of participants to be seen each day for a limited period of time, so we stopped inclusion when the sample size approached the most possible number of participants considering prison setting limitations.
Those who were able to read the consent form and were capable of providing informed consent were evaluated through a sociodemographic questionnaire, including carceral history, medical history, contact with TB cases in the same cell, current illicit drug use or tobacco smoking and alcohol abuse related disorders. Individuals self-reported their HIV status; we did not perform HIV testing. A version of the questionnaire translated to English is available in the supplement (Supplementary Material 2 – Annex 1). There were no other exclusion criteria for participation. After initial evaluation, individuals underwent venous blood collection for interferon-gamma release assay (IGRA) testing with the QuantiFERON-TB Gold Plus®, processed and categorized according to manufacturer instructions. All individuals were asked to provide one individual sputum sample after IGRA for individual testing with GeneXpert Ultra® (Xpert). After data collection and sample processing we excluded patients with missing questionnaire data, missing or indeterminate IGRA results.
Laboratory procedures and results management
The QuantiFERON-TB Gold Plus® assay was performed according to the manufacturer’s instructions. The same operator was responsible for sample processing across all sites. Blood samples from each participant were incubated at 36–38 °C for 16–24 h for cell stimulation. After incubation, supernatants were harvested, stored at − 80 °C, and subsequently sent to Campo Grande for batch ELISA testing, ensuring operational protocol standardization. Results were considered valid only when accepted by the QFT-Plus software. A positive result was defined as an IFN-γ response ≥ 0.35 IU/mL above the Nil value in either TB1 or TB2, and indeterminate results were defined per manufacturer criteria (Nil > 8.0 IU/mL or Mitogen < 0.5 IU/mL).
Sputum samples were stored at 2–8 °C on site and transferred on the same day to the local laboratory for Xpert processing, reflecting routine prison healthcare clinic conditions. We reported to prison healthcare providers individuals with positive and trace results to be evaluated by medical consult. According to the Brazilian guidelines, all individuals with positive Xpert results are considered cases and trace results are considered as positive in people living with HIV or if there is clinical suspicion of extra-pulmonary disease. Considering we did not collect follow-up information for the TB cases after medical consult; we considered TB cases to be only the individuals with a positive Xpert result at first evaluation. Individual follow-up, decision to treat and care was performed by the healthcare clinic professionals in each site, according to Brazilian TB care guidelines [9].
Data management and statistical analysis
The study team collected and stored data in an online RedCap® (Vanderbilt University, Tennessee, USA) database hosted by Fiocruz, Mato Grosso do Sul, Brazil. Data analysis was performed utilizing R statistical software [23]. At first, we conducted Chi-square and Fisher tests for categorical variables to determine differences in the response variables among subgroups of different genders and IGRA results, we also evaluated if there was a trend for positivity with increasing incarceration time through the Cochran–Armitage test. Due to significant differences between male and female groups regarding prison characteristics, demographics and carceral history, we performed a separate analysis of factors associated with IGRA positivity in each group. We utilized the Wilson score interval to obtain confidence intervals for the prevalence of IGRA positivity in the overall population and prison and gender subgroups.
Univariate logistic regression analysis was performed on both groups. A collinearity verification was performed calculating the variance inflation factor for the variables included. The multivariable model aimed to evaluate a predictive model for IGRA positivity in the PDL. We performed a sensitivity analysis of the main model where we selected only the population without previous history of TB, since those individuals might have persistent IGRA positive results after treatment.
For male prisons, we included in the multivariable model all variables that had a strong association with IGRA positivity in the univariate analysis (with a p-value < 0.05), in addition to including age as an a priori confounder; we included prison unit as a random effect in the model to account for the variability between units. For the female prisons analysis, due to a smaller sample size with fewer outcomes, we performed a simpler fixed effects model evaluating previous TB history and variables that are related to period of exposure to the prisons setting (current incarceration time and previous incarceration) due to previous data from Brazilian prisons suggesting an strong association of prison history with TB outcomes [11, 14]. Prison units were also accounted for adjustment as fixed effects in the female model.
Results
After the first randomization, 1040 individuals were selected for initial approach, 130 individuals in each prison site. In total, 945 individuals were able to be approached and provided informed consent to participate. We excluded four individuals due to missing sociodemographic information, 19 individuals with blood samples not collected, and five individuals with indeterminate IGRA results (two at Campo Grande female prison and three at the Venâncio Aires prison). Due to a lab processing error of samples, 33 individuals from the Porto Velho city prison were also excluded. In the final analysis, 884 individuals were included among the eight prisons (Fig. 1). Sputum samples were not collected for the 107 individuals evaluated Salvador prison and 7 individuals from the remaining sites had missing samples, so the Xpert results are presented for the 770 participants with valid sputum results. The response rate varied between 72.3% (Porto Velho) and 90.7% (Campo Grande – Female), the total number of individuals included for the final analysis and missing sputum samples in each study site is reported in Table 1.
Fig. 1.
Cities included in the study evaluation and flowchart of inclusion of study participants. Legend: A – Cities where the study prison units were located within Brazil. Five diferent country regions are represented in different colors, at least one prison unit was evaluated in each region. B – Flowchart of participant inclusion. Abreviations: IGRA: Interferon gamma release assay; Xpert: GeneXpert Ultra
Table 1.
Characteristics of people deprived of liberty evaluated in a tuberculosis cross-sectional survey, included in the final analysis, by prison unit (n = 884)
| Variables | Prison City Site | ||||||||
|---|---|---|---|---|---|---|---|---|---|
|
Overall
N = 884a |
MA-F
N = 113a |
MA-M
N = 113a |
CG-F
N = 118a |
CG-M
N = 115a |
PV N = 94a | VA N = 117a | BH N = 107a | SA N = 107a | |
| Age categories | |||||||||
| 18–25 years old | 154 (17.4%) | 20 (17.7%) | 23 (20.4%) | 20 (16.9%) | 14 (12.2%) | 22 (23.4%) | 15 (12.8%) | 24 (22.4%) | 16 (15.0%) |
| 26–35 years old | 389 (44.0%) | 54 (47.8%) | 64 (56.6%) | 40 (33.9%) | 44 (38.3%) | 41 (43.6%) | 59 (50.4%) | 47 (43.9%) | 40 (37.4%) |
| 36–45 years old | 238 (26.9%) | 22 (19.5%) | 18 (15.9%) | 42 (35.6%) | 40 (34.8%) | 26 (27.7%) | 32 (27.4%) | 28 (26.2%) | 30 (28.0%) |
| Above 46 years old | 103 (11.7%) | 17 (15.0%) | 8 (7.1%) | 16 (13.6%) | 17 (14.8%) | 5 (5.3%) | 11 (9.4%) | 8 (7.5%) | 21 (19.6%) |
| Race | |||||||||
| White | 177 (20.0%) | 17 (15.0%) | 12 (10.6%) | 22 (18.6%) | 26 (22.6%) | 12 (12.8%) | 62 (53.0%) | 16 (15.0%) | 10 (9.3%) |
| Black/Mixed and Others | 707 (80.0%) | 96 (85.0%) | 101 (89.4%) | 96 (81.4%) | 89 (77.4%) | 82 (87.2%) | 55 (47.0%) | 91 (85.0%) | 97 (90.7%) |
| Schooling | |||||||||
| Incomplete basic education | 425 (48.1%) | 38 (33.6%) | 64 (56.6%) | 47 (39.8%) | 63 (54.8%) | 64 (68.1%) | 45 (38.5%) | 48 (44.9%) | 56 (52.3%) |
| Complete basic education or higher | 459 (51.9%) | 75 (66.4%) | 49 (43.4%) | 71 (60.2%) | 52 (45.2%) | 30 (31.9%) | 72 (61.5%) | 59 (55.1%) | 51 (47.7%) |
| Previously incarcerated | 631 (71.4%) | 58 (51.3%) | 96 (85.0%) | 74 (62.7%) | 74 (64.2%) | 85 (90.4%) | 104 (88.9%) | 78 (72.9%) | 62 (57.9%) |
| Previous contact with TB case in the same cell | 291 (32.9%) | 16 (14.2%) | 60 (53.1%) | 45 (38.1%) | 45 (39.1%) | 38 (40.4%) | 59 (50.4%) | 8 (7.5%) | 20 (18.7%) |
| Current incarceration time | |||||||||
| <1 year | 340 (38.5%) | 73 (64.6%) | 24 (21.2%) | 62 (52.5%) | 89 (77.4%) | 11 (11.7%) | 25 (21.4%) | 47 (43.9%) | 9 (8.4%) |
| 1–2 years | 247 (27.9%) | 25 (22.1%) | 32 (28.3%) | 24 (20.3%) | 17 (14.8%) | 48 (51.1%) | 47 (40.2%) | 19 (17.8%) | 35 (32.7%) |
| 3–4 years | 121 (13.7%) | 5 (4.4%) | 32 (28.3%) | 14 (11.9%) | 0 (0%) | 17 (18.1%) | 14 (12.0%) | 10 (9.3%) | 29 (27.1%) |
| 5 years or above | 176 (19.9%) | 10 (8.8%) | 25 (22.1%) | 18 (15.3%) | 9 (7.8%) | 18 (19.1%) | 31 (26.5%) | 31 (29.0%) | 34 (31.8%) |
| Previous TB history | 76 (8.6%) | 7 (6.2%) | 9 (8.0%) | 8 (6.8%) | 13 (11.3%) | 13 (13.8%) | 19 (16.2%) | 1 (0.9%) | 6 (5.6%) |
| Positive IGRA result | 474 (53.6%) | 33 (29.2%) | 78 (69.0%) | 37 (31.4%) | 49 (42.6%) | 79 (84.0%) | 74 (63.2%) | 60 (56.1%) | 64 (59.8%) |
| Xpert result3 | |||||||||
| Positive | 20/770 (2.5%) | 0/111 (0%) | 2/111 (1.8%) | 2/116 (1.7%) | 1/115 (0.8%) | 5/94 (5.3%) | 7/117(5.9%) | 3/106 (2.8%) | 0 (NA) b |
| Trace | 9/770 (1.1%) | 0/111 (0%) | 0/111 (0%) | 1/116 (0.9%) | 0/115 (0%) | 1/94 (1.0%) | 3/117 (2.5%) | 4/106 (3.7%) | 0 (NA) b |
| Negative | 741/770 (96.4%) | 111/111 (100%) | 109/111 (98.2%) | 113/116 (97.4%) | 114/115 (99.2%) | 88/94 (93.7%) | 107/117 (91.6%) | 99/106 (93.5%) | 0 (NA) b |
| Missing | 114 | 2 | 2 | 2 | 0 | 0 | 0 | 1 | 107b |
| Living with HIV | 13 (1.5%) | 3 (2.7%) | 0 (0%) | 2 (1.7%) | 0 (0%) | 0 (0%) | 5 (4.3%) | 2 (1.9%) | 1 (0.9%) |
| Alcohol abuse | 41 (4.6%) | 1 (0.9%) | 0 (0%) | 0 (0%) | 2 (1.7%) | 0 (0%) | 28 (23.9%) | 7 (6.5%) | 3 (2.8%) |
| Current tobacco smoking | 228 (25.8%) | 1 (0.9%) | 2 (1.8%) | 36 (30.5%) | 27 (23.5%) | 15 (16.0%) | 86 (73.5%) | 25 (23.4%) | 36 (33.6%) |
| Current illicit drug use | 137 (15.5%) | 5 (4.4%) | 5 (4.4%) | 16 (13.6%) | 12 (10.4%) | 2 (2.1%) | 77 (65.8%) | 10 (9.3%) | 10 (9.3%) |
Abreviations: MA-F Manaus – Female, MA-M Manaus – Male, CG – F Campo Grande – Female, CG – M Campo Grande – Male, PV Porto Velho, VA Venâncio Aires, BH Belo Horizonte, SA Salvador, IGRA Interferon Gamma release assay, TB Tuberculosis, Xpert GeneXpert Ultra
an (%), bSalvador site did not perform sputum testing
Overall, 26.1% (231/884) of study participants were from female prisons. The evaluated PDL were mostly under 35 years old (61.4%), not white (80.0%) and almost half had incomplete basic education (48.1%). Most individuals had a previous history of incarceration (71.4%) and two thirds of them (66.4%) have been currently incarcerated for up to two years. We encountered significant variation by prison in the proportion of reported history of alcohol abuse (0% to 23.9%), current tobacco smoking (0.9% to 73.5%) and illicit drug use in the last year (2.1% to 65.8%), however the overall proportion of these characteristics was notable (4.6%, 25.8% and 15.5%, respectively). Only 1.5% of individuals reported living with HIV, proportionally similar to the national estimates for the overall population [24]. Among all participants, 8.6% reported previous TB history and 32.9% had previous contact with TB positive individuals in the same cell (Table 1). Individuals differed significantly when compared by gender in most characteristics evaluated, however we found no difference in age, race, previous TB history and HIV status (Supplementary Table 1).
The IGRA positivity was high, with an average of 53.6% (95% confidence interval [CI] 50.3–56.9) positivity across all prisons. There was significant variation between sites, the lowest positivity detected was 29.2% (95% CI 21.6–38.2) in the Manaus female prison and the highest was 84.0% (95% CI 75.3–90.1) in the Porto Velho male prison (Table 1). IGRA positivity was considerably higher in male (61.9%, 95% CI 58.1–65.5) than female individuals (30.3%, 95% CI 24.7–36.5, p < 0.001) (Supplementary Table 1). A stratified analysis across subgroups also evidenced that IGRA positivity was considerably higher in males then females in most subgroups evaluated (Fig. 2).
Fig. 2.
Proportion of IGRA positivity across different subgroups of people deprived of liberty in Brazil, stratified by gender. (n = 884). Legend: Values are presented in the measured percentage of positivity with 95% confidence intervals obtained with the Wilson score interval. Small sample sizes for individuals living with HIV (n = 13) and females with alcohol abuse history (n = 1) distorted the estimates. Abbreviations: IGRA – Interferon gamma release assay; TB – Tuberculosis
Univariate analysis found that in male prisons individuals, a positive IGRA result was associated with lower completion of basic education (44.1% vs. 54.2%, p = 0.012), previous TB history (12.9% vs. 3.6%, p < 0.001), contact with TB cases in the same cell (38.6% vs. 29.7%, p = 0.021) and previous incarceration history (83.7% vs. 64.7%, p < 0.001). We also identified an increasing trend in for IGRA positivity with increasing incarceration time (Cochran–Armitage test for trend, p < 0.001). In female prisons, a positive IGRA results was associated with previous TB history (14.3% vs. 3.1%, p = 0.003) (Table 2). A secondary evaluation for trend stratified by gender and incarceration history identified an increased IGRA positivity by current incarceration time in both males and females without previous incarceration history and males with previous incarceration history (Supplementary Table 2).
Table 2.
Variables associated with IGRA positivity in people deprived of liberty evaluated in a tuberculosis cross-sectional survey, by gender (n = 884)
| Variables | IGRA Result - Male | IGRA Result - Female | ||||
|---|---|---|---|---|---|---|
| Negative N = 249a |
Positive N = 404a |
p-valueb | Negative N = 161a |
Positive N = 70a |
p-valueb | |
| Age categories | > 0.900 | 0.300 | ||||
| 18–25 years old | 42 (16.9%) | 72 (17.8%) | 32 (19.9%) | 8 (11.4%) | ||
| 26–35 years old | 113 (45.4%) | 182 (45.0%) | 64 (39.8%) | 30 (42.9%) | ||
| 36–45 years old | 65 (26.1%) | 109 (27.0%) | 45 (28.0%) | 19 (27.1%) | ||
| Above 46 years old | 29 (11.6%) | 41 (10.1%) | 20 (12.4%) | 13 (18.6%) | ||
| Race | 0.200 | 0.140 | ||||
| White | 59 (23.7%) | 79 (19.6%) | 31 (19.3%) | 8 (11.4%) | ||
| Black/Mixed and Others | 190 (76.3%) | 325 (80.4%) | 130 (80.7%) | 62 (88.6%) | ||
| Schooling | 0.012 | 0.700 | ||||
| Incomplete basic education | 114 (45.8%) | 226 (55.9%) | 58 (36.0%) | 27 (38.6%) | ||
| Complete basic education or higher | 135 (54.2%) | 178 (44.1%) | 103 (64.0%) | 43 (61.4%) | ||
| Previously incarcerated | 161 (64.7%) | 338 (83.7%) | < 0.001 | 87 (54.0%) | 45 (64.3%) | 0.150 |
| Previous contact with TB case in the same cell | 74 (29.7%) | 156 (38.6%) | 0.021 | 41 (25.5%) | 20 (28.9%) | 0.600 |
| Current incarceration time | < 0.001c | 0.324c | ||||
| <1 year | 104 (41.8%) | 101 (25.0%) | 99 (61.5%) | 36 (51.4%) | ||
| 1–2 years | 78 (31.3%) | 120 (29.7%) | 30 (18.6%) | 19 (27.1%) | ||
| 3–4 years | 29 (11.6%) | 73 (18.1%) | 14 (8.7%) | 5 (7.1%) | ||
| 5 years or above | 38 (15.3%) | 110 (27.2%) | 18 (11.2%) | 10 (14.3%) | ||
| Previous TB history | 9 (3.6%) | 52 (12.9%) | < 0.001 | 5 (3.1%) | 10 (14.3%) | 0.003 |
| Xpert resultd | 0.500 | 0.400 | ||||
| Positive | 3/205 (1.4%) | 15/338 (4.4%) | 1/158 (0.6%) | 1/69 (1.4%) | ||
| Trace | 3/205 (1.4%) | 5/338 (1.4%) | 0/158 (0%) | 1/69 (1.4%) | ||
| Negative | 199/205 (97.2%) | 318/338 (94.2%) | 157/158 (99.4%) | 67/69 (97.2%) | ||
| Missing | 44 | 66 | 3 | 1 | ||
| Living with HIV | 3 (1.2%) | 5 (1.2%) | > 0.900 | 4 (2.5%) | 1 (1.4%) | > 0.900 |
| Alcohol abuse | 21 (8.4%) | 19 (4.7%) | 0.053 | 0 (0%) | 1 (1.4%) | 0.300 |
| Current tobacco smoking | 68 (27.3%) | 123 (30.4%) | 0.400 | 26 (16.1%) | 11 (15.7%) | > 0.900 |
| Illicit drug use | 48 (19.3%) | 68 (16.8%) | 0.400 | 16 (9.9%) | 5 (7.1%) | 0.500 |
Abbreviations: IGRA Interferon Gamma release assay, TB Tuberculosis, Xpert GeneXpert Ultra
an (%), bFisher’s exact test, Pearson’s Chi-squared test, cCochran-Armitage test for trend, dSalvador site did not perform sputum testing and 7 other sputum samples were missing from the remaining prisons
The multivariable model for male prisons found that IGRA positivity was associated with incomplete basic education (adjusted odds ratio [aOR] 1.46, 95% confidence interval [CI] 1.03–2.07, p = 0.034), previous incarceration (aOR 2.81, 95% CI 1.85–4.25, p < 0.001), previous TB history (aOR 2.99, 95% CI 1.39–6.42, p = 0.005) and longer current incarceration time (< 1 year as reference, 3–4 years: aOR 2.15, 95% CI 1.18–3.92, p = 0.012; 5 years or above: aOR 2.85, 95% CI 1.66–4.90, p < 0.001). We found no association with age or TB contacts in the same cell for IGRA positivity in the multivariable analysis. The female prison model found that only previous TB history (aOR 5.93, 95% CI 1.97–20.2, p = 0.002) was associated with IGRA positivity, with no significant association with previous prison or current incarceration time (Table 3). The sensitivity analysis findings, excluding individuals with previous TB history, were similar to the main model, with no significant change in the coefficients for time of incarceration and previous incarceration for the male population and both characteristics were not associated with IGRA positivity in females (Supplementary Table 3).
Table 3.
Crude and adjusted odds ratios for IGRA positivity in people deprived of liberty evaluated in a tuberculosis cross-sectional survey, by gender (n = 884)
| Variables | Male (n = 653) | Female (n = 231) | ||||||
|---|---|---|---|---|---|---|---|---|
| cOR (95% CI) | p-value | aOR (95% CI) | p-value | cOR (95% CI) | p-value | aOR (95% CI) | p-value | |
| Age categories | ||||||||
| 18–25 years old | Reference | Reference | Reference | |||||
| 26–35 years old | 0.94 (0.60–1.14) | 0.800 | 0.69 (0.42–1.13) | 0.140 | 1.88 (0.80–4.81) | 0.200 | ||
| 36–45 years old | 0.98 (0.60–1.59) | > 0.900 | 0.79 (0.46–1.36) | 0.400 | 1.69 (0.68–4.53) | 0.300 | ||
| Above 46 years old | 0.82 (0.45–1.52) | 0.500 | 0.66 (0.34–1.30) | 0.200 | 2.60 (0.93–7.64) | 0.073 | ||
| Schooling | ||||||||
| Complete basic education or higher | Reference | Reference | ||||||
| Incomplete basic education | 1.50 (1.10–2.07) | 0.012 | 1.46 (1.03–2.07) | 0.034 | 0.90 (0.50–1.85) | 0.700 | ||
| Race | ||||||||
| White | Reference | Reference | ||||||
| Black/Mixed and Others | 1.28 (0.87–1.87) | 0.200 | 1.85 (0.84–4.53) | 0.150 | ||||
| Current incarceration time | ||||||||
| <1 year | Reference | Reference | Reference | Reference | ||||
| 1–2 years | 1.58 (1.07–2.36) | 0.023 | 1.14 (0.71–1.83) | 0.600 | 1.74 (0.87–3.46) | 0.110 | 1.87 (0.91–3.82) | 0.084 |
| 3–4 years | 2.59 (1.57–4.36) | < 0.001 | 2.15 (1.18–3.92) | 0.012 | 0.98 (0.30–2.77) | > 0.900 | 0.82 (0.23–2.49) | 0.700 |
| 5 years or above | 2.98 (1.90–4.47) | < 0.001 | 2.85 (1.66–4.90) | < 0.001 | 1.53 (0.63–3.57) | 0.300 | 1.30 (0.50–3.19) | 0.500 |
| Previous incarceration | 2.80 (1.94–4.06) | < 0.001 | 2.81 (1.85–4.25) | < 0.001 | 1.53 (0.86–2.76) | 0.150 | 1.42 (0.77–2.64) | 0.250 |
| Previous contact with TB case in the same cell | 1.49 (1.06–2.09) | 0.021 | 1.13 (0.77–1.66) | 0.500 | 1.17 (0.62–2.18) | > 0.900 | ||
| Previous TB history | 3.94 (2.00–8.69) | < 0.001 | 2.99 (1.39–6.42) | 0.005 | 5.20 (1.77–17.3) | 0.004 | 5.93 (1.97–20.2) | 0.002 |
| Current tobacco use | 1.17 (0.82–1.66) | 0.400 | 0.97 (0.43–2.04) | > 0.900 | ||||
| Alcohol abuse* | 0.54 (0.28–1.02) | 0.057 | Not included* | |||||
| Drug use | 0.85 (0.56–1.28) | 0.400 | 0.70 (0.22–1.87) | 0.500 | ||||
Abbreviations: cOR Crude Odds Ratio, aOR Adjusted Odds Ratio, CI Confidence Interval, TB Tuberculosis
*Alcohol abuse was only reported by one female participant and it was not included in the univariate analysis
Considering the 770 sputum samples available in seven prisons, we detected 20 individuals (2.6%, 20/770, 95% CI 1.6–3.9%) with positive Xpert results in our sample, corresponding to a point prevalence of 2597 cases per 100,000 PDL evaluated (95% CI 1687–3977). Three (15%, 3/20) of the positive Xpert tests also detected Rifampicin resistance (two samples from Manaus male prison and one sample from Venâncio Aires prison). Also, nine individuals had a Xpert trace result, these individuals were also referred to clinical evaluation by prison healthcare staff.
Discussion
In this cross-sectional evaluation of eight Brazilian prisons, we found that over half of the individuals evaluated had a positive IGRA result and a high prevalence of active TB disease detected by Xpert. Male prisons had higher IGRA positivity than female prisons. The risk associated with IGRA positivity in male individuals were less education, previous incarceration, longer time of current incarceration and previous TB history; in female prisons, IGRA positivity was only associated with previous TB history.
Previous assessments utilizing tuberculin skin testing (TST) or IGRA in Brazilian prisons have shown TST/IGRA positivity between 11.7 and 73.0% [10, 13, 14, 22, 25–27], with the latest evaluation performed in 2021. To our knowledge, the Porto Velho male prison had a proportion of IGRA positivity (84.0%, 95% CI 75.3–90.1) above the highest rate previously recorded for Brazilian prisons. Although there is evidence that point TST or IGRA evaluation might be biased to assess the true prevalence of MTB infection due to false-positives and a significant proportion of individuals that do not convert despite high exposure to disease [28], these findings can be used to evaluate trends over time within the population and can still be used for estimates of the annual risk of infection of the population [17].
The prevalence of active TB disease detected by positive Xpert in our sample was also high, 2597 cases per 100,000 PDL, within the previously estimates for prevalence in South America region [4]. Similar discussions about bias introduced by false-positive Xpert results in surveys have been debated in the literature [29], however recent evidence has suggested that the Xpert specificity for similar screening studies is above 99% [30]. However, we did not collect information about the initial medical consultations, chest X-rays and culture results or symptom status for the individuals with positive Xpert results. It is important to reiterate that the burden of disease could be overestimated by false positives due to other reasons such as recently finalized TB treatment.
In context of previous evidence that carceral settings influence the national TB epidemic [6, 7], our findings reveal an urgent need for new strategies of TB control in prisons. We also evidence a significant heterogenicity in the IGRA positivity between prisons, suggesting different disease transmission profiles between these populations. New proposed approaches that might mitigate disease burden should account for these differences and could include biomedical interventions such as systematic active case finding efforts incorporating sputum and chest X-ray evaluation [31] or structural reforms of prison conditions and policy [7]. These efforts can not only reduce transmission by early identification and treatment of affected individuals, but also avoid the diagnostic delay of passive detection that could lead to post-TB lung disease and other complications [32].
Currently, national guidelines do not recommend tuberculosis preventive treatment (TPT) in prison settings [9]. Our findings of general high prevalence of IGRA positivity could be used to inform current discussions to update the recommended practices and reconsider including TPT as a strategy of disease control within the population. In some settings, considering that most individuals might already have a positive TST or IGRA result, it could be feasible to perform TPT in all individuals at risk without previous immunological testing, given that active disease is ruled out in those individuals eligible for TPT. These possible approaches have been shown to be well accepted and cost-effective in similar low and middle income countries [33]. Also, since the risk of developing active TB remains high years after incarceration [34], strategies to implement TPT prior to release or after incarceration could reduce the transmission burden to the overall population. However, implementation of such new strategies requires planning and assessment of feasibility with health authorities, accounting for heterogeneity between prisons, to promote highly efficient healthcare interventions in these vulnerable populations.
Our study also support earlier findings that incarceration time is strongly associated with TST/IGRA positivity [11, 14, 35]. This association was stronger for longer periods of exposure, despite previous reports that over a quarter of individuals can present TST conversion in a single year of follow-up [36]. Previous incarceration also indicates longer time at risk in this high burden environment. However, we did not encounter this association when evaluating the female prisons model, despite identifying a trend of increasing positivity over time of incarceration in females without previous prison and reports suggesting that the same is valid for both populations [14]. We hypothesize that this finding is due to the smaller number of females individuals evaluated and that in our sample incarceration time and previous incarceration proportions were significantly lower compared to the male individuals. Despite the absence of national estimates of this data stratified by gender, a survey performed in 12 prisons in a single state identified similar discrepancies in prison profiles of male and female units [14].
These findings should be evaluated in the light of the study limitations. We performed a random selection of individuals for inclusion not accounting for each facility characteristics. It is common for PDLs to be unevenly allocated in cell blocks due to crime profile or gang affiliations, that could lead to different crowding or demographics conditions distributed across the prison introducing bias, considering our small coverage of the total prison population. Moreover, except Venâncio Aires, all cities where the study was conducted are state capitals. We have no reason to believe that prison characteristics significantly differ within each state, however the TB incidence reported in those capitals is higher than the overall state [37]. Also, despite the randomization for individual selection, study sites were not selected randomly but chosen due to the presence of experienced research teams in prison healthcare, which could have influenced individual participation in procedures, previous site healthcare access and might not reflect the reality of other prison settings. Finally, the follow-up of the detected TB cases was performed by the prison healthcare units, so our report doesn’t include information about medical evaluation or further phenotypic and genotypic testing of the resistant TB cases.
Conclusion
In conclusion, this multicenter survey demonstrates that tuberculosis infection and active disease remain highly prevalent in Brazilian carceral settings, with risk strongly associated with cumulative exposure to incarceration. These findings underscore the urgent need for systematic and comprehensive tuberculosis screening, prevention, and control strategies in prisons, including the potential evaluation of new preventative technologies such as new TB vaccines.
Supplementary Information
Acknowledgements
The authors thank the study participants for their kind cooperation during the data collection process, the researchers and workers involved in all cities for the support and excellent work practices and the carceral staff administration in all participating prisons for their support with study procedures and coordination.
Abbreviations
- IGRA
Interferon gamma release assay
- TST
Tuberculin skin testing
- TB
Tuberculosis
- TPT
Tuberculosis Preventive Treatment
- MTB
Mycobacterium tuberculosis
- Xpert
GeneXpert Ultra
- PDL
People deprived of liberty
Authors’ contributions
All authors attest they meet the ICMJE criteria for authorship. RGW, JC and EFL conceived the study design. GGAA, JVBB, KMS, MSV, EFL, EGB, LGP, GLF, BCVO, JSJ, AGC, MCS, MPAV, MGT, RR, BBD, PPC, BBA participated in the data collection process. GGAA, JVBB, KMS, MSV, EFL, EGB, LGP, GLF, BCVO, JSJ, AGC, MCS, MPAV, MGT, RR, BBD, PPC, BBA assisted with the sample collections/processing and data collection process. GGAA, JVBB, KAT and RAC performed the analyses. GGAA, JVBB, RAC and KAT interpreted the results. JVBB, GGAA, EFL wrote the first draft of the manuscript. RAC, KAT, RGW, JC critically revised the first draft of the manuscript. KMS, MSV, EGB, LGP, GLF, BCVO, JSJ, AGC, MCS, MPAV, MGT, RR, BBD, PPC, BBA assisted with further drafting and revisions of the manuscript. All authors reviewed and approved the final version of the manuscript.
Funding
This study was funded by Wellcome Trust (310728/Z/24/Z) (RGW, RAC, KAT) RAC was also funded by Coefficient Giving (GV673606227), BMGF (INV-001754) and NIH (G-202303-69963, R-202309-71190) to LSHTM. RGW is also funded by the Wellcome Trust (218261/Z/19/Z), NIH (1R01AI147321-01, G-202303-69963, R-202309-71190), EDTCP (RIA208D-2505B), UK MRC (CCF17-7779 via SET Bloomsbury), ESRC (ES/P008011/1), BMGF (INV-004737, INV-035506), Coefficient Giving (GV673606227), and the WHO (2020/985800-0). This study was also financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance code 001. JVBB was also funded by CAPES scholarship (88881.126581/2025-01). This work was also supported by the National Council for Scientific and Technological Development (CNPq) [Grant: 445759/2023-2]; the Department of HIV/AIDS, Tuberculosis, Viral Hepatitis, and STDs (Secretariat of Health and Environmental Surveillance, Ministry of Health); and the Department of Science and Technology (Secretariat of Science, Technology, Innovation, and Health Complex, Ministry of Health).
Data availability
The dataset used for the present study are available upon reasonable request to the corresponding author.
Declarations
Ethics approval and consent to participate
This study was submitted, reviewed, and approved by the Research Ethics Committee of the Federal University of Mato Grosso do Sul (#75865823.7.1001.0021), the London School of Hygiene and Tropical Medicine Research Ethics Committee (#31229/RR/37148), and the Research Ethics Committee of Tropical Medicine Research Center of Rondônia (#88090125.4.0000.0011). Study participants provided written informed consent prior to study participation. All study procedures were performed in accordance with relevant local guidelines and regulations.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Ghislaine Gonçalez de Araujo Arcanjo and José Victor Bortolotto Bampi contributed equally to this work.
References
- 1.World Health Organization. Global tuberculosis report 2025. 1st ed. World Health Organization. 2025. https://www.who.int/teams/global-programme-on-tuberculosis-and-lung-health/tb-reports/global-tuberculosis-report-2025. Accessed 26 Feb 2026.
- 2.World Health Organization. Implementing the end TB strategy: the essentials, 2022 update. 1st ed. World Health Organization. 2022.https://www.who.int/publications/i/item/9789240065093. Accessed 26 Feb 2026.
- 3.Dye C, Bassili A, Bierrenbach A, Broekmans J, Chadha V, Glaziou P, et al. Measuring tuberculosis burden, trends, and the impact of control programmes. Lancet Infec Dis. 2008. 10.1016/S1473-3099(07)70291-8. [DOI] [PubMed] [Google Scholar]
- 4.Cords O, Martinez L, Warren JL, O’Marr JM, Walter KS, Cohen T, et al. Incidence and prevalence of tuberculosis in incarcerated populations: a systematic review and meta-analysis. Lancet Public Health. 2021. 10.1016/S2468-2667(21)00025-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Brazil. Boletim Epidemiológico Especial - Tuberculose em pessoas privadas de liberdade 2025. Departamento de HIV, Aids, Tuberculose, Hepatites Virais e Infecções Sexualmente Transmissíveis. 2025. https://www.gov.br/aids/pt-br/central-de-conteudo/boletins-epidemiologicos/2025/boletim_especial_pll_2025.pdf/view. Accessed 13 Feb 2026.
- 6.Sacchi FPC, Praça RM, Tatara MB, Simonsen V, Ferrazoli L, Croda MG, et al. Prisons as Reservoir for Community Transmission of Tuberculosis, Brazil. Emerg Infect Dis. 2015. 10.3201/eid2103.140896. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Liu YE, Mabene Y, Camelo S, Rueda ZV, Pelissari DM, Dockhorn Costa Johansen F, et al. Mass incarceration as a driver of the tuberculosis epidemic in Latin America and projected effects of policy alternatives: a mathematical modelling study. Lancet Public Health. 2024. 10.1016/S2468-2667(24)00192-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.World Health Organization. WHO consolidated guidelines on tuberculosis: module 2: screening: systematic screening for tuberculosis disease. World Health Organization. 2021. https://www.who.int/publications/i/item/9789240022676. Accessed 16 Feb 2025. [PubMed]
- 9.Brazil. Manual de recomendações para o controle da tuberculose no Brasil. 2. ed. Secretaria de Vigilância em Saúde. 2018. https://www.gov.br/saude/pt-br/centrais-de-conteudo/publicacoes/svsa/tuberculose/manual-de-recomendacoes-e-controle-da-tuberculose-no-brasil-2a-ed.pdf/view. Accessed 26 Feb 2026.
- 10.Lemos ACM, Matos ED, Bittencourt CN. Prevalence of active and latent TB among inmates in a prison hospital in Bahia, Brazil. J Bras Pneumol. 2009. 10.1590/s1806-37132009000100009. [DOI] [PubMed] [Google Scholar]
- 11.Valença MS, Scaini JLR, Abileira FS, Gonçalves CV, von Groll A, Silva PEA. Prevalence of tuberculosis in prisons: risk factors and molecular epidemiology. Int J Tuberc Lung Dis. 2015. 10.5588/ijtld.15.0126. [DOI] [PubMed] [Google Scholar]
- 12.Pivetta de Araujo RC, Martinez L, da Silva Santos A, Ferreira Lemos E, Dias de Oliveira R, Croda M, et al. Serial Mass Screening for Tuberculosis Among Incarcerated Persons in Brazil. Clin Infect Dis. 2024. 10.1093/cid/ciae055. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Abrahão RMCM, Nogueira PA, Malucelli MIC. Tuberculosis in county jail prisoners in the western sector of the city of São Paulo, Brazil. Int J Tuberc Lung Dis. 2006;10:203–8. [PubMed] [Google Scholar]
- 14.Carbone Ada, SS, Paião DSG, Sgarbi RVE, Lemos EF, Cazanti RF, Ota MM, et al. Active and latent tuberculosis in Brazilian correctional facilities: a cross-sectional study. BMC Infect Dis. 2015. 10.1186/s12879-015-0764-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kuhleis D, Ribeiro AW, Costa ERD, Cafrune PI, Schmid KB, Costa LL et al. da,. Tuberculosis in a southern Brazilian prison. Mem Inst Oswaldo Cruz. 2012; 10.1590/S0074-02762012000700012 [DOI] [PubMed]
- 16.Martinez L, Warren JL, Harries AD, Croda J, Espinal MA, Olarte RAL, et al. Global, regional, and national estimates of tuberculosis incidence and case detection among incarcerated individuals from 2000 to 2019: a systematic analysis. Lancet Public Health. 2023. 10.1016/S2468-2667(23)00097-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Rickman HM, Kamchedzera W, Schwalb A, Phiri MD, Ruhwald M, Shanaube K, et al. Know your tuberculosis epidemic-Is it time to add Mycobacterium tuberculosis immunoreactivity back into global surveillance? PLOS Glob Public Health. 2022. 10.1371/journal.pgph.0001208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Andrews JR, Charalambous S, Churchyard G, Cobelens F, Fernández-Escobar C, Frick M, et al. The participation of people deprived of liberty in tuberculosis vaccine trials: should they be protected from research, or through research? Lancet Infec Dis. 2025. 10.1016/S1473-3099(25)00305-6. [DOI] [PubMed] [Google Scholar]
- 19.Bampi JVB, Arcanjo GG de A, Santo KM, Ventura MS, Clark RA, Thomas KA et al. Acceptability of a proposed new tuberculosis vaccine among people deprived of liberty in Brazil. Vaccine. 2026; 10;1016/j.vaccine.2026.128492. 10.1016/j.vaccine.2026.128492. [DOI] [PubMed]
- 20.Pires P MCCF, Liu YE, Lemos EF, Silva LF da, Croda MG, Magalhães M, et al. Perceptions of persons deprived of liberty regarding tuberculosis vaccine research. PLOS Glob Public Health. 2025. 10.1371/journal.pgph.0004941. [DOI] [PMC free article] [PubMed]
- 21.Conselho Nacional de Justiça. CNIEP - Estatísticas de Inspeções Prisionais. Conselho Nacional de Justiça. 2026. https://geopresidios.cnj.jus.br/. Accessed 6 Feb 2026.
- 22.de Oliveira RD, Santos Ada, Reis S, Leite CB, de Sacchi A, de Araujo FPC. Primary Prophylaxis to Prevent Tuberculosis Infection in Prison Inmates: A Randomized, Double-Blind, Placebo-Controlled Trial. Am J Trop Med Hyg. 2020. 10.4269/ajtmh.20-0110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing. (2026). https://www.r-project.org/. Accessed 26 Feb 2026.
- 24.Brazil. Boletim Epidemiológico - HIV e Aids. (2024). Departamento de HIV, Aids, Tuberculose, Hepatites Virais e Infecções Sexualmente Transmissíveis. 2024. https://www.gov.br/aids/pt-br/central-de-conteudo/boletins-epidemiologicos/2024/boletim_hiv_aids_2024e.pdf/view. Accessed 23 Feb 2025.
- 25.Ferreira MM, Ferrazoli L, Palaci M, Salles PS, Medeiros LA, Novoa P, et al. Tuberculosis and HIV infection among female inmates in São Paulo, Brazil: a prospective cohort study. J Acquir Immune Defic Syndr Hum Retrovirol. 1996. 10.1097/00042560-199610010-00009. [DOI] [PubMed] [Google Scholar]
- 26.Estevan AO, Oliveira SM do VL de, Croda J. Active and latent tuberculosis in prisoners in the Central-West Region of Brazil. Rev Soc Bras Med Trop. 2013. 10.1590/0037-8682-1441-2013. [DOI] [PubMed]
- 27.Nogueira PA, Abrahão RMC, de Galesi M. Tuberculosis and latent tuberculosis in prison inmates. Rev Saude Publica. 2012. 10.1590/s0034-89102011005000080. [DOI] [PubMed] [Google Scholar]
- 28.Dowdy DW, Behr MA. Are we underestimating the annual risk of infection with Mycobacterium tuberculosis in high-burden settings? Lancet Infec Dis. 2022. 10.1016/S1473-3099(22)00153-0. [DOI] [PubMed] [Google Scholar]
- 29.Houben RMGJ, Lalli M, Kranzer K, Menzies NA, Schumacher SG, Dowdy DW. What if They Don’t Have Tuberculosis? The Consequences and Trade-offs Involved in False-positive Diagnoses of Tuberculosis. Clin Infect Dis. 2019. 10.1093/cid/ciy544. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Houben RMGJ, Veeken LD, Schwalb A, Grint DJ, Wasserman S, van Crevel R, et al. Do no harm - re-evaluating the risks of overtreatment in community-wide tuberculosis screening. medRxiv. 2025. 10.1101/2025.09.18.25335917. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Liu YE, Bampi JVB, Arthur RF, Salindri AD, Busatto C, Jiménez PA, et al. Comparative impacts and cost-effectiveness of tuberculosis active case-finding strategies in prisons in Brazil, Colombia, and Peru: a mathematical modeling study. medRxiv. 2025. 10.1101/2025.09.25.25336625. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Akalu TY, Clements ACA, Liyew AM, Gilmour B, Murray MB, Alene KA. Risk factors associated with post-tuberculosis sequelae: a systematic review and meta-analysis. eClinicalMedicine. 2024. 10.1016/j.eclinm.2024.102898. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Narayan A, Salindri AD, Keshavjee S, Muyoyeta M, Velen K, Rueda ZV, et al. Prioritizing persons deprived of liberty in global guidelines for tuberculosis preventive treatment. PLoS Med. 2023. 10.1371/journal.pmed.1004288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Mabud TS, Alves M, de LD, 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. 10.1371/journal.pmed.1002737. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Nava-Aguilera E, Andersson N, Harris E, Mitchell S, Hamel C, Shea B, et al. Risk factors associated with recent transmission of tuberculosis: systematic review and meta-analysis. Int J Tuberc Lung Dis. 2009;13:17–26. [PubMed] [Google Scholar]
- 36.Paião DSG, Lemos EF, Carbone A da, Sgarbi SS, Junior RVE, da Silva AL. Impact of mass-screening on tuberculosis incidence in a prospective cohort of Brazilian prisoners. BMC Infect Dis. 2016;16:533. 10.1186/s12879-016-1868-5. [DOI] [PMC free article] [PubMed]
- 37.Brazil. Boletim Epidemiológico de Tuberculose - Número Especial | Mar. 2024. Ministério da Saúde. 2025. https://www.gov.br/saude/pt-br/centrais-de-conteudo/publicacoes/boletins/epidemiologicos/especiais/2024/boletim-epidemiologico-de-tuberculose-numero-especial-mar-2024.pdf/view. Accessed 18 Mar 2025.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The dataset used for the present study are available upon reasonable request to the corresponding author.