Abstract
Introduction
Latent tuberculosis infection (LTBI) constitutes a significant reservoir for future tuberculosis (TB) cases, particularly among incarcerated populations, which experience a disproportionately elevated burden. Recognizing modifiable risk factors in correctional facilities is crucial for formulating effective public health strategies to mitigate TB spread.
Methods
This case-control research examined 1,252 incarcerated inmates in Shenzhen, China. Subjects (n = 215) were interferon-gamma release assay (IGRA)-positive, whereas controls (n = 1,037) were IGRA-negative. Data were obtained from computerized medical records. Multivariable logistic regression discerned independent risk variables. A bootstrap mediation study (5,000 iterations) was conducted to investigate potential routes.
Results
Four independent factors of LTBI were identified: present smoking (adjusted odds ratio = 1.57), absence of a bacillus Calmette-Guerin (BCG) scar (aOR = 6.08), initial incarceration (aOR = 2.16), and absence of medical insurance (aOR = 0.05). Mediation analysis indicated that smoking behavior largely mediated the connections between insurance status and first incarceration with LTBI risk, underscoring the interrelated behavioral and structural pathways.
Conclusion
The results indicate that LTBI in correctional facilities is influenced by a confluence of biological, behavioral, and socioeconomic factors. An integrated preventative plan is essential, encompassing mandatory entry screening, verification of BCG vaccination, institutional smoking cessation programs, and measures to guarantee healthcare access. Future longitudinal studies ought to validate these mediation mechanisms and evaluate their influence on the progression to active illness.
Supplementary Information
The online version contains supplementary material available at 10.1007/s44197-026-00527-w.
Keywords: Latent tuberculosis, Interferon-gamma release assay, Risk factors, Infectious disease, Incarcerated populations
Introduction
Tuberculosis (TB) is a significant infectious disease endangering human health. In 2024, the projected incidence of new TB infections globally was 10.7 million, resulting in 1.23 million fatalities [1, 2]. TB is significantly widespread among incarcerated populations, with incidence rates reported to be ten to several dozen times greater than those in the general community [3]. Prison-acquired TB endangers the health of incarcerated populations and provides a risk to the wider community through transmission by prison officials, visitors, and untreated released individuals, thereby obstructing TB control efforts at the population level [4]. The World Health Organization (WHO) estimates that around one-quarter of the global population is asymptomatically infected with mycobacterium tuberculosis (MTB), a condition referred to as latent tuberculosis infection (LTBI) [5].
Individuals with LTBI face a markedly elevated risk of progressing to active TB [6, 7]. Administering tuberculosis preventative treatment (TPT) to persons with LTBI and ensuring adherence to the complete treatment schedule can effectively avert progression to active TB [8]. The WHO end TB Strategy recommends testing for latent MTB infection and providing preventive treatment for populations at high risk of developing TB, including incarcerated populations settings [9, 10]. The WHO guidelines recommend interferon-gamma release assay (IGRA) and tuberculin skin test (TST) as methods for detecting latent MTB infection [11]. Compared to conventional diagnostic methods, IGRA demonstrates significantly higher sensitivity and specificity [12].
Recent investigations have indicated a significant frequency of LTBI within incarcerated populations [13, 14]. Nonetheless, comprehensive investigation of treatment outcomes and the determinants affecting therapy efficacy is still scarce. A thorough comprehension of LTBI-associated risk factors in correctional facilities is crucial for formulating effective prevention and management measures.
This study seeks to clarify the intricate connections between known risk factors and the onset of LTBI using a thorough, multidimensional examination of medical records. The aim is to offer novel insights and pinpoint potential intervention targets for the prevention and management of LTBI. The initiative seeks to augment preventive and therapeutic strategies to reduce TB incidence and increase outcomes and quality of life for incarcerated populations.
Methods
Study Design and Study Population
Shenzhen Bao’an District People’s Hospital is the exclusive institution tasked with executing disease prevention and control screenings and health evaluations for all incarcerated populations in Bao’an District, China. We can centrally gather and systematize these cases for processing. This case-control study entailed the examination of electronic medical record databases at Shenzhen Bao’an District People’s Hospital to identify individual files with LTBI and subsequent monitoring of TB incidence. Patient records were gathered from July 2021 to November 2025.
IGRA Application and Interpretation
Individual files were categorized as IGRA-positive or IGRA-negative based on the interferon (IFN) -γ level response to MTB-specific antigens. IGRA positivity was determined according to known assay thresholds and utilized to ascertain LTBI status in this investigation. Considering that IGRA results can be influenced by the host’s immunological status, negative results were read with caution, and IGRA findings were utilized exclusively for classification rather than for conclusive exclusion of infection.
Outcome Measures
Depending on the IGRA results, the subjects divided into two groups: the case group (LTBI) and the control group (non-LTBI). Exclusion Criteria: Individual files with a incarcerated history of TB or active TB, along with those who have undergone treatment for LTBI or are presently undergoing anti-TB therapy; Patients with substantial systemic comorbidities, such as diabetes and HIV infection; Individual files of discharged TB patients within two years (for whom ongoing TB surveillance and follow-up assessments are unfeasible); Smoking was defined as the current usage of cigarettes at the time of data collection, while former smokers were categorized as non-smokers. A comparative examination of clinical data between the two groups was then performed. The exclusion technique is illustrated in Fig. 1.
Fig. 1.
Flowchart of exclusion process. Abbreviations: TB, tuberculosis; IGRA, interferon-gamma release assay; LTBI, latent tuberculosis infection
Sampling
In the prevalent estimation approach for calculating sample size in multivariable logistic regression, the sample size for the outcome category with the lesser proportion must be no less than 10–12 times the number of independent variables. This study examines a dependent variable with two categories (LTBI and non-LTBI), and initial estimates indicate 20 relevant independent variables. Consequently, the sample size for the case group in this study is roughly 200 instances, calculated as 10 × 20. After examining the literature, the estimated prevalence of latent tuberculosis infection (LTBI) ranges from 26% to 40% [15, 16]. Consequently, the necessary sample size is a minimum of 770 cases, calculated as 200 divided by 26%.
Data Collection
Expert researchers conducted data collection by extracting patients’ medical records in strict accordance with the inclusion and exclusion criteria, and they did not engage in further statistical analysis. To avert data duplication from the same patient, collecting data exclusively from the original hospital admission, notwithstanding the existence of supplementary medical examination records obtained during detention.
Data Analysis
Measurement data were expressed as mean ± SD (standard deviation) when normally distributed, and differences between groups were compared using the independent samples t-test or analysis of variance (ANOVA). For non-normally distributed data, results were presented as median (interquartile range), and differences between groups were assessed using the rank sum test. Categorical data were expressed as frequencies and percentages (%), with group differences assessed using the χ2 test. Subsequently, we performed binary logistic regression analysis utilizing relevant variables from the χ2 test to discover statistically significant predictors. Mediation analysis was conducted via a bootstrap method with 5,000 iterations to determine indirect effects. Indirect, direct, and total effects were calculated based on standard path coefficients (a × b). Significance was defined as a 95% bootstrap confidence interval excluding zero. Data were analyzed using the SPSS version 20.0 statistical software package (SPSS Inc., Chicago, IL, USA).
Results
Incarcerated Population Characteristics
After applying our inclusion and exclusion criteria, the final data analysis encompassed 1252 individuals. The exclusion technique is illustrated in Fig. 1. The study comprised 1097 male patients and 155 female patients, aged between 29 and 53 years, with a median age of 41 years. Of the 1252 individuals incarcerated, 215 developed LTBI, reflecting a prevalence of 17.17% among the incarcerated population.
Comparison of Relevant Factors
This study thoroughly analyzed 14 distinct parameters, including 7 demographic characteristics and 7 medical history and clinical features. A comparative analysis comparing LTBI and non-LTBI groups indicated that 4 out of 14 candidate variables exhibited significant relationships with LTBI development in incarcerated populations (P < 0.05): insurance type, first incarceration, mycobacterium bovis bacillus Calmette-Guerin (BCG) scar, and cigarette smoking use.
Demographic studies indicated substantial correlations between LTBI and insurance type (χ2 = 14.35, P = 0.0002). Regarding medical history and clinical features among incarcerated populations, those with LTBI exhibited a significant prevalence of prior incarceration (χ2 = 5.82, P = 0.016). Correspondingly, they exhibited a reduced proportion of individuals with BCG vaccination scars (χ2 = 15.89, P < 0.0001) and an elevated proportion of smokers (χ2 = 8.77, P = 0.003) (Table 1).
Table 1.
Comparison of demographic parameters, medical history, and clinical aspects between LTBI and non-LTBI groups among incarcerated populations
| Variables | Total (n = 1252) |
non-LTBI (n = 1037) | LTBI (n = 215) |
Statistic | P-value |
|---|---|---|---|---|---|
| Demographic data | |||||
| Age, M (Q1, Q3), years | 41.00 (29.00,53.00) | 41.00 (28.00,52.00) | 42.00 (30.00,53.00) | Z = − 0.89 | 0.373 |
| Gender, n (%) | χ2 = 2.27 | 0.132 | |||
| Male | 1097 (87.62) | 902 (86.98) | 195 (90.70) | ||
| Female | 155 (12.38) | 135 (13.02) | 20 (9.30) | ||
| BMI (kg/cm2), n (%) | χ2 = 3.30 | 0.348 | |||
| 18.5–25 | 985 (78.67) | 808 (77.92) | 177 (82.33) | ||
| < 18.5 | 106 (8.47) | 93 (8.97) | 13 (6.05) | ||
| 25–30 | 147 (11.74) | 123 (11.86) | 24 (11.16) | ||
| ≥ 30 | 14 (1.12) | 13 (1.25) | 1 (0.46) | ||
| Education level, n (%) | χ2 = 2.56 | 0.464 | |||
| Primary school | 297 (23.72) | 249 (24.01) | 48 (22.33) | ||
| Middle school | 366 (29.23) | 310 (29.89) | 56 (26.05) | ||
| Senior high school | 410 (32.75) | 335 (32.31) | 75 (34.88) | ||
| University and above | 179 (14.30) | 143 (13.79) | 36 (16.74) | ||
| Marital status, n (%) | χ2 = 1.45 | 0.484 | |||
| Unmarried | 230 (18.37) | 194 (18.71) | 36 (16.74) | ||
| Married | 947 (75.64) | 778 (75.02) | 169 (78.61) | ||
| Divorced | 75 (5.99) | 65 (6.27) | 10 (4.65) | ||
| Occupation, n (%) | χ2 = 1.49 | 0.222 | |||
| Unemployed | 570 (45.53) | 464 (44.74) | 106 (49.30) | ||
| Employed | 682 (54.47) | 573 (55.26) | 109 (50.70) | ||
| Insurance type, n (%) | χ2 = 14.35 | 0.0002* | |||
| Self-Pay | 986 (78.75) | 796 (76.76) | 190 (88.37) | ||
| Insured | 266 (21.25) | 241 (23.24) | 25 (11.63) | ||
| Medical background and clinical features | |||||
| Residence, n (%) | χ2 = 3.58 | 0.059 | |||
| Local | 849 (67.81) | 715 (68.95) | 134 (62.33) | ||
| Migrant | 403 (32.19) | 322 (31.05) | 81 (37.67) | ||
| First incarceration, n (%) | χ2 = 5.82 | 0.016* | |||
| No | 1068 (85.30) | 896 (86.40) | 172 (80.00) | ||
| Yes | 184 (14.70) | 141 (13.60) | 43 (20.00) | ||
| Contact history, n (%) | χ2 = 0.54 | 0.461 | |||
| No | 1074 (85.78) | 893 (86.11) | 181 (84.19) | ||
| Yes | 178 (14.22) | 144 (13.89) | 34 (15.81) | ||
| Concurrent condition, n (%) | χ2 = 3.06 | 0.080 | |||
| No | 971 (77.56) | 814 (78.50) | 157 (73.02) | ||
| Yes | 281 (22.44) | 223 (21.50) | 58 (26.98) | ||
| Mycobacterium bovis BCG scar, n (%) | χ2 = 15.89 | < 0.0001* | |||
| No | 881 (70.37) | 754 (72.71) | 127 (59.07) | ||
| Yes | 371 (29.63) | 283 (27.29) | 88 (40.93) | ||
| Cigarettes, n (%) | χ2 = 8.77 | 0.003* | |||
| No | 503 (40.18) | 436 (42.04) | 67 (31.16) | ||
| Yes | 749 (59.82) | 601 (57.96) | 148 (68.84) | ||
| Alcohol, n (%) | χ2 = 0.40 | 0.526 | |||
| No | 512 (40.89) | 358 (82.74) | 154 (80.93) | ||
| Yes | 740 (59.11) | 679 (17.26) | 61 (19.07) | ||
Note: Categorical data are presented as frequencies (%), while continuous data are expressed as median (quartile); Z, Mann–Whitney test; χ2, Chi-square test; -, Fisher’s exact test; *, p-value of < 0.05 was considered statistically significant. LTBI, latent tuberculosis infection; BMI, Body mass index; BCG, bacillus Calmette-Guerin
Logistic Regression Analysis Results
Factors deemed statistically significant in univariate analysis were incorporated into multivariate logistic regression. A notable correlation was identified between cigarettes [OR (95% CI) 1.571 (1.104 ~ 2.237)], BCG scar [OR (95% CI) 6.081 (3.732 ~ 9.915)], First incarceration [OR (95% CI) 2.163 (1.149 ~ 4.071)], and insurance type [OR (95% CI) 0.050 (0.026 ~ 0.094)] with LTBI. Figure 2 and Table S1 presents specific statistics.
Fig. 2.
The forest plot of associated factors for LTBI by multiple logistic regression. Variables with P < 0.05 and those deemed clinically significant were incorporated into the multiple logistic regression analysis. From the multiple logistic regression analysis, we computed adjusted odds ratios (aORs) and their corresponding 95% confidence intervals (CIs) to evaluate the relationship between incarcerated population and LTBI, controlling for potential variables. A p-value of < 0.05 was considered statistically significant
Mediation Analysis Results
To further investigate potential collinearity in the model, we conducted a mediation analysis to assess the mediating influence of smoking behavior. Smoking behavior partially buffers the influence of insurance type and first incarceration experience on LTBI. Specifically, the pathway through which insurance type indirectly increased LTBI risk via elevated smoking exposure was significant (indirect effect = 0.054, p < 0.001, 95% CI [0.029, 0.080]). Likewise, the indirect effect associated with first incarceration was significant (indirect effect = 0.022, p = 0.0027, 95% CI [0.003, 0.041]). In contrast, the mediating effect of smoking on the association between BCG scar and LTBI was not significant (indirect effect = 0.016, p = 0.213, 95% CI [–0.009, 0.041]) (Table 2 and Table S2).
Table 2.
Analysis results mediated by cigarettes
| Pathway | c (Overall effect) | a×b (indirect effect) | p-value | 95% Boot CI |
|---|---|---|---|---|
| Insurance type → Cigarettes → LTBI | −0.526** | 0.054** | p < 0.001** | (0.029, 0.080) |
| First incarceration → Cigarettes → LTBI | 0.090* | 0.022* | 0.027* | (0.003, 0.041) |
| BCG scar → Cigarettes → LTBI | 0.426** | 0.016 | 0.213 | (−0.009, 0.041) |
Note: *, p < 0.05, **, p < 0.01; Bootstrap resampling count = 5000; Confidence interval method = Percentile Bootstrap method
Discussion
This study seeks to investigate the risk factors associated with LTBI. Utilizing univariate analysis and logistic regression techniques, we identified four significant risk factors: cigarette smoking, BCG scar, first incarceration, and insurance type.
This study found that the absence of a Mycobacterium bovis BCG scar was significantly associated with an increased risk of LTBI among incarcerated populations. BCG vaccination not only inhibits the progression of LTBI to active TB but also offers protection against initial infection [17]. This indicates that the immune systems activated by BCG are essential for developing host defenses, and the existence of a BCG scar may act as a marker of this immunological response. The enhanced immune elements produced by vaccination, including innate immunity and adaptive immune responses, may be related to this protective effect [18]. Prior research has evidenced the preventive efficacy of BCG immunization against LTBI, indicating that BCG diminishes the risk of latent infection in persons with end-stage renal illness [19]. This discovery corresponds with extensive evidence indicating that BCG-induced immune training improves resistance to latent infections, with the appearance of a BCG scar potentially acting as a signal for this protection. The BCG vaccination is a crucial element of TB control measures, and the existence of a BCG scar may act as an indirect signal of immunological protection against latent infection.
The results demonstrated that insurance coverage was markedly correlated with a reduced incidence of LTBI, exhibiting an odds ratio (OR) of 0.05 (95% CI: 0.03–0.09). The statistics imply that insured persons had an almost 95% reduced chance of infection, indicating that health insurance may serve a protective function by enabling timely TB prevention and screening. This observation aligns with results from prior population-based longitudinal studies [20]. In various Asian countries, insurance coverage is essential to national TB control programs. In South Korea, advancements in TB prevention and treatment have been partially ascribed to insurance-funded programs, encompassing extensive TB management and contact tracing efforts [21]. These insurance-supported initiatives enhance the identification and management of both active and latent infections, therefore contributing to the reduction of the overall TB burden. Economic factors regarding LTBI screening and treatment are also significant. Cost-effectiveness analyses of several LTBI screening strategies under Thailand’s Universal Coverage Scheme (UCS) indicated that the substantial expense of IGRA restricts its implementation in routine screening under insurance programs [22]. This signifies that insurance coverage decisions directly affect the choice and execution of LTBI screening tests, thereby influencing the identification and management of LTBI in high-risk populations. Health insurance coverage profoundly impacts all facets of LTBI management, encompassing screening protocols, treatment safety oversight, and program execution [23]. The health insurance system is a crucial factor in the proper management of LTBI, influencing policy formulation and clinical decision-making within healthcare organizations. Insurance status may act as a proxy indication for wider socioeconomic factors, such as income stability, employment history, and access to medical care, rather than serving as a direct cause factor. Secondly, we recognize the potential for residual confounding, as certain characteristics associated with socioeconomic status, healthcare-seeking behavior, and comorbidities may not be comprehensively represented in the current data. In a constrained population, access to medical treatments and documentation may vary consistently based on insurance status, resulting in selection bias.
Conversely, first incarceration was positively correlated with LTBI risk, indicating that primary exposure to correctional settings may heighten infection risk via increased environmental contact or stress-induced immunological mechanisms. Cross-sectional studies indicate that screening and preventive measures in correctional facilities can diminish the likelihood of LTBI progressing to active TB, particularly in prison environments where early diagnosis and intervention are more feasible [24, 25]. The research revealed that lung immune cells, including alveolar macrophages and T cells, exhibit distinct DNA methylation patterns that can forecast the transition of individuals with elevated exposure risk to LTBI, indicating that epigenetic modifications may function as biomarkers for identifying high-risk LTBI populations [26, 27]. In the setting of first incarceration, detained populations provide a critical demographic for LTBI screening and intervention. The significant incidence noted in correctional facilities, along with molecular and epigenetic research, indicates the feasibility of implementing targeted preventive strategies and creating predictive biomarkers to identify persons susceptible to progressing from latent to active TB. These findings offer preliminary information that could guide future research on the creation of biomarkers, especially in incarcerated populations where transmission dynamics are intricate and options for focused intervention are essential.
Smoking increases the risk of LTBI, with smokers demonstrating an infection rate around 57% higher. This discovery corroborates current data that smoking may has immunosuppressive effects on the respiratory system [28]. Smoking influences the likelihood of getting active TB and may also correlate with illness severity, thereby impacting transmission dynamics and disease treatment of TB [29]. The possible mechanisms linking smoking and TB have been investigated concerning immunological regulation and pulmonary disease. While the literature lacks specific mechanistic research, the observed correlations indicate that smoking may compromise lung defenses, thereby heightening the likelihood of latent infection and the progression to active TB. This aligns with the prevailing notion that smoking diminishes immune responses and facilitates the onset and reactivation of LTBI [30]. We performed additional mediation analysis on cigarette smoking, revealing that cigarette smoking partially mediated the influence of insurance type and first incarceration on LTBI. Our findings indicate that insufficient healthcare may intensify smoking behaviors, consequently increasing the risk of infection. Individuals covered by certain insurance types may experience higher smoking prevalence due to underlying socioeconomic disadvantage, occupational stress, or differential health-related behaviors, which in turn increases susceptibility to MTB infection by impairing respiratory immune defenses; likewise, the indirect impact of initial incarceration was also notable, implying that stress or lifestyle alterations within the prison environment may indirectly facilitate LTBI by encouraging smoking. Current research indicates that smoking heightens vulnerability to MTB infection and is independently correlated with LTBI [31]. Simultaneously, individuals with a history of incarceration exhibit both an elevated burden of LTBI and a greater prevalence of smoking [32]. Furthermore, the underinsured and uninsured exhibit elevated smoking rates and diminished access to smoking cessation programs and LTBI screening and treatment. While direct mediation analyses quantifying the indirect pathway from incarceration or insurance type to LTBI via cigarette smoking were not conducted in this study, existing mechanistic and epidemiological evidence substantiates cigarette smoking as a plausible mediator connecting social determinants to LTBI risk, a hypothesis that necessitates further validation in future research [33].
This study, utilizing a comprehensive single-center real-world electronic medical record database, systematically examined LTBI among the incarcerated populations in the Pearl River Delta region of China, identifying risk factors and elucidating the indirect influence of social health determinants on LTBI through smoking behavior. Recognized risk factors for TB, including confinement, overcrowding, and inadequate nutritional status, are acknowledged as key contributors to transmission in correctional environments. These structural variables establish conditions that enable prolonged exposure to MTB. In this perspective, the current findings do not undermine the significance of classical risk factors but rather enhance them by emphasizing the potential role of behavioral and social determinants, such as smoking exposure and insurance-related markers of social vulnerability. Collectively, our findings highlight the multifaceted nature of LTBI risk and indicate that effective preventative interventions may necessitate integrated methods that consider both structural factors and individual-level causes. Furthermore, we discovered quantitative evidence indicating that BCG scars serve as a protective factor against LTBI, corroborating the study of trained immunity mechanisms and offering empirical support within the incarcerated populations, which holds significant additional significance. Health insurance, as a socioeconomic variable, has garnered less focus in TB research. This study incorporated health insurance coverage into the analytical model and found that it was significantly associated with a lower risk of LTBI, suggesting a potential protective role that warrants further investigation.
Notwithstanding its numerous strengths, this study also has some limitations. A case-control study is essentially structured to identify connections between variables, making it challenging to demonstrate causality. The relationship between risk variables and LTBI cannot be entirely elucidated as causal, particularly the mediation role of insurance type and cigarette smoking, which requires additional validation through prospective cohort studies. The study sample was derived from a singular location and a specific prison system, and the external promotion was restricted, perhaps influencing the generalizability of the risk factors. Several significant high-risk indicators are absent, including detention duration, cell overcrowding, ventilation conditions, dietary state, and mental stress levels. We acknowledge that stress linked to depression and anxiety may influence immune function and health-related habits, such as smoking and drinking, and could therefore act as a potential confounder in the observed relationships. The absence of direct stress evaluation may have led to residual confounding. These characteristics may represent the principal exposures of LTBI in prisons; however, they cannot be incorporated into the model due to the constraints of electronic medical records. Subsequent research should incorporate standardised evaluations of psychological stress, depression, and anxiety to clarify the psychosocial mechanisms behind the identified associations. The mediation analysis demonstrated significant effects; however, it lacks sensitivity analysis and validation of alternative models for potential confounding variables. Due to the retrospective case-control design, these assumptions cannot be completely verified. Consequently, the identified indirect effects should be regarded as exploratory results, necessitating additional evaluation of the model’s robustness through prospective or longitudinal research. The present findings solely indicate the likelihood of infection, and they are unable to evaluate whether these risk factors influence illness progression. In the future, we will monitor the follow-up data of LTBI that eventually progresses to active TB and further validate it in a larger-scale, multi-center prospective study.
Conclusions
This study identifies and characterizes four independent determinants—cigarette smoking, lack of a BCG scar, first incarceration, and absence of medical insurance—for LTBI in incarcerated populations. Mediation analysis reveals that smoking partially mediates the effects of insurance status and first incarceration on LTBI risk, highlighting interconnected behavioral and structural pathways. The findings underscore the necessity for integrated entry screening (integrated screening refers to the routine inclusion of LTBI screening as part of standard health assessments conducted when individuals access healthcare services within the facility), BCG vaccination verification, smoking cessation support, and equitable healthcare access in correctional facilities. Future longitudinal studies are needed to confirm these mediating pathways and assess their impact on disease progression.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The authors express gratitude to all members of the study team for their collaboration and assistance. We express our gratitude to the physicians and nursing personnel of Bao’an District People’s Hospital in Shenzhen, China.
Author contributions
Conception and design: Z.L., Y.C. and A.H.; Methodology: Z.L., A.H.; Data collection: Y.C., J.X. and Z.D.; Data analysis and interpretation: Z.L., J.X. and Z.D.; Manuscript writing: all authors; Reviewing and editing the manuscript: all authors; Supervision: A.H.; Fund financial support: A.H.; Final approval of manuscript: all authors. All authors have read and agreed to the published version of the manuscript.
Funding
This work was funded by Shenzhen Medical Research FundThis work was funded by Shenzhen Medical Research Fund (No. B2303003).
Data Availability
Data are accessible upon a reasonable request. The datasets produced and/or examined in this study can be obtained from the corresponding author upon a reasonable request.
Declarations
Ethical Approval
The study was approved by the Ethics Committee of the Shenzhen Bao’an District People’s Hospital (Approval No. BYL20250218). Our research adheres strictly to the ethical principles outlined in the Declaration of Helsinki. This study constitutes a retrospective review of existing medical records, utilizing data extracted from incarcerated populations’ clinical files, usually produced during conventional medical procedures. No direct contact with the subjects occurred during this investigation. We guaranteed that access to the raw data was restricted solely to approved study team members. All incarceration identity data has been eliminated to safeguard patient confidentiality.
Consent to Participate
Not applicable.
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.
Zhenyang Liu and Yawei Cui contributed equally to this work.
References
- 1.Digby F W, Amy K B, Maximiliano G G, Valerie M. Mycobacterium tuberculosis biology, pathogenicity and interaction with the host. Nat Rev Microbiol. 2025;23(12). [DOI] [PMC free article] [PubMed]
- 2.Gregory P B, Brian A, Anthony B, Gunar G, Celso K, Neelima N, et al. Post-tuberculosis lung disease: a case definition for use in research studies. Lancet Infect Dis. 2025(0). [DOI] [PubMed]
- 3.Nor Zam Azihan MH, Asmah R, Mohd Shaiful Jefri MNSK, Farhana A. A symptomatic approach to tuberculosis screening for high-risk groups in Malaysia: cost-effectiveness and budget impact analysis. J Clin Tuberc Other Mycobact Dis. 2022;29(0). [DOI] [PMC free article] [PubMed]
- 4.Reiko M, Pundharika P, Boonchai C, Worarat I, Supalert N, Hideki Y, et al. Risk for prison-to-community tuberculosis transmission, Thailand, 2017–2020. Emerg Infect Dis. 2023;29(3). [DOI] [PMC free article] [PubMed]
- 5.Mariko H, Hayato T, Tomotada I, Aki T, Atsushi H, Takahiro T, et al. Higher genome mutation rates of Beijing lineage of Mycobacterium tuberculosis during human infection. Sci Rep. 2020;10(1). [DOI] [PMC free article] [PubMed]
- 6.Priya B S, Amy S T, Jennifer F. Screening for latent tuberculosis infection among non-US-born adults in the US: a path toward elimination. JAMA. 2023;329(17). [DOI] [PubMed]
- 7.Randy M S, Monica P, Lana D, David M, Zainab G, Jonathan W, et al. Tuberculosis outbreak in a state prison system - Washington, 2021–2022. MMWR Morb Mortal Wkly Rep. 2023;72(12). [DOI] [PMC free article] [PubMed]
- 8.Anca V, Leonardo M, Rishi K G, Yohhei H, Tara N, Alexander K, et al. Tuberculosis prevention: current strategies and future directions. Clin Microbiol Infect. 2023;30(9). [DOI] [PMC free article] [PubMed]
- 9.Saskia J, Melissa M, Caryn U, Brian A, Andreas H D. Tuberculosis: an update for the clinician. Respirology. 2025;30(3). [DOI] [PMC free article] [PubMed]
- 10.Andrei M, Platon E. Drug-resistant tuberculosis threatens WHO’s End-TB strategy. Lancet Infect Dis. 2017;17(7). [DOI] [PubMed]
- 11.Qing-Hua G, Hong-Bo C, Yun H, Shun-Li C, Li-Ping C, Yong-Ning Y, et al. Predictive value of interferon-gamma release assays and tuberculin skin test for latent tuberculosis infection: a systematic review and meta-analysis of head-to-head comparative tests. J Microbiol Immunol Infect. 2025(0). [DOI] [PubMed]
- 12.Rishi K G, Marc L, Charlotte J, Alice J S, Jo S, Francis D, et al. Quantitative IFN-γ release assay and tuberculin skin test results to predict incident tuberculosis. A prospective cohort study. Am J Respir Crit Care Med. 2019;201(8). [DOI] [PMC free article] [PubMed]
- 13.Jingxian N, Peng L, Yuchen P, Yilin L, Yu Z, Wenxin J, et al. The impact of body mass index on latent tuberculosis infection: combined assessment in people living with HIV. Pathogens. 2025;14(11). [DOI] [PMC free article] [PubMed]
- 14.Xinru F, Shanshan W, Zhan W, Xinsong H, Cheng C, Limei Z, et al. Diagnostic accuracy of the ESAT6-CFP10 skin test for latent tuberculosis infection among jail detainees. Microbiol Spectr. 2025;13(10). [DOI] [PMC free article] [PubMed]
- 15.Kenneth J R, Stephan L, Susan T S. The reporting odds ratio and its advantages over the proportional reporting ratio. Pharmacoepidemiol Drug Saf. 2004;13(8). [DOI] [PubMed]
- 16.Gahee K, Gyuri P, Yujin K, Hyeran J, Young-Joon P. [Results of the 2023 screening project for latent tuberculosis infection among temporary day laborers in care facilities]. Jugan Geongang Gwa Jilbyeong. 2025;18(7). [DOI] [PMC free article] [PubMed]
- 17.Cynthia B E C, Randall R, Ying Z, Robert B. Latent tuberculosis infection: opportunities and challenges. Respirology. 2018;23(10). [DOI] [PubMed]
- 18.Anthea L K, Charlotte J, Jo S, Rishi K G, Francis D, Ajit L, et al. Effectiveness of BCG vaccination against mycobacterium tuberculosis infection in adults: a cross-sectional analysis of a UK-based cohort. J Infect Dis. 2019;221(1). [DOI] [PubMed]
- 19.Ping-Huai W, Shu-Yung L, Hung-Hsiang L, Chien-Chia C, Chin-Chung S, Chih-Yuan L, et al. Protective effect of BCG and neutrophil-to-lymphocyte ratio on latent tuberculosis in end stage renal disease. Infect Dis Ther. 2023;12(7). [DOI] [PMC free article] [PubMed]
- 20.Arisa I, Jun T, Hayato Y, Takehiro S, Takashi Y, Yasuki K. Tuberculosis screening and management of latent tuberculosis infection prior to biologic treatment in patients with immune-mediated inflammatory diseases: a longitudinal population-based analysis using claims data. Health Sci Rep. 2020;3(4). [DOI] [PMC free article] [PubMed]
- 21.Global, regional, and national burden of injuries, and burden attributable to injuries risk factors, 1990 to 2019: results from the Global Burden of Disease study 2019. Public Health. 2024;237(0). [DOI] [PubMed]
- 22.Panida Y, Olivia W, Jiraphun J, Montarat T, Sitaporn Y, Naiyana P, et al. Economic evaluation of diagnosis and treatment for latent tuberculosis infection among contacts of pulmonary tuberculosis patients in Thailand. Sci Rep. 2024;14(1). [DOI] [PMC free article] [PubMed]
- 23.Ye-Jee K, Jiyeon K, Jiwon L, Tae Sun S, Sang Hyoung P, Kyung-Wook J. Latent tuberculosis infection screening in patients with inflammatory bowel disease: a nationwide retrospective cohort study in South Korea comparing IGRA alone versus a combination of TST and IGRA. Intest Res. 2025;23(4). [DOI] [PMC free article] [PubMed]
- 24.Sivaporn G, Weerakit H, Ruamthip S, Gompol S, Jiratchaya S, Sasiwimol U, et al. Prevalence of latent tuberculosis infection and feasibility of TB preventive therapy among Thai prisoners: a cross-sectional study. BMC Public Health. 2021;21(1). [DOI] [PMC free article] [PubMed]
- 25.Carmine I, Annunziata M, Giuseppe DM, Sebastiana DB, Anna DC, Antonio Maria P, et al. Not only COVID-19: prevalence and management of latent mycobacterium tuberculosis infection in three penitentiary facilities in Southern Italy. Healthcare (Basel). 2022;10(2). [DOI] [PMC free article] [PubMed]
- 26.Lovisa K, Jyotirmoy D, Moa N, Amanda T, Isabelle P, Nina I, et al. A differential DNA methylome signature of pulmonary immune cells from individuals converting to latent tuberculosis infection. Sci Rep. 2021;11(1). [DOI] [PMC free article] [PubMed]
- 27.Monica D-S, Vinicius M F, Stephanus T M, Candice E M, Marianna O, Elouise E K, et al. Mycobacterium tuberculosis resisters despite HIV exhibit activated T cells and macrophages in their pulmonary alveoli. J Clin Invest. 2025;135(7). [DOI] [PMC free article] [PubMed]
- 28.Xinsong H, Jiongya L, Yan S, Guoli L, Honghuan S, Qiao L, et al. Smoking exposure and the risk of latent tuberculosis infection: results from NHANES 2011–2012. Toxics. 2024;12(1). [DOI] [PMC free article] [PubMed]
- 29.Mahesh S, Sangramsingh D, Mihir P, Atharva S, Juily S, Yogeshwari P, et al. Relationship between smoking and pulmonary tuberculosis based on positive sputum smears. GMS Hyg Infect Control. 2024;19(0). [DOI] [PMC free article] [PubMed]
- 30.Elizabeth L C, Abbas Z, Yin Bun C, Tsitsi B, Ethel D, Shungu S M, et al. Provider-initiated symptom screening for tuberculosis in Zimbabwe: diagnostic value and the effect of HIV status. Bull World Health Organ. 2010;88(1). [DOI] [PMC free article] [PubMed]
- 31.Camila Agra Gomes de L, Driele PB, Camila da Silva B, Patrícia BR, Maristela Pinheiro F, Edson A, et al. Clinical profile and mortality predictors for tuberculosis disease among patients with solid tumours. Infect Dis (Lond). 2022;54(11). [DOI] [PubMed]
- 32.Jacob B, Gregory H C, Sandro G. Population health in an era of rising income inequality: USA, 1980–2015. Lancet. 2017;389(10077). [DOI] [PubMed]
- 33.Choon Kin L, Sunny H V W, Grace L, Whitney T, Lai San T, Margaret I, et al. A prospective study to monitor for tuberculosis during anti-tumour necrosis factor therapy in patients with inflammatory bowel disease and immune-mediated inflammatory diseases. J Crohns Colitis. 2018;12(8). [DOI] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are accessible upon a reasonable request. The datasets produced and/or examined in this study can be obtained from the corresponding author upon a reasonable request.