Therapy for latent tuberculosis infection (LTBI) is recommended for all children with positive tuberculin skin tests (TSTs) or interferon gamma release assays (IGRA) who have not been treated previously, unless specific contraindications exist.1 Long courses of therapy for asymptomatic patients have low completion rates.2 Although shorter-course therapy using 2 antituberculosis drugs to promote adherence has been proposed, few longitudinal efficacy data exist for children. In lieu of abbreviated therapy, adherence to LTBI therapy also can be enhanced by the use of health departments to deliver directly observed preventive therapy (DOPT) to children. We evaluated factors associated with successful completion of LTBI therapy in a pediatric cohort.
MATERIALS AND METHODS
Children aged 0 to 18 years seen at the Children's Tuberculosis Clinic at Ben Taub General Hospital with LTBI from January 1, 2007 to December 31, 2010 were included. LTBI was defined as a positive TST3 or positive IGRA in an asymptomatic child with a normal examination and chest radiograph. Children could receive medications in 1 of the following 3 ways: (1) daily medication obtained and administered by families (“self-administration”); (2) daily administration by families with monthly medication delivery through health departments and weekly phone reminders (“enhanced self-administered therapy”); or (3) medication given twice weekly in the presence of a public health worker (DOPT). Clinician decision between the 3 options was nonrandom and protocol based. Children identified during public health department contact investigations received DOPT or enhanced self-administered therapy, as did most infants and young children and recent immigrants, even if no source case was identified. Older (>5 years of age) children without identifiable source cases often received self-administered therapy. Information sheets on LTBI and medications were available in English and Spanish; in-room phone translation services were available for other languages.
Completion was defined as 6 months of rifampin (used for isoniazid-resistant infections or isoniazid intolerance) or 9 months of isoniazid (first-line therapy in the absence of an isoniazid-resistant source case) or pyrazinamide/levofloxacin (for children whose source case had multidrug-resistant TB). Adherence was defined by guardian/child report for children receiving self-administered medications and by health department report of 52 and 78 doses received for 6- and 9-month courses, respectively. Pill counts and drug levels were not used. Information collected retrospectively included demographics (age, gender, country of origin, race/ethnicity), number of medications used, and method of administration. Statistical analyses were performed using STATA 10 (Stata, Inc., College Station, TX). Institutional review board approval was obtained.
A total of 289 children were evaluated for LTBI; all were diagnosed by TST. Of these, 35 (12%) returned to their primary care provider for follow-up, 6 families (2%) refused to initiate therapy, and 248 (86%) were followed in our clinic, where longitudinal data were available. Of the latter group, 186 of 248 (75%) completed therapy and 62 (25%) did not have documentation of therapy completion (38 stopped therapy prematurely, 13 did not return for clinic follow-up, and 11 moved out of the area during the treatment course). Features of the population followed at our clinic are shown (Table 1). The demographic characteristics of children who were followed in the TB clinic were not significantly different from children who followed-up with their primary care providers in terms of age, race, or region of origin (data not shown).
Children completing therapy finished a mean of 8.9 months (95% confidence interval [CI], 8.8–9.1 months) of therapy, whereas for the 32 of 62 children who did not complete therapy and for whom treatment duration was known, default occurred after a mean of 3.9 months (95% CI, 2.9–4.9 months). Univariate analysis indicated that foreign birth (odds ratio [OR], 1.9; 95% CI, 1.1–3.8), self-administered (vs. DOPT) therapy (OR, 11.6; 95% CI, 5.7–23.6), and child's TBI diagnosis not obtained through contact investigation (OR, 7.7; 95% CI, 2.9–20.1) were associated with decreased adherence. On multivariate analysis, only the use of DOPT was associated with therapy completion (OR, 7.2; 95% CI, 3.8–13.8). None of the children treated for LTBI were known to have developed TB disease to date.
The primary finding of this study was that the only factor associated with increased pediatric adherence with LTBI therapy in multivariate analysis was medication administration through health departments. We noted that default in therapy tended to occur early in the proposed course. One possibility is that decreased adherence late in a treatment course is more amenable to structured interventions, as described below. In contrast, early adherence issues may be because of families not understanding the rationale behind therapy. We were concerned that the latter may have been a problem in recent immigrants from non-Spanish–speaking countries, as clinicians only are conversant in English and Spanish. However, adherence was not associated with country of origin, potentially because of the availability of in-room phone translation services.
A number of other barriers to LTBI completion should be considered. First, unlike in patients with TB disease, LTBI children are asymptomatic, so families do not see a response to therapy. Second, most TB clinics care for a diverse international patient base, where parents are cautioned as to the poor utility of TSTs in BCG recipients. Third, preventive medicine for TB has been underemphasized in many countries because of resource constraints and emphasis on treatment of disease. Finally, therapy duration is inversely correlated with adherence. In adults, these factors are reflected in adherence rates for LTBI of <50%.2
These factors may be addressed in turn. For the asymptomatic child identified through a contact investigation, the family already has witnessed the impact of TB disease, making preventive therapy more palatable. Incentives and enablers also can be given to the child. One study gave children calendars and stickers. For each day the child took medication, a sticker was placed on that date; a completed calendar was redeemable for a toy at the subsequent clinic visit. Children receiving this intervention were more than twice as likely to complete therapy as controls.4 For international families with preconceptions about the TST, IGRAs offer enhanced specificity, helping demystify LTBI, and increasing adherence.5
The final 2 factors are perhaps the hardest to address in the absence of health department infrastructure. TB control has focused on TB disease and has led to reductions in TB mortality, with an estimated 4.6 to 6.3 million lives saved globally from 1995 to 2009.6 Similar inroads have not been made into LTBI control. In industrialized nations, where the majority of TB disease cases occur in foreign-born persons, we may have reached the limits of TB control without addressing the LTBI population. However, budgetary constraints may limit the ability of a health department to provide DOPT or enhanced self-administered therapy services. Historical precedence indicates that cuts in TB control programs are associated with increase in total and drug-resistant TB disease cases, as illustrated by the New York City experience in the 1990s.7
Our findings differ from prior studies. One study, conducted among adults with TB disease in Malawi, demonstrated that adherence was higher for patients receiving guardian-based versus health cente–rbased directly observed therapy (94% vs. 84%), emphasizing the role of social support in TB care.8 A second study evaluated adherence to isoniazid prophylaxis among HIV-infected children receiving daily or thrice-weekly self-administered therapy. Adherence was 95% and was independent of dosing regimen; decreased adherence was seen in younger children and in children from overcrowded homes.9 It is unclear why most studies in the United States have shown lower adherence.2 One possibility is that first-hand experience with the consequences of therapy default provides motivation for adherence in high-burden TB regions.
This study had several limitations. First, the retrospective nature precluded collection of all variables (eg, frequency and type of translation services used). Second, adherence for children on self-administered therapy was based on parental/guardian report of adherence, subject to recall bias. However, self-report for antiretroviral therapy is commonly used in HIV clinics, and it has shown correlation with virologic response to therapy.10 Third, reasons for default were unknown. Fourth, IGRAs were not used, given the few pediatric specific data available for much of the study period. Knowledge of IGRA results may have increased adherence for children on self-administered therapy. Fifth, there could have been bias in terms of which children received DOPT/enhanced self-administered therapy versus self-administered medications. However, bias here probably would have been in the direction of families with adherence concerns receiving DOPT/enhanced self-administered therapy. Sixth, children whose follow-up was unknown were included in the default group. Although this could have lead to classification bias, we deemed it the more conservative approach; as many of our patients have no medical home, we felt it unreasonable to assume they were receiving LTBI therapy in another medical venue. Finally, choice of DOPT versus self-administration was not randomized or controlled; instead, it was based on perceived risk of disease development and available resources. Although a study limitation, it reflects the reality of risk-stratifying patients for limited public health resources.
In conclusion, health department–assisted therapy for LTBI should be considered for children at high risk for progression to TB disease. Adequate TB control in the United States cannot function independently of health departments.
The authors thank our local, regional, and state health departments for enabling us to optimize care for our children and Larry D. Teeter, PhD and Edward Graviss, PhD, MPH.
1. American Academy of Pediatrics. Tuberculosis. In: Pickering LK, Baker CJ, Kimberlin DW, et al., eds. Red Book: 2009 Report of the Committee on Infectious Diseases. 28th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2009:691.
2. Horsburgh CR Jr, Goldberg S, Bethel J, et al.. Latent TB infection treatment acceptance and completion in the United States and Canada. Chest. 2010;137:401–409.
3. American Thoracic Society. Targeted tuberculin testing and treatment of latent tuberculosis infection. Am J Respir Crit Care Med. 2000;161:S221–S247.
4. Cass AD, Talavera GA, Gresham LS, et al.. Structured behavioral intervention to increase children's adherence to treatment for latent tuberculosis infection. Int J Tuberc Lung Dis. 2005;9:415–420.
5. Mazurek GH, Jereb J, Vernon A, et al.. Updated guidelines for using interferon gamma release assays to detect Mycobacterium tuberculosis
infection—United States, 2010. MMWR Recomm Rep. 2010;59:1–25.
6. Glaziou P, Floyd K, Korenromp EL, et al.. Lives saved by tuberculosis control and prospects for achieving the 2015 global target for reducing tuberculosis mortality. Bull World Health Organ. 2011;89:573–582.
7. Munsiff SS, Joseph S, Ebrahimzadeh A, et al.. Rifampin-monoresistant tuberculosis in New York City, 1993–1994. Clin Infect Dis. 1997;25:1465–1467.
8. Manders AJ, Banerjee A, van de Borne HW, et al.. Can guardians supervise TB therapy as well as health workers? A study on adherence during the intensive phase. Int J Tuberc Lung Dis. 2001;5:838–842.
9. Le Roux SM, Cotton MF, Golub JE, et al.. Adherence to isoniazid prophylaxis among HIV-infected children: a randomized controlled trial comparing 2 dosing schedules. BMC Med. 2009;7:67.
10. Berg KM, Arnsten JH. Practical and conceptual challenges in measuring antiretroviral adherence. J Acquir Immune Defic Syndr. 2006;43(suppl 1):S79–S87.