Home Tuberculosis Risk of developing active tuberculosis following tuberculosis screening and preventive therapy for...

Risk of developing active tuberculosis following tuberculosis screening and preventive therapy for Tibetan refugee children and adolescents in India: An impact assessment



Methods and findings

A mobile team annually screened children and staff for TB at the 7 boarding schools in Himachal Pradesh, India, using symptom criteria, radiography, molecular diagnostics, and tuberculin skin tests. TB infection (TBI) was treated with short-course regimens of isoniazid and rifampin or rifampin. TB disease was treated according to Tibetan and Indian guidelines. Between April 2017 and December 2019, 6,582 schoolchildren (median age 14 [IQR 11–16] years) and 807 staff (median age 40 [IQR 33–48] years) were enrolled. Fifty-one percent of the students and 58% of the staff were females. Over 13,161 person-years of follow-up in schoolchildren (median follow-up 2.3 years) and 1,800 person-years of follow-up in staff (median follow-up 2.5 years), 69 TB episodes occurred in schoolchildren and 4 TB episodes occurred in staff, yielding annual incidence rates of 524/100,000 (95% CI 414–663/100,000) person-years and 256/100,000 (95% CI 96–683/100,000) person-years, respectively. Of 1,412 schoolchildren diagnosed with TBI, 1,192 received TPT. Schoolchildren who received TPT had 79% lower risk of TB disease (adjusted hazard ratio [aHR] 0.21; 95% CI 0.07–0.69; p = 0.010) compared to non-recipients, the primary study outcome. Protection was greater in recent contacts (aHR 0.07; 95% CI 0.01–0.42; p = 0.004), the secondary study outcome. The prevalence of recent contacts was 28% (1,843/6,582). Two different TPT regimens were used (3HR and 4R), and both were apparently effective. No staff receiving TPT developed TB. Overall, between 2017 and 2019, TB disease incidence decreased by 87%, from 837/100,000 (95% CI 604–1,129/100,000) person-years to 110/100,000 (95% CI 36–255/100,000) person-years (p < 0.001), and TBI prevalence decreased by 42% from 19% (95% CI 18%–20%) to 11% (95% CI 10%–12%) (p < 0.001). A limitation of our study is that TB incidence could be influenced by secular trends during the study period.

Author summary


It is estimated that every year 1.5 million children and adolescents 0–19 years old develop tuberculosis (TB) disease, and 7.5 million children and young adolescents (0–14 years) develop TB infection (TBI) [14]. Approximately 233,000 children and young adolescents die of TB annually [1,5]. Despite this disease burden, implementation of TB diagnostic and preventive services in children and adolescents remains suboptimal. Tibetan refugee children and adolescents in India constitute a vulnerable population having several demographic, socioeconomic, biomedical, and political risk factors for TB [69]. A drug-resistant TB survey carried out in the Tibetan population in India in 2010–2011 showed a high prevalence of multidrug-resistant TB (MDR-TB)—14.5% among new TB cases and 31.4% among previously treated cases [8]. In 2011–2013, during a community-wide active case finding campaign in India, we detected a high prevalence of TB disease in the Tibetan schools (394/100,000 schoolchildren) [6]. Driven by an urgent need to reduce the TB burden for this vulnerable population, and in response to the global campaign to end TB, we initiated a comprehensive TB screening and treatment program—Zero TB Kids (ZTBK)—for Tibetan children in India in 2017 [9]. During initial screening in 2017–2018, we detected a high prevalence of TB disease (916/100,000 schoolchildren) and TBI (19%) in children at Tibetan boarding schools in Himachal Pradesh, where children reside in congregate settings [9]. As a core component of this population-level program across schools, TB preventive treatment (TPT) was implemented using short-course regimens, with a high treatment completion rate (95%). ZTBK then continued with annual screening, treatment, and preventive treatment for TB. We present here the results and impact of 3 years of TB screening and TPT implementation under the program.


Screening for TBI and disease

Between April 2017 and December 2019, 7 Tibetan boarding schools in Himachal Pradesh had periodic screening for TBI and TB disease. The schoolchildren and staff constitute an open cohort, with participants entering and exiting through new admissions in the schools and matriculation to college, respectively. Three waves of screening and treatment for latent TBI have been completed. In the starting year, 2017, all children and staff from the 7 boarding schools were screened for TBI, and children with TBI were offered TPT. In 2018, children newly admitted to the schools were screened for TBI and offered TPT. In 2019, all children and staff previously not screened for TBI or negative for TBI were screened for TBI and offered TPT. Participants with history of previous TB treatment were not screened for TBI. In addition to the above annual schedule for screening for latent TBI, active case finding using symptom criteria and molecular diagnostics was performed twice a year for all students and staff in all the schools, one round conducted simultaneously with the screening for latent TBI and the second round conducted after 6 months. All TB cases detected during the screenings for latent TBI and active disease, and during the year, were recorded. Because of the school-wide screening, contacts of recent TB cases were automatically screened for TB disease. Screenings were timed such that TPT administration and completion happened at the schools before the start of schools’ annual winter breaks. Screening and diagnosis of TBI and TB disease and TPT implementation were carried out according to a prespecified algorithm (S1 Fig) [9]. Each child or staff member was interviewed by project staff using a standardized questionnaire, evaluated by a medical officer, and screened for TBI using tuberculin skin testing (TST) with 5 TU of purified protein derivative RT23 (Span Diagnostics, Surat, India), read after 2–3 days. TST-positive (≥10 mm) individuals had a chest X-ray, following which TPT was offered to those without evidence of active TB disease. Participants were evaluated for active TB using WHO symptom criteria (cough, fever, night sweats, weight loss, tiredness, chest pain), chest X-ray, Xpert MTB/RIF IV assay (Cepheid, Sunnyvale, California), and TB culture. Active TB was treated according to Tibetan and Indian guidelines with 2 months of daily isoniazid, rifampin, ethambutol, and pyrazinamide followed by 4 months of daily isoniazid, rifampin, and ethambutol for all drug-susceptible TB. According to India’s National Strategic Plan for Tuberculosis Elimination, TPT is encouraged for high-risk groups including child pulmonary TB contacts, people living with HIV, immunosuppressed patients, and high-risk adult contacts [10].


All participants detected with TBI were offered TPT, initially with 3 months of daily isoniazid and rifampicin (3HR) and later in the study with 4 months of daily rifampicin (4R), due to side effects encountered by children and staff with 3HR. Participants with exclusive contact with MDR-TB cases were not given TPT but underwent monitoring according to WHO guidelines. Participants with chronic hepatitis B virus (HBV) infection had liver function testing (serum bilirubin, alanine transaminase [ALT], and aspartate aminotransferase [AST]) and abdominal ultrasonographic exam to rule out liver parenchymal disease, according to the local standard of care. Liver function testing was not done for participants without HBV infection. HBV infection status was available with the school clinic as a part of their routine wellness check. Project staff conducted monthly follow-up of the TPT recipients. Hepatotoxicity was determined by the clinician based on development of (1) clinical symptoms of nausea, vomiting, loss of appetite, tiredness, upper abdominal pain, or jaundice; (2) elevated liver enzymes (≥5 times the upper limit of normal if asymptomatic or ≥3 times the upper limit of normal if symptomatic); and (3) subsidence of symptoms and laboratory values following treatment discontinuation or withholding. TPT was temporarily withheld in the event of hepatoxicity or other adverse drug reactions, as decided by the clinician, and then restarted, as appropriate. Those who had intolerance to 3HR were offered 4R, and those developing intolerance to 4R were offered 6 months of isoniazid. Therapy modification or premature termination was recommended by the physician in consultation with the participant, school nurse, home mothers, and parents.

Statistical analysis

Data from individual schools were separately maintained year-wise in secure databases. Demographic, clinical, treatment, and laboratory data of schoolchildren and staff members from enrollment to December 31, 2019, were available. A participant ID was issued to each participant and served as the unique identifier for the longitudinal data. Baseline characteristics of schoolchildren and staff members stratified by TPT receipt were separately described. Annual incidence of TB disease was calculated as total number of new TB cases in students for the year divided by total number of students registered in the school. In addition to incidence proportion, incidence rates—overall and stratified by TPT receipt—were calculated as TB disease development over child-years at risk in the specified time. The data were analyzed according to a prospectively specified statistical analysis plan (S1 Appendix).

For the primary analysis, we assessed whether schoolchildren who received TPT had a lower risk of developing active TB as compared to schoolchildren who did not receive TPT. Participants were followed from the time of enrollment into the project and censored at the earliest of the following: (1) first development of active TB disease, (2) graduation or exit from school, or (3) December 31, 2019. Hazard ratios were calculated using the Cox proportional hazard regression model, with 95% confidence intervals using the Huber–White robust estimator of the variance. Person-time was calculated as the total follow-up time in years for a person from enrollment into the project until censoring. Persons who developed active TB ceased to contribute person-time for the calculation of incidence rates or Cox regression. Persons who developed active TB were censored on the date of diagnosis. This is an open cohort, and children graduate from high school after grade 12. Otherwise, the dropout or loss to follow-up rate is less than 5%. In secondary analyses, we compared the hazard of TB disease for schoolchildren based on TPT receipt restricting the analysis to schoolchildren with recent TB contact, and compared the 2 TPT regimens, 3HR versus 4R. Recent contact was defined as contact with a TB case in the classroom or dormitory in school, or with a family member at home, in the previous 2 years. Age, TPT receipt status, TST positivity, TB contact history, and TB-related symptoms were time dependent. All models were adjusted for age and sex. In separate analyses, the hazard of TB disease development was independently modeled as a function of each of age, sex, enrollment year, contact history, TST positivity, TST conversion, and TB-related symptoms, separately. Prompted by feedback from peer reviewers, we further assessed the risk of TB disease or infection based on (1) whether TB exposure happened at school or at home, (2) contact with drug-resistant TB, and (3) bacille Calmette–Guérin (BCG) vaccination status. The relationship between participant characteristics or risk factors and risk of TBI among schoolchildren was modeled using logistic regression, calculating Huber–White robust standard error to account for clustering in the data. Insufficient follow-up years and disease outcomes precluded investigation of the effect of TPT in adult staff members. Risk of TBI for staff members as a function of the above risk factors has been previously described [9]. Data were processed and analyzed using Stata version 13.1 software (StataCorp, College Station, Texas). This study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (S1 STROBE Checklist).


Demographic characteristics

Between April 1, 2017, and December 31, 2019, 6,582 schoolchildren and 807 staff members participated in the program, contributing 13,161 and 1,800 person-years of follow-up, respectively (Table 1 and S1 Table). The median follow-up time was 2.3 years for students and 2.5 years for staff members. The majority of students (76%) were enrolled in 2017. Students were equally distributed between males and females, with a median age of 13 years. Three percent of students were previously treated for TB disease, and 28% reported a TB contact within the previous 2 years. The majority of the schoolchildren (87%) were BCG vaccinated.

Risk of TB disease and infection

There were 73 TB cases in total, 69 in schoolchildren and 4 in staff members (Fig 1). The median age of schoolchildren who developed TB was 16 years (Table 2). Of the 69 TB cases in students, 64 were pulmonary (60 Xpert-confirmed) and 1 was MDR-TB. Eighty percent (n = 55) of the TB cases in schoolchildren were TST positive, and 12% (n = 8) were TST negative. Fifty-nine percent (41/69) of the cases in schoolchildren did not have cough and 35% (24/69) were asymptomatic, representing a significant burden of subclinical disease (Table 3). Risk of disease increased with age (adjusted hazard ratio [aHR] 1.24; 95% CI 1.19–1.29; p < 0.001); recent contact history (aHR 2.55; 95% CI 1.52–4.27; p < 0.001); TST positivity (aHR 42; 95% CI 16.87–105.75; p < 0.001) (Fig 2); and TST conversion (aHR 3.00; 95% CI 1.37–6.56; p = 0.006) (Table 2). The risk of TB disease in children did not differ based on previous TB treatment (aHR 0.79; 95% CI 0.19–3.35; p = 0.748). Of 6,374 children and adolescents screened for TBI over 3 years, 1,412 (22%) were TST positive. Increasing age and male sex were associated with increased risk of TBI. Among children and adolescents who had recent exposure to TB (i.e., in the last 2 years) (n = 1,843), 1,610 (87%) had exposure at school, 141 (8%) had exposure at home, and 92 (5%) had exposure both at school and at home. We did not find a significant difference in the risk of TB disease or TBI based on exposure at home versus school. The risk of TBI was significantly higher for children who had exposure at both school and home (aHR 1.74; 95% CI 1.09–2.78; p = 0.021). BCG vaccination was not associated with any protective effect or increased risk of TBI and TB disease in the schoolchildren. There were 4 TB cases among staff members, of which 1 was extensively drug-resistant TB.


Fig 1. Care cascade for TBI screening and TPT in children and adolescents from 2017 through 2019.

TB, tuberculosis; TBI, tuberculosis infection; TPT, tuberculosis preventive treatment; TST, tuberculin skin testing.


TPT completion and safety

Of 6,374 children and adolescents screened for TBI over 3 years, 1,412 were TST positive, of whom 1,192 received TPT; 37 were not recommended TPT by the physician, and 183 refused TPT (Fig 1). There were 41 children with chronic HBV infection who were found to have latent TBI. Of these 41 children with TB and HBV coinfection, 11 (27%) did not receive TPT; they either refused treatment or were not recommended TPT by the clinician. None of the children with HBV infection who underwent liver function testing before initiation of TPT had elevated transaminases. Few of the children were receiving antiviral treatment for chronic HBV at the time of screening. While children and adolescents who did (n = 1,192) and did not (n = 5,390) receive TPT were mostly similar in characteristics, TB exposure was higher in TPT recipients (42% versus 25%). Of the 1,192 students and 125 staff members who received TPT, TPT completion information was available for 1,043 students and 115 staff; the remaining were undergoing TPT at the time of analysis (S2 Table). Of 1,043 students and 115 staff who received TPT, 98% (1,025) and 100% completed TPT, respectively. TPT completion was similar between 3HR and 4R for schoolchildren and staff. For both children (5–14 years) and adolescents (10–19 years), the prevalence of any side effects related to various body systems was significantly lower for 4R (range 0%–4%) than for 3HR (range 0%–14%) (S3 Table). Prevalence of hepatotoxicity was less than 0.5% (n = 3) for both 3HR and 4R regimens in schoolchildren.

TB incidence following TPT

The average annual incidence rate for schoolchildren over 13,161 person-years of follow-up was 524/100,000 (95% CI 414–663/100,000) person-years (Table 4). Among persons receiving TPT (2,164 person-years), 4 participants developed TB disease, and among persons not receiving TPT (10,997) person-years), 65 participants developed TB disease, yielding average annual TB incidence rates of 185/100,000 (95% CI 69–492/100,000) person-years and 591/100,000 (95% CI 464–754/100,000) person-years, respectively (p = 0.009). In multivariate analysis adjusted for age and sex, participants receiving TPT had a 79% lower hazard of TB disease development compared to those who did not receive TPT (aHR 0.21; 95% CI 0.07–0.69; p = 0.010) (Fig 3). The preventive effect was more pronounced for participants with a recent contact history, with incidence rates of 125/100,000 (95% CI 18–890/100,000) person-years for those receiving TPT versus 1,643/100,000 (95% CI 1,190–2,267/100,000) person-years for those who were not treated (aHR 0.07; 95% CI 0.01–0.42; p = 0.004). Of all the children and adolescents who received TPT, 4 developed TB disease, and they had all received 3HR. No 4R recipients developed TB disease. The TB that occurred in the 4 TPT recipients was drug susceptible. The children who developed TB post-TPT were adherent to treatment. The mean number of days to TB development after TPT completion was 302 days (SD 118 days). The TB incidence rate difference for use of 3HR versus 4R (0.00222/100,000 person-years; 95% CI 0.00004–0.00439/100,000) was not statistically significant (p = 0.384); however, the follow-up time was short for 4R (343 person-years). Four staff members developed TB disease over 1,800 person-years of follow-up (annual incidence rate 256/100,000 person-years; 95% CI 96–683/100,000); 1 case was extensively drug-resistant TB. No staff who developed TB disease had received TPT. We calculated a number needed to screen of 5 persons to detect 1 TBI, number needed to treat (NNT) with TPT of 248 persons for all schoolchildren to prevent 1 TB disease, and NNT of 64 for schoolchildren with recent TB contact.


Fig 3. Risk of TB disease for children and adolescents who did and did not receive TPT.

3HR, 3 months of daily isoniazid and rifampin; 4R, 4 months of daily rifampin; TB, tuberculosis; TPT, tuberculosis preventive treatment.


Impact on TB burden

Among schoolchildren who were annually screened for latent TBI, the prevalence of latent TBI in the Tibetan boarding schools decreased from 19% (913/4,860) in 2017 to 15% (124/827) in 2018 to 11% (383/3,608) in 2019 (Fig 4; Table 5). The overall incidence of TB disease in schoolchildren declined from 837/100,000 person-years (42/5,020) in 2017 to 431/100,000 person-years (23/5,339) in 2018 and 110/100,000 person-years (5/4,566) in 2019, an 87% decrease in incidence from 2017 to 2019.


Fig 4. Decline in tuberculosis (TB) burden in children and adolescents in the Tibetan boarding schools in India from 2017 through 2019.

(A) Decrease in annual incidence of TB disease. (B) Decrease in annual prevalence of TB infection. *Incidence of annual TB disease represents the proportion of new episodes of TB cases per 100,000 students per year among all students enrolled in the schools (Table 5). #Prevalence of annual TB infection represents the proportion of students testing positive for tuberculin skin test per year among all students at the schools who were not tested previously or had tested negative previously.



Motivated by the United Nations Sustainable Development Goals and India’s commitment toward ending TB, we undertook an initiative to eliminate TB in Tibetan refugee children by implementing a comprehensive screening and treatment program for both TBI and TB disease in a network of boarding schools. Between 2017 and 2019, we observed an average annual incidence rate of TB of 524/100,000 person-years in schoolchildren and 256/100,000 person-years in adult staff members in the Tibetan boarding schools in Himachal Pradesh. In the initial year of screening in 2017, an incidence of TB disease of 837/100,000 person-years and prevalence of TBI of 19% were observed in schoolchildren. After 3 years of program implementation, TB incidence declined by 87%, and prevalence of TBI declined by 42%, in children and adolescents in the schools. None of the staff members who received TPT developed TB disease while 4 staff members who did not receive TPT developed TB disease. The reductions in TB associated with 3HR and 4R TPT regimens were similar. TPT implementation had the greatest value in preventing TB cases in the first 2 years. Our work demonstrates that population-wide active case finding and TPT implementation using a community-based approach can effectively control TB in high-transmission settings where the chances of reinfection are high. Since 2011, active case finding of TB has been annually carried out in the Tibetan boarding schools under the Department of Health of the CTA. However, despite the effort, TB case rates have remained high over the years. Mass screening and TPT—by limiting disease progression, reactivation, and transmission—have driven this exponential reduction in TB rates. The successful control in this population is comparable to the reduction in TB incidence (>75%) achieved in Peru in 1985–1986 and during the community-wide isoniazid prophylaxis trial in the Bethel area in Alaska in the 1960s [11,12].

To date, very few studies have assessed TPT efficacy or effectiveness in children ≥ 5 years or in adolescents [13,14]. The reduction in TB disease risk of 79% associated with TPT use in our child and adolescent population is similar to or better than that observed in independent IPT trials in children 2 months–15 years (67%) in the Bethel area in Alaska in 1962 [15]; children 5–15 years (60%) in France between 1959 to 1969 [16]; children 0–15 years (60%) in the US, Canada, and Mexico in 1961 [17]; and children 0–15 years (62%) in Kenya in 1965 [18]. Applying the TB incidence of 837/100,000 person-years observed in the initial year in 2017 to the average annual student population of 5,000, 126 TB cases (42 annually) could have resulted over 3 years in the absence of the ZTBK program instead of the observed 69 cases. As such, 56 schoolchildren were spared from developing TB disease over 3 years. Despite less follow-up time for staff members, TPT showed protection in adult staff members as well; none of the adult staff who received TPT (n = 240 person-years) developed active TB whereas 4 TPT non-recipients (n = 1,560 person-years) developed active TB. The long-term benefit of TPT has been debated in the past. Investigators in South Africa observed waning of protection in a mass IPT intervention study in adult gold miners in South Africa [19,20]. This may be partly attributable to older age, higher HIV prevalence, high rates of silicosis, and constant interactions with a wider community harboring undiagnosed TB cases. In this study, we observed a sustained benefit over 3 years. The closed nature of the boarding schools, with their physical boundaries, restricted visitors, and less frequent interactions with the community may have helped to sustain the benefit of the TPT for our population. The overall reduction in TB rates observed in our study, including in those who did not receive TPT, suggests a herd benefit to other students, staff, and members of community.

Prevalence of recent TB exposure was high for schoolchildren (28%) and staff (13%). Three percent of schoolchildren and 19% of staff were treated for active TB in the past, reflecting a high disease burden and transmission in the community. In children and adolescents, TB risk was 42-fold higher for those with positive TST during the latest screening (p < 0.001). Of child and adolescent participants who had at least 1 longitudinal follow-up, 10% (347/3,643) converted from TST negative to positive over the 3-year period, reflecting an approximately 3.3% annual rate of acquiring new infection. The annual rate of acquiring new infections was 8% in the Bethel area in 1957, in keeping with its higher TB burden (annual incidence of approximately 1%) [21]. We observed a 3-fold higher disease risk in TST converters and 93% risk reduction from TPT in recent TB contacts. These findings underscore the importance of targeting recent TB contacts for preventive treatment in general. This is also evident from the lower number of schoolchildren who would need to be treated to prevent 1 TB disease in recent TB contacts (NNT = 68) as compared to that for the general population of schoolchildren (NNT = 253). Due to an insufficient number of outcomes of TB disease, we could not assess the rationale for TPT implementation in children ≥5 years with TB exposure without doing TST. With accumulated person-time, we intend to assess this in the future. Currently, WHO recommends TPT for child TB contacts <5 years of age regardless of TST status. Data from prevalence surveys across several communities in Africa have shown a high prevalence of TB disease (18.5%) among adults who were previously treated for TB [22]. We observed a similar prevalence of previous TB of 25% (1/4) in adults, although the sample size is small for adults in our study. For children and adolescents in our study, the prevalence of previous TB was 6.3%. We did not find a significant relationship between previous TB history and the risk of TB disease in children and adolescents.

Globally, the number of people with recent TBI is estimated at 56 million, as compared to the nearly 2 billion total people with latent TBI, making recent contacts a practically targetable population for TPT [23]. For intervention by active case finding alone, the number of people needed to screen to prevent 1 TB disease was calculated to be 1,002 in 1 study [24]. The data on TB rates in recent contacts and TST converters suggest ongoing TB transmission in schoolchildren, supporting the need to study TB transmission dynamics for appropriate public health intervention.

As a result of side effects to the 3HR TPT regimen reported by participants during the initial phase, the regimen was switched to 4R starting in March 2018. Tiredness, sleepiness, dizziness or headache, nausea, and epigastric discomfort were the more common side effects faced by the participants who received 3HR (S3 Table). The side effects were more prevalent in adolescents than in children less than 10 years old. We observed significantly fewer side effects associated with 4R as compared to 3HR for both children (5–14 years) and adolescents (10–19 years). No participant had a serious adverse event that required hospitalization. Risk of hepatotoxicity was very low (<0.5%) for both the TPT regimens. We observed high treatment completion (>95%) for both 3HR and 4R as a result of effective community engagement and mobilization, the dedication and hard work of program staff and school nurses, supervision by the home mothers, and the responsibility shouldered by individual participants. Although the study was not randomized for the comparison, we compared the 3HR and 4R regimens, and the difference in incidence rate reduction associated with the 3HR regimen compared to the 4R regimen was not statistically significant (95% CI 0.00004–0.00439/100,000 person-years; p = 0.384). Given the high completion rate, and better safety profile and similar reduction of TB as 3HR, 4R likely presents an effective treatment regimen for wide TPT implementation, and especially in adolescents and young adults, in whom isoniazid-related toxicity rates are higher.

A limitation of our study is that the reduction in TB incidence could be influenced by a secular trend in TB incidence during the study period. To test this possibility, we adjusted the hazard model (dependent variable: TB disease; independent variable: TPT receipt) for year of enrollment in addition to age and sex. The risk reduction from TPT use remained unchanged (aHR 0.21; 95% CI 0.07–0.68; p = 0.009) after this additional adjustment for calendar year. We considered the possibility that the reduction in TB disease in the second and third years could be partly due to a survival bias whereby schoolchildren having greater risk factors had developed TB disease earlier in the project period, resulting in an apparent reduction in incidence in later years. However, had there been no overall reduction in TB burden in the schools, these same children would also be at risk of reinfection and recurrence in the later years due to ongoing transmission. Therefore, it stands to reason that the reduction in TB incidence was largely the impact of our comprehensive program. Additionally, the global lack of a diagnostic tool prevented us from assessing the risk of TB reinfection for individuals who completed TPT after a positive test for TBI. We are unable to assess the reduction of TB disease as a result of TPT implementation in adult staff members due to a limited sample size. Inadequate exposure time for individual TPT regimens also precluded us from confidently comparing the effectiveness and safety of the regimens. In implementing this population-level initiative, some of the operational challenges we faced related to coordinating X-rays for children outside of schools and ensuring their safety, the need to perform gastric aspirations, especially given the high rate of subclinical disease, the inability to screen schools in parallel due to limited resources and program staff, coordinating activities vis-à-vis school schedule, and the lack of operational guidelines for implementation of TPT. These and additional challenges have been described in more detail previously [9].

Despite clinical studies and mathematical models showing the efficacy, effectiveness, and impact of TPT [2527], its implementation has been neglectfully suboptimal [1]. Our study sheds light on various aspects of TPT implementation including screening strategies, treatment effectiveness, choice of regimens, and safety profiles that could inform national and global guidelines, specifically for children and adolescents. For the first time, to our knowledge, the TB epidemic curve is bent for this vulnerable refugee children population, setting the community on a path to TB elimination, and presenting a model for communities nationally and globally.


  1. 1.
    World Health Organization. 2019 global TB report. Geneva: World Health Organization; 2019 [cited 2020 Feb 11]. https://www.who.int/tb/publications/global_report/en/.
  2. 2.
    Snow KJ, Sismanidis C, Denholm J, Sawyer SM, Graham SM. The incidence of tuberculosis among adolescents and young adults: a global estimate. Eur Respir J. 2018;51(2):1702352. pmid:29467206
  3. 3.
    Jenkins HE, Yuen CM, Rodriguez CA, Nathavitharana RR, McLaughlin MM, Donald P, et al. Mortality in children diagnosed with tuberculosis: a systematic review and meta-analysis. Lancet Infect Dis. 2017;17(3):285–95. pmid:27964822
  4. 4.
    Dodd PJ, Gardiner E, Coghlan R, Seddon JA. Burden of childhood tuberculosis in 22 high-burden countries: a mathematical modelling study. Lancet Glob Health. 2014;2(8):e453–9. pmid:25103518
  5. 5.
    Dodd PJ, Yuen CM, Sismanidis C, Seddon JA, Jenkins HE. The global burden of tuberculosis mortality in children: a mathematical modelling study. Lancet Glob Health. 2017;5(9):e898–906. pmid:28807188
  6. 6.
    Dierberg KL, Dorjee K, Salvo F, Cronin WA, Boddy J, Cirillo D, et al. Improved detection of tuberculosis and multidrug-resistant tuberculosis among Tibetan refugees, India. Emerg Infect Dis. 2016;22(3):463–8. pmid:26889728
  7. 7.
    Nelson LJ, Naik Y, Tsering K, Cegielski JP. Population-based risk factors for tuberculosis and adverse outcomes among Tibetan refugees in India, 1994–1996. Int J Tuberc Lung Dis. 2005;9(9):1018–26. pmid:16158895
  8. 8.
    Salvo F, Dorjee K, Dierberg K, Cronin W, Sadutshang TD, Migliori GB, et al. Survey of tuberculosis drug resistance among Tibetan refugees in India. Int J Tuberc Lung Dis. 2014;18(6):655–62. pmid:24903934
  9. 9.
    Dorjee K, Topgyal S, Dorjee C, Tsundue T, Namdol T, Tsewang T, et al. High prevalence of active and latent tuberculosis in children and adolescents in Tibetan schools in India: the Zero TB Kids initiative in Tibetan refugee children. Clin Infect Dis. 2019;69(5):760–8. pmid:30462191
  10. 10.
    Central TB Division. National strategic plan for tuberculosis elimination 2017–2025. New Delhi: Ministry of Health and Family Welfare; 2017 [cited 2020 Oct 27]. https://tbcindia.gov.in/WriteReadData/NSP%20Draft%2020.02.2017%201.pdf.
  11. 11.
    Accinelli RA, Romero LR, Garcia RF, Sanchez R. Sustained benefit of community-based tuberculosis interventions after 30 years. Am J Respir Crit Care Med. 2015;191(10):1202–3. pmid:25978574
  12. 12.
    Comstock GW, Baum C, Snider DE Jr. Isoniazid prophylaxis among Alaskan Eskimos: a final report of the bethel isoniazid studies. Am Rev Respir Dis. 1979;119(5):827–30. pmid:453704
  13. 13.
    Ayieko J, Abuogi L, Simchowitz B, Bukusi EA, Smith AH, Reingold A. Efficacy of isoniazid prophylactic therapy in prevention of tuberculosis in children: a meta-analysis. BMC Infect Dis. 2014;14:91. pmid:24555539
  14. 14.
    Gray DM, Zar H, Cotton M. Impact of tuberculosis preventive therapy on tuberculosis and mortality in HIV-infected children. Cochrane Database Syst Rev. 2009(1):CD006418. pmid:19160285
  15. 15.
    Comstock GW. Isoniazid prophylaxis in an undeveloped area. Am Rev Respir Dis. 1962;86:810–22. pmid:14022524
  16. 16.
    Debre R, Perdrizet S, Lotte A, Naveau M, Lert F. Isoniazid chemoprophylaxis of latent primary tuberculosis: in five trial centres in France from 1959 to 1969. Int J Epidemiol. 1973;2(2):153–60. pmid:4204799
  17. 17.
    Mount FW, Ferebee SH. Preventive effects of isoniazid in the treatment of primary tuberculosis in children. N Engl J Med. 1961;265:713–21. pmid:14476667
  18. 18.
    Egsmose T, Ang’awa JO, Poti SJ. The use of isoniazid among household contacts of open cases of pulmonary tuberculosis. Bull World Health Organ. 1965;33(3):419–33. pmid:5321762
  19. 19.
    Churchyard GJ, Fielding KL, Lewis JJ, Coetzee L, Corbett EL, Godfrey-Faussett P, et al. A trial of mass isoniazid preventive therapy for tuberculosis control. N Engl J Med. 2014;370(4):301–10. pmid:24450889
  20. 20.
    Golub JE, Cohn S, Saraceni V, Cavalcante SC, Pacheco AG, Moulton LH, et al. Long-term protection from isoniazid preventive therapy for tuberculosis in HIV-infected patients in a medium-burden tuberculosis setting: the TB/HIV in Rio (THRio) study. Clin Infect Dis. 2015;60(4):639–45. pmid:25365974
  21. 21.
    Comstock GW, Ferebee SH, Hammes LM. A controlled trial of community-wide isoniazid prophylaxis in Alaska. Am Rev Respir Dis. 1967;95(6):935–43. pmid:6026165
  22. 22.
    Marx FM, Floyd S, Ayles H, Godfrey-Faussett P, Beyers N, Cohen T. High burden of prevalent tuberculosis among previously treated people in Southern Africa suggests potential for targeted control interventions. Eur Respir J. 2016;48(4):1227–30. pmid:27390274
  23. 23.
    Houben RM, Dodd PJ. The global burden of latent tuberculosis infection: a re-estimation using mathematical modelling. PLoS Med. 2016;13(10):e1002152. pmid:27780211
  24. 24.
    Marks GB, Nguyen NV, Nguyen PTB, Nguyen TA, Nguyen HB, Tran KH, et al. Community-wide screening for tuberculosis in a high-prevalence setting. N Engl J Med. 2019;381(14):1347–57. pmid:31577876
  25. 25.
    Dye C, Glaziou P, Floyd K, Raviglione M. Prospects for tuberculosis elimination. Annu Rev Public Health. 2013;34:271–86. pmid:23244049
  26. 26.
    Dowdy DW, Golub JE, Saraceni V, Moulton LH, Cavalcante SC, Cohn S, et al. Impact of isoniazid preventive therapy for HIV-infected adults in Rio de Janeiro, Brazil: an epidemiological model. J Acquir Immune Defic Syndr. 2014;66(5):552–8. pmid:24853308
  27. 27.
    Golub JE, Saraceni V, Cavalcante SC, Pacheco AG, Moulton LH, King BS, et al. The impact of antiretroviral therapy and isoniazid preventive therapy on tuberculosis incidence in HIV-infected patients in Rio de Janeiro, Brazil. AIDS. 2007;21(11):1441–8. pmid:17589190


This site uses Akismet to reduce spam. Learn how your comment data is processed.