Background: To eliminate tuberculosis (TB) in the United States, more information is needed on how to gain access to difficult-to-reach, high-risk populations to evaluate people who would benefit from treatment for latent TB infection (LTBI).
Methods: A field study was conducted of people at risk for co-infection with TB and the human immunodeficiency virus (HIV) and to demonstrate that treating LTBI in inmates is feasible. Inmates were tested for LTBI using the Mantoux tuberculin skin test (TST). Outcomes measured were skin test results and the start and completion of treatment for LTBI.
Results: In 49 correctional facilities in 12 states, 198102 inmates had a skin test read. The mean skin test positivity rate among inmates was 17.0%. Of those who had a known HIV test result, 14.5% tested HIV positive. Inmates with a positive TST were 4.2 times more likely than those with a negative TST to be HIV infected (95% confidence interval [CI]=3.9-4.4). Therapy for LTBI was completed in 55.9% of patients started on treatment. Patients who were HIV positive and started on a 12-month treatment regimen were less likely than HIV-negative patients (40.0% vs 68.1%, respectively) to complete treatment (odds ratio [OR]=0.24, 95% CI=0.20-0.28). Patients treated in jails were less likely than those treated in prisons (33.6% vs 57.7%, respectively) to complete treatment (OR=0.29, 95% CI=0.26-0.32).
Conclusions: Correctional facilities offer a venue for identifying and treating high-risk individuals for LTBI. However, completing treatment is more problematic in jails than in prisons.