Tuberculosis in Incarcerated Youth in TexasRESEARCH LETTER

To the Editor: Incarcerated youth are said to be at high risk for tuberculosis (TB) infection,¹ but this population has not been well studied. Guidelines recommend that only youth meeting specific new screening criteria be tested for TB.² In a group of incarcerated adolescents, we determined the number of cases of active TB disease and latent TB infection (LTBI), risk factors for infection, and documented compliance with therapy after release.

We retrospectively studied youth 10 to 17 years of age at the Harris County Juvenile Detention Center in Texas and at 4 county facilities used for long-term incarceration (1-6 mo; mean, 3 mo) between January 1, 1999, and December 31, 2004. Youth who are adjudicated to long-term facilities are required by the State of Texas to receive a tuberculin skin test (TST).

The TST was considered positive if induration was 10 mm or more. Youth who are TST-positive undergo chest radiographs and those with negative radiographs begin a 9-month course of isoniazid therapy for LTBI. At release from incarceration, youth receiving isoniazid are referred to the City of Houston Health Department’s Directly Observed Therapy Program. Patients living outside of Houston are referred to the University of Texas-Houston Children’s TB Clinic at Lyndon B. Johnson General Hospital for follow-up.

We determined the number of youth who had active TB disease and LTBI. We recorded age, sex, duration of incarceration prior to the TST, self-reported race/ethnicity, place of birth, and history of exposure to active TB. We determined the percentage of patients who had documented completion of treatment for LTBI. The study had institutional review board approval.

During the study, 12 651 youth were sentenced to long-term incarceration in facilities where our physicians continued to provide medical care. Fifteen of the youth were positive for human immunodeficiency virus (HIV). No youth had or developed clinical evidence of active TB disease.

Two hundred (1.58%) of the 12 651 patients were TST-positive (Table). Of the currently recommended screening risk factors,² the only one for which we had reliable information was birth in a high-risk area (67/200, or 33.5%). Two additional adolescents (1.0%) reported exposure to active TB. No one had an abnormal chest radiograph finding. All began isoniazid therapy for LTBI. One patient developed severe hepatitis that resolved after stopping isoniazid therapy. The cost of the 200 radiographs was $37 000 ($185 each).

Although overall success of the Houston Directly Observed Therapy Program exceeds 90%, the health department reported that because of inability to locate the released youth very few (< 10%) completed the program. Less than 10% of those referred to Lyndon B. Johnson General Hospital kept follow-up appointments.

Our study population had a low prevalence of active TB or LTBI. For comparison, in a national sample of young US Navy recruits in 1997-1998, 3.5% had LTBI, compared with 1.6% of our group.³ The recruits had a greater proportion reported as white (55.3%) than our population (22%).

The apparent poor adherence to therapy raises questions about the value of identifying and treating LTBI in this population. Treatment for less than 6 to 9 months is ineffective.⁴ Moreover, 1 patient developed a severe hepatic reaction to isoniazid, and the potential benefit in this group must be balanced against the risk of physical harms, costs of serious adverse events, and costs of follow-up. We would have avoided radiographs and isoniazid therapy for LTBI in 65% of the 200 TST-positive youth if we had tested only adolescents who met the newly recommended screening criteria related to birth or prolonged residence in a high-risk region outside the United States, exposure to TB or LTBI, or close contact with someone who is TST-positive.² However, when considering the benefits of screening in a closed population, even 1 case of active contagious pulmonary TB can have disastrous consequences.

This study has several limitations. These results are from 1 county in Texas and TB prevalence may be different in other areas of the country, with different implications. Similar studies should be conducted in other regions. More of the youth may have been exposed to individuals with TB disease or LTBI but parents were not readily available to provide this history. The number of adolescents identified with risk factors would have been higher if we had asked about additional exposures² considered to be possibly useful in recent expert recommendations: exposure to persons who had been in jail, prison, or a shelter; used illegal drugs; had HIV; or were exposed to household members born or having traveled outside the United States. However, the importance of these risk factors in settings such as ours is unclear.

Our results suggest that mandatory TB skin testing in incarcerated youth in our area who do not meet the new risk screening criteria should be discontinued, and possibly replaced by risk questionnaires and screening for symptoms of active TB. Further research should determine unique risk factors for TB infection in this population, cost-effectiveness of screening programs, and interventions to improve adherence to therapy for LTBI.