Evaluation of a school-based program for diagnosis and treatment of latent tuberculosis infection in immigrant children

Journal of Infection and Public Health (2010) 3, 67—75 Evaluation of a school-based program for diagnosis and treatment of latent tuberculosis infection in immigrant children Philippe Minodier a,c, Valérie Lamarre a, Marie-Eve Carle b, Denis Blais a, Philippe Ovetchkine a, Bruce Tapiero a,b,∗ a Infectious Diseases Division, Department of Pediatrics, CHU Sainte Justine — Université de Montréal, 3175 Côte Sainte Catherine, Montréal, QC, Canada H3T 1C5 b Intercultural Pediatric Unit, Department of Pediatrics, CHU Sainte Justine — Université de Montréal, 3175 Côte Sainte Catherine, Montréal, QC, Canada H3T 1C5 c Pediatric Emergency, CHU Nord, Chemin des Bourrelly, 13015 Marseille, France Received 14 January 2010 ; received in revised form 5 February 2010; accepted 8 February 2010 KEYWORDS Latent tuberculosis infection; Child; Mass screening; Emigration and immigration; Tuberculin skin test; Patient compliance; Patient acceptance of health care Summary Objective: To evaluate a 10-year school-based latent tuberculosis infection (LTBI) screening program, targeting immigrant children in Montreal, Canada, and to identify predictive factors for refusal and, poor adherence to treatment. Methods: Immigrant children were screened for LTBI with Tuberculin Skin Test (TST). Isoniazid was, given when LTBI was diagnosed. Predictors of LTBI, of refusal of followup and treatment and of poor, adherence to isoniazid were analyzed. Results: Four thousand three hundred and seventy-five children were offered screening, 82.3% consented to TST and 22.8% were positive. An, older age at migration (odds ratio (OR) = 1 [95% CI: 1.0—1.01]), as well as migration from a none, established market economy country (OR varying from 2.41 to 4.23) were significantly associated with, positive TST. Among positive children, further evaluation was refused in 5.7%, mainly in migrants from, Eastern Europe (OR = 4.05 [95% CI: 2.14—7.69]). Refusal of treatment (11.2%) was more frequent in, Eastern European when compared to Southeastern Asian (OR = 6.91 [95% CI: 1.56—30.75]), in, blended families (OR = 3.25 [95% CI: 1.25—8.46]) and when the first visit to hospital was delayed (OR = 1.01 [95% CI: 1.0—1.02]). Adequate completion of treatment was noted in 61.3%. Age > 16 years (OR = 1.82 [95% CI: 1.82—2.99]), a delay between TST and first visit > 15 days (OR = 1.6 Abbreviations: TB, tuberculosis; LTBI, latent tuberculosis infection; TST, tuberculin skin test; BCG, bacillus Calmette-Guérin; OR, odds ratio; CI, confidence interval. ∗ Corresponding author at: Intercultural Pediatric Unit, Department of Pediatrics, CHU Sainte Justine—Université de Montréal, 3175 Côte Sainte Catherine, Montréal, QC, Canada H3T 1C5. Tel.: +1 514 345 4931; fax: +1 514 345 4822. E-mail address: bruce tapiero@ssss.gouv.qc.ca (B. Tapiero). 1876-0341/$ — see front matter © 2010 King Saud Bin Abdulaziz University for Health Sciences. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jiph.2010.02.001 68 P. Minodier et al. [95% CI: 1.12—2.28]), as well as the presence of relative > 18 years in the household (OR = 1.56 [95% CI: 1.0—2.43]), were associated with poor adherence to treatment. Conclusion: Sociocultural and behavioural factors are involved in acceptance of LTBI treatment in, immigrant children. Adherence to treatment is challenging and requires comperhension of sociocultural beliefs and accessibility to TB clinic. © 2010 King Saud Bin Abdulaziz University for Health Sciences. Published by Elsevier Ltd. All rights reserved. Targeted tuberculin skin testing is an important part of tuberculosis (TB) control strategy in low burden countries [1,2]. Latent tuberculosis infection (LTBI), which is defined as a positive tuberculin skin test (TST) and no evidence of TB disease, has to be detected and treated to prevent extension and dissemination of TB, especially in children. Indeed, lifelong risk of reactivation is about 10—20% in young children and adolescents with TST ≥10 mm [3]. LTBI is frequent in people emigrating from highly endemic countries [1,4]. In Québec, Canada, the incidence of active TB was 3.4/100,000 in 2003 but almost 7 times higher in areas of Montreal, a city in which immigrants concentrate [5]. Although foreign-born people represent only 9.9% of Québec’s population, they account for 61.7% of TB cases. In immigrant children, active TB annual incidence was 43/100,000 in 15—19-year olds and 6/100,000 in 1—4-year olds in 2000—2003 [5]. Considering the TB risk in immigrant children, a prospective school-based TB-screening program was implemented in 1997 at Sainte Justine Hospital, a large pediatric tertiary care teaching hospital in Montréal, Québec (Canada). The program targeted immigrant children living in the most densely populated area of the city. In this area, immigrants actually represent 45% of the total population, and half of them have entered Canada in the last 10 years. Although 45% of immigrants had some university education, half have an annual income <15000 Ca$ [6]. In Québec, most immigrant children are not fluent in French and are placed in welcoming classes to learn the common language, whereas French-speaking immigrants usually attended regular classes. The TB-screening program selected both welcoming classes and regular classes with immigrants. A 10-year analysis of the program was performed. LTBI was defined and treated in accordance to current Canadian guidelines [7]. As adherence is challenging in pediatric LTBI treatment [8—15], we tried to identify factors contributing to acceptance and to completion of treatment. Methods The TB-screening program In selected classes, school nurses informed children and parents about TB-screening by mail. Written parents’ consent was collected before testing for children aged less than 14 years old. On the planned day, consent of each child was required. A trained nurse from the TB-Clinic of Sainte Justine Hospital visited the children at school and asked them for their age, country of birth, age at emigration, and foreign countries where they had resided. Children in fifth and sixth grades (10—12 years old) in primary schools were specifically targeted for screening. Only two of those trained nurses were involved in the program during the period. They also performed TST using the Mantoux technique, according to Canadian tuberculosis standards [16]. Five Tuberculin Units (0.1 ml) of purified protein derivative standard (Tubersol® , Sanofi Pasteur) were injected in the dermis of the left forearm. TST were read using the ‘‘ball point’’ technic, 48 h later at school by the TB-nurse. A positive TST was defined as an induration ≥10 mm in transverse diameter at the site of injection. Previous bacillus Calmette-Guérin (BCG) vaccination was not taken into account for TST interpretation. On the day of TST reading, children with positive TST were scheduled for a follow-up appointment in the TB-Clinic of Sainte Justine Hospital. The day of the first visit at the TB-Clinic, all the TST positive children underwent a chest X-ray and a medical evaluation. Chest X-rays were read by a paediatric radiologist. Gender, age, age at time of immigration, country of birth, foreign travels, native language, ability to speak French or English, type of family (presence of the parents, relatives >18 years living in the household), siblings under 18 years living in the household, refugee status (Canadian, permanent resident, refugee), known contact with TB-infected persons, BCG vaccination and previous TB evaluation and treatment were recorded with families. BCG vaccine status was evaluated Immigrants with latent tuberculosis infection looking at any existing documentation, such as a vaccination record, and/or by the presence of scars on BCG vaccine common sites. Patients with active TB were not included in the study. Children with LTBI were offered treatment with isoniazid 10—15 mg/kg/d (max. 300 mg/d), if not contraindicated by history and physical examination. The planned duration of treatment was 6 months from 1997 to 2000, then changed to 9 months following a report of the Centers for Disease Control & Prevention [17]. If treatment was accepted, a 6 (then 9)-month isoniazid supply was prescribed and distributed each month. Visits at TB-Clinic were scheduled at 1, 3, 6 and 9 months of treatment. When an appointment was missed, a TB-clinic nurse phoned the family to schedule another appointment. Treatment withdrawal was considered when no contact with family was obtained after two phone calls and two letters. When INH treatment was refused, children were discharged after medical counseling. Adherence to isoniazide was evaluated by: 1) by self-reported adherence at each appointment (doses usually taken during a week, i.e. 0—7 doses/7 days), 2) by pills counts whenever possible, 3) by phone calls to pharmacies during treatment in order to verify regular medication supplies. Adherence was considered adequate if the child took ≥80% of the prescribed doses in a time ≤120% of the duration planned at first visit (i.e. 29 weeks for a 6 months treatment and 43 weeks for a 9 months one). The threshold of 80% of the dose is commonly chosen for good adherence [18]. As others [15], we have added limits in the duration of treatment to include children that had taken all the doses in a slightly longer time. However, when it took a very long time (>120% of the planned duration) to assess ≥80% of the prescribed doses, children were considered poor compliant. Complaints during treatment were noted at each appointment. Finally, when the treatment was complete, a certificate was given and the child was discharged. Data collection Data was collected prospectively using the standard TB-Clinic form sheets, which were then transferred anonymously onto a computerized database (SPSS 15.0 for Windows). Countries of migration were grouped according to WHO subregions [19]. Complaints during treatment were compiled with no regard to potential causality. This study was approved by the local ethics committee and the institutional review board. 69 Statistics SPSS 15.0 for Windows was used for statistics. For categorical data and group means comparisons, we used a binary regression model. Independent clinical and biological data that were significant or close to significance in univariate analysis were then retained for a multivariate logistic regression analysis. The significance p value always was p < 0.05. Results From 1997 to 2007, 4375 children in 21 schools were offered TB-screening. Seven hundred and seventythree (17.7%) refused testing. Three thousand six hundred and two consenting children received TST and 3401 tests (94.4%) were read, of which 777 (22.8%) were ≥10 mm. The mean age of screened children was 146 (S.D. ±33) months, 52% were males. Data was available for analysis in 724 children (93.2%). Forty-one children (5.7%) refused the first TB-clinic visit. Among the 683 evaluated children, none had chest X-ray compatible with active TB. BCG vaccination was reported in 89%. Thirty-eight children (5.6%) had been previously treated for LTBI and were discharged. Seventy-two (11.2%) refused to be treated. Isoniazid was prescribed in 573 children. Twenty-eight children (4.9%) were referred to another hospital during the follow-up. Among the 545 evaluable children (70.1% of all the positive TST), 334 (61.3%) finally completed the treatment with an adequate adherence. Fig. 1 summarizes the school-based TB-screening. Risk factors for LTBI Compared to children born in an established market economy country (Canada, United States, Western Europe and Israel mainly), children emigrating from Eastern Europe, African countries with high HIV prevalence, Latin America, the Western Pacific region, Central Europe, African countries with low HIV prevalence, the Eastern Mediterranean region or South-eastern Asia were significantly more likely to have TST ≥10 mm, both in univariate and multivariate analysis (Fig. 2). Mean ages at emigration was 118 (S.D. ±49) months. Rates of positive TST increased with age at emigration (Fig. 3). Odds ratio (OR) and 95% confidence intervals (95% CI) are reported in Table 1. Compared to TST negative children by a multivariate analysis, mean age at time of emigration was significantly higher among 70 P. Minodier et al. Figure 1 School-based LTB1 screening program (1997—2007). Figure 2 Number of children tested according to WHO subregions of migration and percents of TST ≥ 10 mm. Immigrants with latent tuberculosis infection 71 Figure 3 Rates of TST ≥ 10 mm according to age at time of migration. Table 1 Features associated with a positive TST risk (TST ≥ 10 mm). TST ≥ 10 mm TST < 10 mm Univariate analysis, OR [95% CI] Multivariate analysis, OR [95% CI] Country of migration (n/%)a Established market economy South east Asia East Mediterranean Africa low HIV Central Europe Western Pacific Latin America Africa high HIV East Europe 29 (2.7) 92 (26) 59 (27.1) 34 (31.5) 16 (32) 203 (32.8) 146 (37.6) 19 (38) 174 (39.4) 1062 (97.3) 262 (74) 159 (72.9) 74 (68.5) 34 (68) 416 (67.2) 242 (62.4) 31 (62) 268 (60.6) 1 12.86 13.59 16.83 17.23 17.87 22.09 22.45 23.78 1 2.41 [1.15—5.06] 2.65 [1.23—5.69] 3 [1.33—6.78] 3 [1.18—7.64] 3.44 [1.67—7.08] 4.03 [1.93—8.39] 3.9 [1.57—9.73] 4.23 [2.04—8.77] Mean age at TST (months) 153 (±32) 144 (±32) [1.01—1.01] Mean age at emigration (months) 126 (±48) 114 (±49) [8.29—19.94] [8.45—21.85] [9.72—29.12] [8.56—34.68] [11.92—26.8] [14.48—33.7] [11.37—44.3] [15.7—36.02] 1.01 [1—1.01] 1 [1—1.01] (NS) 1 [1—1.01] NS: non-significant. a Grouped according to WHO subregion [19]. TST positive children (126 (S.D. ±48) months versus 114 (S.D. ±49) months, OR = 1 [95% CI: 1—1.01]). Mean age at time of the TST testing was significantly higher among positive children in univariate analysis only (153 (S.D. ±35) months versus 144 (S.D. ±32) months, OR = 1.01 [95% CI: 1.01—1.01]). Refusal of the initial visit at TB-Clinic Forty-one children (5.7% of evaluable positive children) refused to visit the TB-Clinic. Eastern European people were more likely to refuse the visit than others (OR = 4.05 [95% CI: 2.14—7.69]) whereas those from the Western Pacific were more likely to accept (OR = 0.21 [95% CI: 0.07—0.69]). Russian (5.5% of TST positive but 12.2% of refusals), Ukrainian (3.5% and 12.2%), Moldavian (2.3% and 9.8%) and Bulgarian (3.1% and 7.3%) children were the most frequent to refuse the TB-Clinic visit. Refusal of treatment Treatment was refused by 72 children (11.2% of children visiting TB-Clinic and not previously treated). In children emigrating from South-eastern Asia and Eastern Europe, treatment refusal rates were respectively 3.6% and 23.7%. In univariate analysis, significant features for refusal were emigrating from Eastern Europe (when compared to Southeastern Asia), living with one of the parents and a parent in law (blended family) and having no or only one sibling <18 years at home. The mean 72 P. Minodier et al. Table 2 Features associated with refusal of treatment (non-significant features in univariate analysis were not reported in the table). Treatment’s refusal, n (%) or mean (S.D.) Treatment’s acceptance, n (%) or mean (S.D.) Univariate analysis, OR [95% CI] Multivariate analysis, OR [95% CI] 3 (4.2) 32 (44.4) 81 (13.3) 103 (16.9) 1 8.39 [2.48—28.38] 1 6.91 [1.56—30.75] Blended family 7 (12.7) 23 (3.8) 3.67 [1.50—8.98] 3.25 [1.25—8.46] >2 children under 18 at home 6 (10.9) 191 (31.6) 0.27 [0.11—0.63] 0.42 [0.17—1.05] (NS) 1.01 [1—1.02] 1.01 [1—1.02] Place of migration South east Asia East Europe Mean delay TST/first visit 28.6 (32.5) 21.8 (20.5) NS: non-significant. Table 3 Features associated with poor adherence to LTBI treatment (non-significant features in univariate analysis were not reported in the table). Age >16 years Delay TST-Visit >15 days Presence relatives >18 years in household Adequate completion, n (%) Incomplete completion, n (%) Univariate analysis, OR [95% CI] Multivariate analysis, OR [95% CI] 37 (11.1) 166 (50.2) 54 (16.2) 40 (19) 129 (61.1) 49 (23.3) 1.88 [1.16—3.05] 1.56 [1.1—2.22] 1.58 [1.02—2.43] 1.82 [1.11—2.99] 1.6 [1.12—2.28] 1.56 [1.0—2.43] Adequate completion was defined as ≥80% of total dose in a duration ≤120% of these planned at first visit. delay between the TST and the first visit to the TBClinic was also significantly longer among refusing children. In multivariate analysis, significant independent factors for refusal were Eastern European origin (OR = 6.91 [95% CI: 1.56—30.75]), blended family (OR = 3.25 [95% CI: 1.25—8.46]) and mean delay between TST and first visit (OR = 1.01 [95% CI: 1—1.02]) (Table 2). Completion of treatment Of the 545 children evaluated for adherence, 61.3% adequately completed the treatment. Factors decreasing adherence were age >16 years, delay between TST and first visit >15 days and the presence of relatives >18 years living in the household in both univariate and multivariate analysis. Origin of migration, planned treatment’s duration (6 or 9 months), formulation of isoniazid (pills or syrup) were, among other factors, not significant for completion (Table 3). During treatment, 25.1% of the patients reported complaints. The most frequent were abdominal pain (4.7%), headaches (4.5%), cough (4.3%), weight loss (2.8%), asthenia (2%) and vomiting (1.9%). Complaints were not correlated to age or gender. Their presence did not affect compliance. No case of hepatic toxicity or serious adverse events were noted. Discussion In pediatrics, the goal of TB-screening is to identify children with LTBI who are at risk for progression to active TB [1,2]. Early treatment of LTBI prevents extended and disseminated disease [1,4]. Universal TB school-testing [20,21], with a reported positivity rate of 2—3%, is not cost-effective [22]. Conversely, targeted programs have been shown to be costeffective if thresholds of 20% positive children and 60% completion rates are reached [2,23]. Our targeted school-based screening program achieved a TST positivity rate of 22.8% and good compliance in 61.8%, therefore fulfilling these criteria. Our rate of LTBI was identical to other school-based series [8,15], but, with better case selection, it could have been improved. We chose to target at-risk classrooms rather than at-risk children [7] in order to rapidly test a large number of children. However, the percentage of positive TST would have been higher in regular classes if we had tested immigrant children only. Indeed, children coming from an established market economy country represented 31.8% of tested children, but only 3.7% of positive ones. If these low-risk children had been excluded from the TB-screening, the rate of positive TST in the program would have been 32.2%. Previous studies also found that positive TST was associated with migration from a highly endemic Immigrants with latent tuberculosis infection country [8,20,21,24—27]. As reported by others [8], the risk of having a positive TST was higher in children that had resided in an endemic country for a longer period of time. Older age at time of testing was not significant in our multivariate analysis, but was reported as a predictor by others [20,21,25,28] and in adults [29]. In only one study of children born in Mexico and immigrating to the United States, positive TST rate and age were not associated [27]. As our program targeted fifth and sixth grades in primary schools, the lack of significance for age may be related to the age homogeneity in our targeted group. To improve the accuracy of the testing in regular classes, one could also use a previously validated questionnaire [30,31] including contacts with cases of TB, country of origin or travels (child and household’s members) in TB-endemic areas, regular contacts with adults at risk for TB or known HIV infection. Educational levels of parents and ethnicity have also been reported as predictors of positive TST [30]. In such approach, only at-risk children would require TST and the program’s effectiveness would subsequently improve. We reported that 89% of positive children were vaccinated with BCG, often at birth. Vaccine status was not collected in negative children. As a great number of negative children were born in Canada (where BCG is neither mandatory nor recommended), we can hypothesize that BCG vaccination rate was low in this group. Previous studies [20,21,26,27] report BCG vaccination as a predictor for positive TST, especially in younger foreign-born children. In our series, the rate of a positive TST increased with age at migration. We believe, as others [30], that the positive TST is more likely related to an increased duration of TB exposure in the country of birth, rather than to previous BCG vaccination. We report a refusal of initial evaluation at the TB-Clinic and treatment refusal in 16.9% of the cases. Eastern Europeans were more likely to decline the treatment. Communication was not a barrier to evaluation or treatment, as interpreters were available when needed. The refusal might be related to socio-cultural factors: perception of the disease, personal beliefs concerning the risks of LTBI, or parent perception that the previous BCG vaccination protected their child. Other parent refusals seemed based on the perception that the positive TST reflected the previous BCG vaccination history rather than LTBI. In some European countries, TST is used to verify the BCG vaccine immunogenicity and a positive result is said to confer protection. The reason for increased refusal among blended families in our series was unclear. 73 Perhaps, fear of conflict with the other parent resulting from a positive test and treatment might have been an influential factor. It is also possible that familial structure reflects socio-economical levels and related beliefs. Socio-anthropologic studies are required for a better understanding of TB perception and related behaviours among newly immigrating populations. Quality improvement in LTBI management requires a timely evaluation [4,32]. We have found that a long delay between TST and the TB-clinic visit was associated with treatment refusal and poor compliance. Delayed management could lead parents to doubt the seriousness of a positive LTBI [4]. Our treatment completion rate in LTBI children was 61.8%. In previous studies, it varied from 20 to 27.6% in south African children <5 years [13,14], to 56.8% in Israelis 12—13 years old [10], 50—80% in foreign-born adolescents living in the United States [11,12] and 82—92% in children recently immigrating to Canada [8,9,15]. Our lower rate of treatment completion compared to other Canadian series can be explained by our rigid definition for adequate adherence: children taking >80% of the total dose within 120% of the allotted time. However, using a larger definition for completion in children treated during 9 months (>80% of the total dose within 1 year), adherence rate poorly differed (65.1%). Other studies have tried to determine factors for failure of LTBI treatment completion, both in adults [29,33—43] and in children [12,14,15,44,45]. Somatic complaints during isoniazid treatment have been reported in poorly adherent adolescents, especially in females [46], but we did not find such an association. In our study, significant social variables for poor completion were age >16 years, and the presence of relatives >18 years in the household. We believe this is related to age homogeneity in our population and to the collection of socio-demographic features rather than psychosocial ones. Others have reported good compliance to be associated with adherence to social norms and behaviours and to acculturation processes (as reflected by a lower age, a higher grade, a better parents’ stability and supervision (especially for the youngest children), bicultural backgrounds, or lesser risk behaviours reported in compliant children) [45]. Confidence in the self-ability to adhere treatment and expectations of health benefits also play a role in treatment completion. Incentives have been used to improve compliance [47,48]. They have been effective when combined with nurse management in adults [49] or with peer counseling in adolescents [48]. Short- 74 term therapy with isoniazid and rifampicin for 3—4 months may also improve adherence to treatment [4,50] and has been shown to be as effective as 6—12 months of isoniazid [51]. In-school supervised administration of treatment could also boost treatment completion rate. In the future, specificity may be improved by combining interferon ␥-based assays and TST [52,53]. The high cost of the assay might be counterbalanced by a reduction in the number of patients treated. In summary, a school-based TB-screening program is effective if targeted towards recent immigrant children. Factors of refusal of testing and treatment seem essentially related to beliefs and behaviours concerning protection by BCG vaccination, risks of LTBI, and ability of TST to detect disease. Young age contributed to compliance. Improving adherence to treatment requires a comprehension of socio-cultural beliefs and behaviours involved in LTBI, as well as accessibility to TBClinics. Acknowledgements The authors want to acknowledge Dr. Lorne Aaron for manuscript reading, Dr. Guilhem Noël for his help in SPPS software use, Suzanne Desrochers and Patricia Germain for their help in TB-Clinic’s organization and in data collection, nurses and physicians of the TB-Clinic at Sainte Justine Hospital, and nurses of local community health centers involved in children management in schools. Conflict of interest statement The authors declare that they have no conflict of interest. References [1] Pediatric Tuberculosis Collaborative Group. Targeted tuberculin skin testing and treatment of latent tuberculosis infection in children and adolescents. Pediatrics 2004;114:1175—201. [2] Wobeser W, Yuan L, Gushulak B. Surveillance and screening in tuberculosis control. In: Long R, Ellis E, editors. Canadian tuberculosis standards. 6th ed. Ottawa, ON: Minister of Health; 2007. p. 274—97. [3] Horsburgh Jr CR. Priorities for the treatment of latent tuberculosis infection in the United States. N Eng J Med 2004;350:2060—7. [4] American Thoracic Society. Targeted tuberculin testing and treatment of latent tuberculosis infection. Am J Respir Crit Care Med 2000;161:S221—47. [5] Comité Québécois sur la tuberculose. Epidémiologie de la tuberculose au Québec de 2000 à 2003. ed. Québec, QC: Ministère de la Santé et des Services Sociaux du Québec; 2005. P. Minodier et al. [6] Portrait des populations immigrante et non immigrante, 2001. Arrondissement Côte-des-Neiges-Notre-Damede-Grâce. http://ville.montreal.qc.ca [updated July 2008]. [7] Kitai I, Malloy P, Kowalczyk A. Pediatric tuberculosis. In: Long R, Ellis E, editors. Canadian tuberculosis standards. 6th ed. Ottawa, ON: Minister of Health; 2007. p. 182—99. [8] Yuan L, Richardson E, Kendall PRW. Evaluation of a tuberculosis screening program for high-risk students in Toronto schools. Can Med Assoc J 1995;153:925—32. [9] Doering D, Kocuipchyk R, Lester S. A tuberculosis screening and chemoprophylaxis project in children from a high risk population in Edmonton, Alberta. Can J Public Health 1999;90:152—5. [10] Bibi H, Weiler-Ravell D, Shoseyov D. Compliance to treatment of latent tuberculosis infection in a region of Israel. Isr Med Assoc J 2002;4:13—6. [11] Hovell MF, Sipan CL, Blumberg EJ. Increasing Latino adolescents’ adherence to treatment for latent tuberculosis infection: a controlled trial. Am J Public Health 2003;93:1871—7. [12] Coly A, Morisky D. Predicting completion of treatment among foreign-born adolescents treated for latent tuberculosis in Los Angeles. Int J Tuberc Lung Dis 2004;8:703—10. [13] Marais BJ, Van Zyl S, Schaaf HS. Adherence to isoniazid preventive chemotherapy: a prospective community based study. Arch Dis Child 2006;91:762—5. [14] Van Zyl S, Marais BJ, Hesseling AC. Adherence to antituberculosis chemoprophylaxis and treatment in children. Int J Tuberc Lung Dis 2006;10:13—8. [15] Brassard P, Steensma C, Cadieux L. Evaluation of a school-based tuberculosis-screening program and associate investigation targeting recently immigrated children in a low-burden country. Pediatrics 2006;117:e148—56. [16] Menzies D, Khan K. Diagnosis of tuberculosis infection and disease. In: Long R, Ellis E, editors. Canadian tuberculosis standards. 6th ed. Ottawa, ON: Minister of Health; 2007. p. 53—91. [17] Center for Disease Control and Prevention. Targeted tuberculin testing and treatment of latent tuberculosis infection. MMWR 2000;49(RR-6):26—39. [18] Hoeppner VH, Ward H, Elwood K. Treatment of tuberculosis disease and infection. In: Long R, Ellis E, editors. Canadian tuberculosis standards. 6th ed. Ottawa, ON: Minister of Health; 2007. p. 114—45. [19] World Health Organization. Global tuberculosis control: surveillance, planning, financing. WHO Report 2007. Geneva: World Health Organization (WHO/HTM/TB/2007.376). [20] Scholten JN, Fujiwara PI, Frieden TR. Prevalence and factors associated with tuberculosis infection among new school entrants, New York City, 1991—1993. Int J Tuberc Lung Dis 1999;3:31—41. [21] Gounder CR, Driver CR, Scholten JN. Tuberculin testing and risk of tuberculosis infection among New York City schoolchildren. Pediatrics 2003;111:e309—15. [22] Reznik M, Ozuah PO. Tuberculin skin testing in children. Emerg Infect Dis 2006;12:725—8. [23] Mohle-Boetani JC, Miller B, Halpern M. School-based screening for tuberculosis infection. A cost—benefit analysis. J Am Med Assoc 1995;274:613—9. [24] Henry PM, Mills WA, Holtan NR. Screening for tuberculosis infection among secondary school students in MinneapolisSt Paul: policy implications. Minn Med 1996;79:43—9. [25] Pong AL, Bronwen JA, Moser KS. Tuberculosis screening at 2 San Diego high schools with high-risk populations. Arch Pediatr Adolesc Med 1998;152:646—50. Immigrants with latent tuberculosis infection [26] Saiman L, San Gabriel P, Schulte J. Risk factors for latent tuberculosis infection among children in New York City. Pediatrics 2001;107:999—1003. [27] Young J, O’Connor ME. Risk factors associated with latent tuberculosis infection in Mexican American children. Pediatrics 2005;115:e647—53. [28] Chang S, Wheeler LSM, Farrell KP. Public health impact of targeted tuberculosis screening in public schools. Am J Public Health 2002;92:1942—5. [29] Menzies D, Chan CH, Vissandjée B. Impact of immigration on tuberculosis infection among Canadian-born schoolchildren and young adults in Montreal. Am J Respir Crit Care Med 1997;156:1915—21. [30] Froehlich H, Ackerson LM, Morozumi PA. Targeted testing of children for tuberculosis: validation of a risk assessment questionnaire. Pediatrics 2001;107:e54. [31] Ozuah PO, Ozuah TP, Stein REK. Evaluation of a risk assessment questionnaire used to target tuberculin skin testing in children. J Am Med Assoc 2001;285:451—3. [32] Fos PJ, Lee JH, Zuniga MA. The role of quality improvement in disease management: a statewide tuberculosis control success story. J Public Health Manage Pract 2005;11:222—7. [33] Levesque JF, Dongier P, Brassard P. Acceptance of screening and completion of treatment for latent tuberculosis infection among refugees claimants in Canada. Int J Tuberc Lung Dis 2004;8:711—7. [34] Lavigne M, Rocher I, Steensma C. The impact of smoking on adherence for latent tuberculosis infection. BMC Public Health 2006;6:66. [35] Menzies D, Dion MJ, Francis D. In closely monitored patients, adherence in the first month predicts completion of therapy for latent tuberculosis infection. Int J Tuberc Lung Dis 2005;9:1343—8. [36] Ailinger RL, Moore JB, Nguyen N. Adherence to latent tuberculosis infection therapy among Latino immigrants. Public Health Nurs 2006;23:307—13. [37] Goldberg SV, Wallace J, Jackson JC. Cultural case management of latent tuberculosis infection. Int J Tuberc Lung Dis 2004;8:76—82. [38] Shieh FK, Snyder G, Horsburgh CR. Predicting noncompletion of treatment for latent tuberculous infection. A prospective survey. Am J Respir Crit Care Med 2006;174:717—21. [39] LoBue PA, Moser KD. Use of isoniazid for latent tuberculosis infection in a Public Health Clinic. Am J Respir Crit Care Med 2003;168:443—7. [40] Nyamathi A, Stein JA, Schumann A. Latent variable assessment of outcomes in a nurse-managed intervention to increase latent tuberculosis treatment completion in homeless adults. Health Psychol 2007;26:68—76. [41] Carvalho ACC, Saleri N, El-Hamad I. Completion of screening for latent tuberculosis infection among immigrants. Epidemiol Infect 2005;133:179—85. 75 [42] Shukla SJ, Warren DK, Woeltje KF. Factors associated with the treatment of latent tuberculosis infection among health-care workers at Midwestern Teaching Hospital. Chest 2002;122:1609—14. [43] Parsyan AE, Saukkonen J, Barry MA. Predictors of failure to complete treatment for latent tuberculosis infection. J Infect 2007;54:262—6. [44] Morisky DE, Ebin VJ, Malotte CK. Assessment of tuberculosis treatment completion in an ethnically diverse population using two data sources. Implications for treatment interventions. Eval Health Prof 2003;26:43—58. [45] Hovell M, Blumberg E, Gil-Trejo L. Predictors of adherence to treatment for latent tuberculosis infection in high-risk Latino adolescents: a behavioral epidemiological analysis. Soc Sci Med 2003;56:1789—96. [46] Berg J, Blumberg EJ, Sipan CL. Somatic complaints and isoniazid (INH) side effects in Latino adolescents with latent tuberculosis infection (LTBI). Patient Educ Couns 2004;52:31—9. [47] Cass AD, Talavera GA, Gresham LS. Structured behavioral intervention to increase children’s adherence to treatment for latent tuberculosis infection. Int J Tuberc Lung Dis 2005;9:415—20. [48] Kominski GF, Varon SF, Morisky DE. Costs and costeffectiveness of adolescent compliance with treatment for latent tuberculosis infection: results from a randomized trial. J Adolesc Health 2007;40:61—8. [49] Nyamathi AM, Christiani A, Nahid P. A randomized controlled trial of two treatment programs for homeless adults with latent tuberculosis infection. Int J Tuberc Lung Dis 2006;10:775—82. [50] Spyridis NP, Spyridis PG, Gelesme A. The effectiveness of a 9-month regimen of isoniazid alone versus 3- and 4-month regimens of isoniazid plus rifampicin for treatment of latent tuberculosis infection in children: results of an 11-year randomized study. Clin Infect Dis 2007;45: 715—22. [51] Ena J, Valls V. Short-course therapy with rifampicin plus isoniazid, compared with standard therapy with isoniazid, for latent tuberculosis infection: a meta-analysis. Clin Infect Dis 2005;40:670—6. [52] Taylor REB, Cant AJ, Clark JE. Potential effect of NICE tuberculosis guidelines on paediatric tuberculosis screening. Arch Dis Child 2008;93:200—3. [53] Drobniewski F, Cobelens F, Zellweger JP, et al. Use of gamma-interferon assays in low- and mediumprevalence countries in Europe: a consensus statement of Wolfheze Workshop organised by KNCV/EuroTB, Vilnius Sept 2006. Euro Surveill 2007;12(30):pii=3242. http://www.eurosurveillance.org/ew/2007/070726.asp#2 [updated July 2008]. Available online at www.sciencedirect.com