Abstract
Background
Despite low sensitivity and implementation challenges, the tuberculin skin test (TST) remains the standard-of-care tuberculosis (TB) infection test in Mexico. Interferon gamma release assays (IGRA) may overcome TST-related challenges. Within the confines of the local programmatic setting, this cross-sectional study evaluated the prevalence of TB infection (TBI) and concordance of TST and IGRA in three high-risk populations in Mexicali, Baja California, Mexico.
Methods
Household contacts (HHC) of individuals with TB, people who use drugs (PWUD), people deprived of liberty (PDL) and prison employees underwent evaluation for TBI using TST and QIAreach, a novel IGRA. Prevalence of infection, concordance of test results and reactivity trends of time-to-results (TTR) by TST-induration size were assessed.
Results
In total, 214 of 411 (52.07%) people who had TST and 269 of 460 (58.48%) people who had IGRA tested positive for TBI. Frequency of infection varied across risk groups (HHC 29 (29.6%); PWUD 67 (70.53%); PDL 111 (56.06%) and prison employees 7 (35.0%), p<0.001). Overall concordance of TST and IGRA was 74.4%. Among people without a BCG vaccination, TTR decreased as TST induration size increased (from 1200 s in 0–4 mm to 808 s in >20 mm, p=0.05).
Conclusion
All risk groups had a high frequency of TBI, necessitating locally tailored guidelines for screening, treatment and management of TBI to optimise care for vulnerable populations.
Keywords: Tuberculosis, Respiratory infections, Diagnostic microbiology, Vulnerable Populations, Prisons
STRENGTHS AND LIMITATIONS OF THIS STUDY.
We were able to evaluate tuberculosis (TB) infection in three different groups who are typically at high risk of TB, due to exposures in their living environments and restricted access to routine health services, in Mexicali, Baja California.
Using programmatic data, we were able to assess the real-world performance of the tuberculin skin test and interferon gamma release assay in high-risk groups.
Due to the cross-sectional nature of the study and our inability to follow individuals over time, we were not able to ascertain when individuals were infected.
The active patient-finding approach may reduce the generalisability of test performance to the broader population.
The study lacks statistical power for formal TB infection prevalence estimates in people who use drugs, household contacts or people deprived of liberty, limiting generalisability.
Introduction
One-quarter of the world’s population is estimated to be infected with tuberculosis (TB).1 2 Approximately 10% of people infected with TB develop active disease; the majority do so within the first 2 years of infection.3 Identifying individuals infected with TB and administering TB preventive therapy (TPT) is essential to reduce global TB incidence.4 Rapid TPT initiation can reduce the risk of progression to disease by 60%.5 Several tests, including tuberculin skin test (TST) and interferon gamma release assay (IGRA), are used to detect TB infection (TBI); the latter group includes QuantiFERON TB and T-SPOT.
Certain groups, such as people living with HIV (PLHIV), household contacts (HHC) of individuals with TB, people who use drugs (PWUD) and people deprived of liberty (PDL), are at increased risk of TBI and progressing to active disease than the general population. 6The WHO recommends that high-risk populations be prioritised for TBI testing; however, in practice, coverage of testing and linkage to care is often variable by setting based on local epidemiology, guidelines and available resources.7
The Mexican state of Baja California has a TB incidence of >50 cases per 100 000 population annually, one of the highest TB rates in the country and in North America.2 8 Locally, TB risk has been associated with low socioeconomic status, drug use, HIV infection and diabetes.8 As of 2023, the Mexican and Baja California guidelines for TB management do not provide guidance on screening and treating TBI for most high-risk groups. TPT is currently only recommended for HHC <5 years, HHC between 5–14 years without documented BCG vaccination and PLHIV with a known exposure to someone with TB.9 These recommendations systematically exclude other high-risk groups.
We assessed TBI prevalence and concordance of two TBI tests—the standard of care, TST and a novel IGRA, QIAreach Quantiferon-TB or QIAreach—in three high-risk groups: HHCs, PWUD and people in prisons. QIAreach (QIAGEN GmbH, Hilden, Germany), a semiautomated lateral-flow assay, presents several advantages over both TST and other IGRA tests, such as simplified operations and rapid results.10 11 Our objectives were to characterise the TB burden and inform the future development of locally tailored guidelines for high-risk groups.
Methods
Study setting and population
The city of Mexicali, the capital of Baja California, is home to roughly 1 million people. The Mexicali health jurisdiction implements an intensified TB control and elimination strategy that expands on the Mexican national TB guidelines as part of routine, programmatic efforts to reduce TB locally. This strategy includes (1) active patient-finding efforts in high-risk groups, (2) expanded testing for TBI among high-risk groups and (3) improved TPT prescription practices for people with TBI and without TB disease. Locally, TST is used to diagnose TBI; a cut-off of >5 mm induration is used for PLHIV and HHC <5 years old, while >10 mm induration is used for all other groups. When TBI is suspected, 6 months of isoniazid is routinely prescribed.
Between December 2020 and January 2022, QIAreach was prospectively piloted as part of the programmatic active patient-finding efforts. This cross-sectional study was conducted within three high-risk groups identified by the Mexicali programme:
Group 1: PWUD
PWUD and were residents of a rehabilitation facility (for treatment and recovery from addiction to alcohol, methamphetamine, cocaine and/or heroin) were screened for TB in December 2020. All individuals in the facility were male and >18 years old.
Group 2: HHC of patients with TB
HHC of individuals diagnosed with TB disease in July 2021 were enumerated from the programme’s TB register, and clinicians and community health workers were deployed to the households for TB evaluation.
Group 3: PDL and prison employees
Individuals in the Mexicali prison, including PDL and prison employees, were screened for TB in January 2022. Individuals were enumerated by the prison’s sentencing and employment rosters and were screened to determine if they were serving a sentence or were employed and working at the prison at the time. All individuals were >18 years old; PDL were male only. Unless otherwise indicated, prison employees were included with PDL in analyses due to the shared built environment.
Individuals from all three groups were eligible for TB screening if they had no self-reported previous history of active TB, no previous diagnosis of latent TBI or associated treatment, and had not received a BCG vaccination or TST administration within the last 6 months. Individuals’ responses were verified with the TB programme’s records by the study team.
Our intended sample size was based on the availability of QIAreach testing units at the time of programmatic implementation. 500 total testing units were available during the study timeline, and we intended to allocate one test per study participant. Therefore, our planned sample size was 500.
Key procedures and measures
Eligible individuals provided written informed consent. For child HHC under 18 years old, written informed consent was provided by a guardian, and verbal assent was obtained from the individual child. Registration and evaluation of consenting individuals were completed according to local TB control guidelines. Individuals who did not consent to participate received a standard-of-care evaluation from the TB programme. Standard-of-care evaluation included symptom screening followed by chest radiography and sputum collection for bacteriologic confirmation using GeneXpert MTB/RIF (Cepheid, Sunnyvale, USA).
Participants received a medical consultation with a clinician and were screened for TB using the WHO-recommended symptom screen, asking about the presence of cough, fever, weight loss or night sweats.6 Age, sex, self-reported status as a close contact of someone with TB, HIV status and BCG vaccination status were collected. All individuals then received chest radiography using the MiniXray battery-powered portable X-ray unit (MinXray, Northbrook, USA). Radiographs were read by a clinician, and individuals were categorised as suggestive or not suggestive of TB disease based on radiographic findings and the presence of at least one sign or symptom.6 12 If initial screening or chest radiograph was suggestive of TB disease, a sputum sample was collected for testing using GeneXpert. Individuals diagnosed with TB disease per Gene Xpert result or clinical diagnosis were linked to care within the local TB programme.
Individuals who were not diagnosed with TB on the date of screening were assessed for TBI using TST and IGRA. An intradermal injection of TST was placed; TST induration was measured and interpreted by TB programme staff after 72 hours. TBI was defined as a positive TST per the national TB guidelines.12 Additionally, a blood sample was drawn for IGRA analysis and transported to the Universidad Autonoma de Baja California laboratory to be processed and analysed per the QIAreach user manual.13 Results—positive, negative and time elapsed for analysis to complete (time-to-results, TTR)—were automated via the QIAreach software platform. TTR was a continuous numeric value ranging from 0 to 1200 s; negative samples were automatically analysed for 1200 s.
Analysis
Participant characteristics and TB disease and infection evaluation results were reported using descriptive statistics. The frequency of observed TST induration sizes (0–4, 5–9, 10–14, 15–19, 20+ mm and not available), median TST induration size, frequency of IGRA results, median TTR for all IGRA results and median TTR for only positive IGRA results were summarised, in total, by risk group, and by BCG vaccination status. For individuals with valid test results, the frequency of IGRA and TST positivity and negativity was compared across risk groups and by BCG vaccination status using Pearson’s χ2 Test. Individuals with indeterminate IGRA or TST results, or with missing results, were included in summary statistics but were excluded from concordance analyses.
In the absence of a gold standard for identifying TBI, concordance in test results was selected for this study’s measure of diagnostic accuracy. TST and IGRA results were assessed for concordance under two scenarios: (1) a positive TST defined as ≥10 mm for all individuals without HIV and ≥5 mm for PLHIV and (2) a positive TST defined as ≥5 mm for all. We tested for differences in median TTR across risk groups using the Kruskal-Wallis test. Median TTR was assessed for differences by TST induration size group using Spearman’s rank correlation. Analyses were performed in RStudio V.2022.07.0.
Patient and public involvement
Patients and the public were not formally involved in the design or choice of outcomes of this study. Patients were involved in the research as recipients of TB services offered by the Mexicali jurisdiction TB program. They were first engaged in active case-finding outreach offered by the jurisdiction and were offered to participate in this study during those routine services.
Results
A total of 515 individuals were evaluated for TB, including 103 PWUD, 98 HHC, 289 PDL and 25 prison employees (table 1). The median age of participants was 36 (IQR 27–49). Due to only males being included in the PWUD and PDL groups, 458 (88.9%) of participants were male. BCG vaccination was documented for 469 (91.1%) and 30 (5.8%) individuals were living with HIV.
Table 1. Demographic characteristics of PWUD and HHC who were tested for TBI using TST and/or IGRA.
| PWUD (n=103) | HHC (n=98) | PDL (n=289) | Prison employees (n=25) | TOTAL (n=515) | |
|---|---|---|---|---|---|
| Median age (IQR) | 34 (27–42) | 27 (8.75–45) | 40 (30–53) | 42 (36–49) | 36 (27–49) |
| Age group | |||||
| 0–19 | 1 (0.97%) | 35 (35.71%) | 3 (1.04%) | 0 (0) | 39 (7.57%) |
| 20–39 | 63 (61.17%) | 35 (35.71%) | 140 (48.44%) | 9 (36.00%) | 247 (47.96%) |
| 40–59 | 39 (37.86%) | 21 (21.43%) | 102 (35.29%) | 16 (64.00%) | 178 (34.56%) |
| 60+ | 0 (0) | 7 (7.14%) | 44 (15.22%) | 0 (0) | 51 (9.90%) |
| Sex: male | 103 (100.00%)* | 51 (52.04%) | 289 (100.0%)* | 15 (60.00%) | 458 (88.93%) |
| BCG vaccinated | 86 (83.50%) | 94 (95.92%) | 265 (91.70%) | 24 (96.00%) | 469 (91.07%) |
| HIV-positive | 19 (18.45%) | 1 (1.02%) | 10 (3.46%) | 0 (0) | 30 (5.83%) |
| Close contact of TB patient | 26 (25.24%) | 98 (100.00%) | 15 (5.19%) | 4 (16.00%) | 143 (27.77%) |
All males due to the nature of the institution.
HHC, household contacts; IGRA, interferon gamma release assay; PDL, people deprived of liberty; PWUD, people who use drugs; TB, tuberculosis; TBI, TB infection; TST, tuberculin skin test.
Results of TB evaluation
Chest radiography indicated TB-suggestive results for 29 (5.6%) participants; 14 (13.6%) PWUD, 5 (5.1%) HHCs and 10 (3.5%) PDL (table 2). TB was suspected by medical evaluation for 35 (6.8%) participants; 10 (9.7%) PWUD, 24 (8.3%) PDL and 1 (4.0%) prison employee. 5 (0.97%) people were diagnosed with TB after screening and therefore included in the TB infection analysis.
Table 2. Clinical characteristics of study participants.
| PWUD (n=103) | HHC (n=98) | PDL (n=289) | Prison employees (n=25) | Total (n=515) | |
|---|---|---|---|---|---|
| Medical evaluation | |||||
| TB indicated | 10 (9.71%) | 0 (0) | 24 (8.30%) | 1 (4.00%) | 35 (6.80%) |
| TB not indicated | 93 (90.29%) | 98 (100.00%) | 265 (91.70%) | 24 (96.00%) | 480 (93.20%) |
| Chest radiography | |||||
| TB indicated | 14 (13.59%) | 5 (5.10%) | 10 (3.46%) | 0 (0) | 29 (5.63%) |
| TB not indicated | 89 (86.41%) | 93 (94.90%) | 279 (96.54%) | 25 (100.00%) | 486 (94.37%) |
| TB disease diagnosed | 4 (3.88%) | 0 (0) | 1 (0.35%) | 0 (0) | 5 (0.97%) |
HHC, household contacts; PDL, people deprived of liberty; PWUD, people who use drugs; TB, tuberculosis.
Results of TST
TST was done in 411 (79.8%) individuals; results were not available for eight PWUD who were lost to follow-up after TST placement and for 91 PDL and five prison employees due to a national tuberculin shortage. A positive TST result was observed in 214 (52.07%) individuals per standard guidelines and 294 (71.5%) using a lowered threshold of >5 mm (table 3). PWUD had the highest proportion of TST positivity per national guidelines, with 67 (70.5%) testing positive, compared with HHC (29.6%), PDL (56.1%) and prison employees (35.0%). We found a statistically significant difference in positivity (p<0.001). Median TST induration size was 10 mm (IQR 2–15), highest among PWUD and lowest among prison employees (table 3).
Table 3. Results of TST and IGRA testing.
| PWUD (n=103) | HHC (n=98) | PDL (n=289) | Employees (n=25) | P value | Total (n=515) | |
|---|---|---|---|---|---|---|
| TST induration | ||||||
| Not available | 8 (7.77%) | 0 (0) | 91 (31.49%) | 5 (20.00%) | 104 (20.19%) | |
| 0–4 mm | 19 (18.44%) | 47 (47.96%) | 40 (13.84%) | 11 (44.00%) | 117 (22.72%) | |
| 5–9 mm | 13 (12.62%) | 22 (22.45%) | 48 (16.61%) | 2 (8.00%) | 85 (16.50%) | |
| 10–14 mm | 21 (20.38%) | 15 (15.31%) | 51 (17.65%) | 2 (8.00%) | 89 (17.28%) | |
| 15–19 mm | 27 (26.21%) | 7 (7.14%) | 40 (13.84%) | 4 (16.00%) | 78 (15.15%) | |
| 20+ mm | 15 (14.56%) | 7 (7.14%) | 19 (6.57%) | 1 (4.00%) | 42 (8.16%) | |
| Median (IQR) | 12 (8–17.5) | 5 (0–10) | 10 (6–15) | 4 (0–12.75) | 10 (2–15) | |
| TST positive (assuming >10 mm for HIV-negative, ≥5 mm for PLHIV) | 67 (70.53%) | 29 (29.59%) | 111 (56.06%) | 7 (35.00%) | <0.001 | 214 (52.07%) |
| TST positive (≥5 mm) | 76 (80.00%) | 51 (52.04%) | 158 (79.80%) | 9 (45.00%) | <0.001 | 294 (71.53%) |
| TST positive (≥10 mm) and documented BCG | 54 (56.84%) | 29 (29.59%) | 101 (51.01%) | 7 (35.00%) | <0.001 | 191 (46.47%) |
| IGRA results | ||||||
| Positive | 69 (66.99%) | 23 (23.47%) | 165 (57.09%) | 12 (48.00%) | <0.001 | 269 (52.23%) |
| Negative | 26 (25.24%) | 75 (76.53%) | 77 (26.64%) | 13 (52.00%) | 191 (37.09%) | |
| Not performed | 8 (7.77%) | 0 (0) | 14 (4.84%) | 0 (0) | 22 (4.27%) | |
| Error | 0 (0) | 0 (0) | 33 (11.42%) | 0 (0) | 33 (6.41%) | |
| IGRA TTR: median (IQR) | 1200 (675–1200) | 1200 (1200–1200) | 1197.5 (685–1200) | 1200 (910–1200) | 1200 (815–1200) | |
| IGRA TTR: positive median (IQR) | 1010 (561.25–1200) | 1200 (1020–1200) | 1085 (670–1200) | 960 (683.75–1200) | 1085 (655–1200) |
HHC, household contacts; IGRA, interferon gamma release assay; PDL, people deprived of liberty; PLHIV, people living with HIV; PWUD, people who use drugs; TST, tuberculin skin test; TTR, time-to-results.
Results of IGRA
IGRA was completed for 460 (89.3%) individuals; results were unavailable for 55 (10.7%) samples due to operational error (6.4%) or phlebotomy-related challenges (4.3%) (table 3). A total of 269 (52.2%) had a positive IGRA result. PWUD had the highest proportion of positivity results (67.0%) compared with HHC (23.5%), PDL (57.1%) and prison employees (48.0%), and these differential positivity rates were statistically significant (p<0.001). Median TTR for all IGRA-positive results was 1085 s (IQR 655–1200); PWUD TTR was 1010 s (IQR 561.25–1200), HHC TTR was 1200 s (IQR 1020–1200), PDL TTR was 1085 s (IQR 670–1200) and prison employees TTR was 960 s (IQR 684–1200).
Comparison and concordance of TB infection test results
Across all individuals for whom both TST and IGRA results were available (n=386), TST and IGRA were both positive in 159 (41.2%) individuals and were both negative in 128 (33.2%) individuals; there was an overall agreement of 74.4% (table 4). Among PWUD, HHC and a combined population of PDL and prison employees, positive and negative agreement was 79.5%, 73.5% and 72.5%, respectively. Using a lower induration cut-off of ≥5 mm to define a positive TST, per cent agreement between positive and negative results increased to 79.8% (table 4).
Table 4. Concordance in TST and IGRA results, per patient group and BCG vaccination status.
| TST results | IGRA positive | IGRA negative | % Total agreement | Difference in agreement | |
|---|---|---|---|---|---|
| PWUD (n=88) | TST+ ≥10 mm** | 54 (61.36%) | 8 (9.09%) | 79.54% | 7.96% |
| TST− <10 mm | 10 (11.36%) | 16 (18.18%) | |||
| TST+ ≥5 mm | 62 (70.45%) | 9 (10.23%) | 87.50% | ||
| TST− <5 mm | 2 (2.27%) | 15 (17.05%) | |||
| HHC (n=98) | TST+ ≥10 mm | 13 (13.27%) | 16 (16.33%) | 73.47% | −10.20% |
| TST− <10 mm | 10 (10.20%) | 59 (60.20%) | |||
| TST+ ≥5 mm | 19 (19.39%) | 32 (32.65%) | 63.27% | ||
| TST− <5 mm | 4 (4.08%) | 43 (43.88%) | |||
| PDL and prison employees (n=200)* | TST+ ≥10 mm | 92 (46.00%) | 14 (7.00%) | 72.50% | 12.00% |
| TST− <10 mm | 41 (20.50%) | 53 (26.50%) | |||
| TST+ ≥5 mm | 126 (63.00%) | 24 (12.00%) | 84.50% | ||
| TST− <5 mm | 7 (3.50%) | 43 (21.50%) | |||
| BCG (n=349) | TST+ ≥10 mm** | 141 (40.40%) | 37 (10.60%) | 73.64% | 5.73% |
| TST− <10 mm | 55 (15.76%) | 116 (33.24%) | |||
| TST+ ≥5 mm | 185 (53.01%) | 61 (17.48%) | 79.37% | ||
| TST− <5 mm | 11 (3.15%) | 92 (26.36%) | |||
| No BCG (n=37) | TST+ ≥10 mm | 18 (48.65%) | 1 (2.70%) | 81.01% | 2.77% |
| TST− <10 mm | 6 (16.22%) | 12 (32.43%) | |||
| TST+ ≥5 mm | 22 (59.46%) | 4 (24.32%) | 83.78% | ||
| TST− <5 mm | 2 (5.41%) | 9 (24.32%) | |||
| Total (n=386) | TST+ ≥10 mm** | 159 (41.19%) | 38 (9.84%) | 74.35% | 5.44% |
| TST− <10 mm | 61 (15.80%) | 128 (33.16%) | |||
| TST+ ≥5 mm | 207 (53.63%) | 65 (16.58%) | 79.79% | ||
| TST− <5 mm | 13 (3.37%) | 101 (26.17%) |
*PDL and Prison Employees reported as a single group due to small population size of prison employees and nature of their shared environment.
** Includes 4 PLHIV with TST induration between 5-10mm
HHC, household contacts; IGRA, interferon gamma release assay; PDL, people deprived of liberty; PWUD, people who use drugs; TST, tuberculin skin test.
Assessing correlation between TST induration and IGRA TTR
TST induration size and IGRA TTR were not correlated, with a Spearman’s Correlation Coefficient of −0.031 (p=0.65) (table 5). In the subgroup of people without BCG vaccination, a larger TST induration was associated with a shorter TTR (p=0.05).
Table 5. Median TTR (in seconds) for positive IGRA tests per patient group, BCG vaccination status and TST induration group.
| Median TTR for all positive IGRA (IQR) | Total | PWUD | HHC | PDL | Employees | No BCG | BCG | |
|---|---|---|---|---|---|---|---|---|
| 1110 (655–1200) | 1010 (561.25–1200) | 1200 (1020–1200) | 1097.5 (675–1200) | 960 (683.75–1200) | 825 (460–1200) | 140 (675–1200) | ||
| P value=0.06875 | P value=0.1037 | |||||||
| Median TTR by TST (IQR) | 0–4 mm | 1200 (1140–1200) | 1170 (1155–1185) | 1200 (1200–1200) | 1155 (462.5–1200) | n/a | 1200 (1200–1200) | 1200 (842.5–1200) |
| 5–9 mm | 930 (670–1200) | 1200 (712.5–1200) | 1040 (876.25–1200) | 862.5 (650–1200) | 965 (850–1080) | 1037.5 (775–1200) | 930 (670–1200) | |
| 10–14 mm | 1160 (835–1200) | 967.5 (651.25–1166.25) | 1200 (1200–1200) | 1200 (852.5–1200) | 1055 (982.5–1127.5) | 982.5 (767.5–1166.25) | 1160 (857.5–1200) | |
| 15–19 mm | 1085 (653.75–1200) | 932.5 (483.75–120) | 790 (585–995) | 1140.00 (798.75–1200) | 1105 (920–1200) | 460 (460–1050) | 1170 (762.5–1200) | |
| 20+ mm | 1200 (760–1200) | 815 (545–1200) | 1200 (760–1200) | 1200 (790–1200) | 1200 (1200–1200) | 807.5 (702.5–918.75) | 1200 (760–1200) | |
| Spearman’s rank correlation | P value=0.65 | P value=0.13 | P value=0.53 | P value=0.24 | P value=0.49 | P value=0.05 | P value=0.65 | |
P values from statistical tests are reported in italics. Those that were less than or equal to 0.05 are bolded.
HHC, household contacts; IGRA, interferon gamma release assay; PDL, people deprived of liberty; PWUD, people who use drugs; TST, tuberculin skin test; TTR, time-to-results.
Discussion
We observed a high frequency of TB infection among PWUD, HHC, PDL and prison employees in Mexicali. Although TB disease burden is known to be more prevalent in this border region,8 and TB infection positivity has been established for high-risk populations in other areas in Baja California,14 the specific prevalence of TB infection in these groups in Mexicali has not previously been quantified. Illicit drug use is a significant risk factor for TB, given its association with high-risk behaviours, low socioeconomic status, social isolation and immunological vulnerability.15,17 Global prevalence among PWUD ranges from 10% to 59%, per TST18; in Tijuana, the prevalence among people who inject drugs was 67% using IGRA.19 The observed prevalence in Mexicali (70.5% via TST and 72.5% via IGRA) is higher, potentially due to the carceral nature of the compulsory drug detention centre.20
Prisons are known reservoirs for TB due to poor nutrition, drug use, overcrowding and insufficient ventilation.20 21 A meta-analysis found a pooled prevalence of TB infection in prison populations to be 44.4%,22 although the prevalence in Latin American prison settings ranges from 22.5% among incarcerated males in Brazil,23 to 33.2% among Chilean inmates,24 to 77.6% in Colombian prisons.21 The observed prevalence in Mexicali’s prisons (56.1% via TST and 68.2% via IGRA) aligns with the higher end of global estimates, indicating ongoing transmission.23 Similarly, a meta-analysis found the pooled prevalence of TB infection among prison employees to be 26%, but as high as 44% in high-burden settings.25 A study in Brazil found that 22.6% of prison employees had TB infection, per IGRA.26 In Mexicali, we observed prevalence higher than the global estimates, with 35.0% via TST and 48.0% via IGRA.
Estimates of TB infection prevalence among HHC vary considerably from 5% to 81%, with a pooled estimate for TB infection of 41%.27 A study in Zacatecas, Mexico, found a TST-determined prevalence of 34.1% using a cut-off of >5 mm, 53.6% using a cut-off of >10 mm and 41.4% using IGRA.28 Another study found 39.7% of TB contacts had a positive IGRA, despite only 3.2% having a positive TST.29 In another study conducted in Mexican border states, 49.7% of contacts were positive through IGRA, but only 21.1% were positive through TST.30 The burden observed in this Mexicali population, 29.6% via TST and 23.5% via IGRA, was consistent with global estimates but lower than estimates from other states in Mexico. Over one-third of HHC in this study were <20 years old, indicating a population that would benefit from programmatic TPT administration.
Concordance between IGRA and TST was 74.4% (TST cut-off >10 mm) and 79.8% (cut-off >5 mm), consistent with previous research comparing the QuantiFERON-TB-Gold In-Tube to TST and found 73.1% and 74.8% agreement with the same cut-offs.28 Higher concordance in PWUD and those unvaccinated with BCG points to possible reduced accuracy of TST in BCG-vaccinated populations, particularly younger groups.31 Results are consistent with research indicating IGRA–TST concordance is higher among close contacts using the >10 mm cut-off compared with the ≥5 mm cut-off.28 While literature on TST-IGRA concordance is scarce, results suggest that setting-specific TST thresholds for TB infection could improve diagnostic accuracy.32
This study has a handful of limitations. Because TB infection tests were administered as part of programmatic care, we cannot truly assess the advantages of using one diagnostic tool over another due to unmeasured confounding and lack of standardised procedures for delivering care across the included populations; this may also reduce the generalisability of the measured test performance for the general population. Nor were we able to follow individuals over time or ascertain when individuals were infected due to the cross-sectional nature of the study. Moreover, the tests were deployed alongside an active patient-finding exercise, which may limit the generalisability of these tests’ performance in the general population. Additionally, newer diagnostic tools such as computer-aided diagnosis (CAD) are increasingly used in programmatic settings, and evaluation of newer TB infection detection tools should be paired alongside CAD; unfortunately, CAD was not available in this setting at the time of this programme’s implementation. Finally, these results are not powered for a formal estimate of TB infection prevalence in PWUD, HHC or PDL; thus, caution should be used when generalising these disease estimates across similar populations.
The study’s strengths include providing new data on TB infection in vulnerable populations in Baja California and comparing the performance of TST and IGRA in a real-world setting. Additionally, this study leverages programmatic data to assess the real-world performance of these TB infection tests. This evidence is critical for tailoring clinical practices for people with restricted access to care, and it further underscores the need for TB services among these populations. Given the uncertainty around the comparative effectiveness of TST and IGRA to identify TB infection globally31,34 and in Mexico,35 these findings may be useful to other TB elimination programmes in high TB-burden settings.
Our findings suggest that locally defined guidelines for diagnosing TB infection may benefit Mexicali, where there is a high burden of TB, a high BCG vaccination coverage and political interest in a comprehensive TB elimination strategy, including prioritisation of TB infection evaluation and treatment. Further qualitative research is needed to explore preferences among patient groups, providers and programme implementers, and improve TB testing and treatment strategies.
Footnotes
Funding: This research was conducted as part of Advance Access & Delivery’s routine programmatic and operational research initiatives in partnership with its Zero TB Initiative partners. Programmatic activities were funded through a charitable donation made by the Tuberculosis Foundation of Virginia, and the IGRA tests used in this study were donated by QIAGEN. No additional funding was secured for the design or conduct of this research specifically.
Prepublication history for this paper is available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2024-097505).
Data availability free text: Study data will be made available upon reasonable request.
Patient consent for publication: Consent obtained directly from patient(s).
Ethics approval: IRB approval was received under the General Hospital Ethical Committee of Mexicali, with the number 02-01-HGMXL//UABC/ISESALUD/QIAGEN/2020-11-20-279. Written informed consent was provided by all participants prior to enrolment; all data was managed under the privacy policies of the healthcare services of the Mexicali Jurisdiction.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient and public involvement: Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.
Data availability statement
Data are available upon reasonable request.
References
- 1.Houben R, Dodd PJ. The Global Burden of Latent Tuberculosis Infection: A Re-estimation Using Mathematical Modelling. PLoS Med. 2016;13:e1002152. doi: 10.1371/journal.pmed.1002152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.World Health Organization Global tuberculosis report. 2023 https://iris.who.int/bitstream/handle/10665/373828/9789240083851-eng.pdf?sequence=1 Available.
- 3.Biraro IA, Kimuda S, Egesa M, et al. The Use of Interferon Gamma Inducible Protein 10 as a Potential Biomarker in the Diagnosis of Latent Tuberculosis Infection in Uganda. PLoS ONE. 2016;11:e0146098. doi: 10.1371/journal.pone.0146098. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Rangaka MX, Cavalcante SC, Marais BJ, et al. Controlling the seedbeds of tuberculosis: diagnosis and treatment of tuberculosis infection. Lancet. 2015;386:2344–53. doi: 10.1016/S0140-6736(15)00323-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Smieja M, Marchetti C, Cook D, et al. Isoniazid for preventing tuberculosis in non-HIV infected persons. Cochrane Database Syst Rev. 1999 doi: 10.1002/14651858.cd001363. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.WHO . WHO Consolidated Guidelines on Tuberculosis Module 2: Screening Systemic Screening for Tuberculosis Disease. 2020. [PubMed] [Google Scholar]
- 7.WHO . Latent tuberculosis infection: updated and consolidated guidelines for programmatic management. World Health Organization; 2018. [PubMed] [Google Scholar]
- 8.Bello-López JM, León-García G, Rojas-Bernabé A, et al. Morbidity Trends and Risk of Tuberculosis: Mexico 2007-2017. Can Respir J. 2019;2019:8295261. doi: 10.1155/2019/8295261. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Burgos JL, Kahn JG, Strathdee SA, et al. Targeted screening and treatment for latent tuberculosis infection using QuantiFERON-TB Gold is cost-effective in Mexico. Int J Tuberc Lung Dis. 2009;13:962–8. [PMC free article] [PubMed] [Google Scholar]
- 10.Stieber F, Howard J, Manissero D, et al. Evaluation of a lateral-flow nanoparticle fluorescence assay for TB infection diagnosis. Int J Tuberc Lung Dis. 2021;25:917–22. doi: 10.5588/ijtld.21.0391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Mahmoudi S, Nourazar S. Evaluating the diagnostic accuracy of QIAreach QuantiFERON-TB compared to QuantiFERON-TB Gold Plus for tuberculosis: a systematic review and meta-analysis. Sci Rep. 2024;14:14455. doi: 10.1038/s41598-024-65663-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.CENAPRECE . Programa Sect Salud; 2013. Programa de acción específico prevención y control de la tuberculosis.http://www.cenaprece.salud.gob.mx/descargas/pdf/PAE_PrevencionControlTuberculosis2013_2018.pdf Available. [Google Scholar]
- 13.QIAGEN QIAreach ehub user manual. 2021:1–40.
- 14.Velasquez MG, Laniado-Laborin R, Rodwell TC, et al. Tuberculosis testing among populations with high HIV risk in Tijuana, Baja California, Mexico. Rev Panam Salud Publica . 2012;32:30–5. doi: 10.1590/S1020-49892012000700005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.World Health Organization . WHO consolidated guidelines on tuberculosis. Module 1: Tuberculosis preventive treatment. 2020. [Google Scholar]
- 16.Silva DR, Muñoz-Torrico M, Duarte R, et al. Risk factors for tuberculosis: diabetes, smoking, alcohol use, and the use of other drugs. J Bras Pneumol. 2018;44:145–52. doi: 10.1590/s1806-37562017000000443. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Myers B, Carney T, Rooney J, et al. Smoked drug use in patients with TB is associated with higher bacterial burden. Int J Tuberc Lung Dis. 2023;27:444–50. doi: 10.5588/ijtld.22.0650. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.D RG, R TC, G RS. Tuberculosis and illicit drug use: Review and update. Clin Infect Dis. 2009;48:72–82. doi: 10.1086/594126.Tuberculosis. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Garfein RS, Lozada R, Liu L, et al. High prevalence of latent tuberculosis infection among injection drug users in Tijuana, Mexico. Int J Tuberc Lung Dis. 2009;13:626–32. [PMC free article] [PubMed] [Google Scholar]
- 20.Dolan K, Wirtz AL, Moazen B, et al. Global burden of HIV, viral hepatitis, and tuberculosis in prisoners and detainees. The Lancet. 2016;388:1089–102. doi: 10.1016/S0140-6736(16)30466-4. [DOI] [PubMed] [Google Scholar]
- 21.Rueda ZV, Arroyave L, Marin D, et al. High prevalence and risk factors associated with latent tuberculous infection in two Colombian prisons. Int J Tuberc Lung Dis. 2014;18:1166–71. doi: 10.5588/ijtld.14.0179. [DOI] [PubMed] [Google Scholar]
- 22.Placeres AF, de Almeida Soares D, Delpino FM, et al. Epidemiology of TB in prisoners: a metanalysis of the prevalence of active and latent TB. BMC Infect Dis. 2023;23:20. doi: 10.1186/s12879-022-07961-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Carbone A da S, Paião DSG, Sgarbi RVE, et al. Active and latent tuberculosis in Brazilian correctional facilities: a cross-sectional study. BMC Infect Dis. 2015;15:24. doi: 10.1186/s12879-015-0764-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Aguilera XP, González C, Nájera-De Ferrari M, et al. Tuberculosis in prisoners and their contacts in Chile: estimating incidence and latent infection. Int J Tuberc Lung Dis. 2016;20:63–70. doi: 10.5588/ijtld.15.0056. [DOI] [PubMed] [Google Scholar]
- 25.Grenzel ML, Grande AJ, Paniago AMM, et al. Tuberculosis among correctional facility workers: A systematic review and meta-analysis. PLoS ONE. 2018;13:e0207400. doi: 10.1371/journal.pone.0207400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Trevisol M, Moreira TP, Sanvezzo GHB, et al. Latent Tuberculosis Infection Diagnosis Using QuantiFERON-TB Gold Plus Kit Among Correctional Workers: A Cross-Sectional Study in Francisco Beltrão-PR, Brazil. J Community Health . 2023;48:600–5. doi: 10.1007/s10900-023-01201-z. [DOI] [PubMed] [Google Scholar]
- 27.Sagili KD, Muniyandi M, Shringarpure K, et al. Strategies to detect and manage latent tuberculosis infection among household contacts of pulmonary tb patients in high tb burden countries ‐ a systematic review and meta‐analysis. Tropical Med Int Health . 2022;27:842–63. doi: 10.1111/tmi.13808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Serrano-Escobedo CJ, Enciso-Moreno JA, Monárrez-Espino J. Performance of tuberculin skin test compared to QFT-IT to detect latent TB among high-risk contacts in Mexico. Arch Med Res. 2013;44:242–8. doi: 10.1016/j.arcmed.2013.02.004. [DOI] [PubMed] [Google Scholar]
- 29.González-Salazar F, Vargas-Villarreal J, Garcialuna-Martínez FJ, et al. Snapshot of Quantiferon TB gold testing in Northern Mexico. Tuberculosis (Edinb) 2011;91 Suppl 1:S34–7. doi: 10.1016/j.tube.2011.10.007. [DOI] [PubMed] [Google Scholar]
- 30.Oren E, Alatorre-Izaguirre G, Vargas-Villarreal J, et al. Interferon Gamma-Based Detection of Latent Tuberculosis Infection in the Border States of Nuevo Leon and Tamaulipas, Mexico. Front Public Health. 2015;3:1–6. doi: 10.3389/fpubh.2015.00220. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Wang L, Turner MO, Elwood RK, et al. Vaccination on Tuberculin Skin Test Measurements. Thorax. 2002;57:804–9. doi: 10.1136/thorax.57.9.804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Gualano G, Mencarini P, Lauria FN, et al. Tuberculin skin test - Outdated or still useful for Latent TB infection screening? Int J Infect Dis. 2019;80S:S20–2. doi: 10.1016/j.ijid.2019.01.048. [DOI] [PubMed] [Google Scholar]
- 33.Auguste P, Tsertsvadze A, Pink J, et al. Comparing interferon-gamma release assays with tuberculin skin test for identifying latent tuberculosis infection that progresses to active tuberculosis: systematic review and meta-analysis. BMC Infect Dis. 2017;17:200. doi: 10.1186/s12879-017-2301-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Cohen A, Mathiasen VD, Schön T, et al. The global prevalence of latent tuberculosis: a systematic review and meta-analysis. Eur Respir J. 2019;54:1–14.:1900655. doi: 10.1183/13993003.00655-2019. [DOI] [PubMed] [Google Scholar]
- 35.Monárrez-Espino J, Enciso-Moreno JA, Laflamme L, et al. Serial QuantiFERON-TB Gold In-Tube assay and tuberculin skin test to diagnose latent tuberculosis in household Mexican contacts: conversion and reversion rates and associated factors using conventional and borderline zone definitions. Mem Inst Oswaldo Cruz. 2014;109:863–70. doi: 10.1590/0074-0276140085. [DOI] [PMC free article] [PubMed] [Google Scholar]