Incidence of TB in inmates with latent TB infection: 5-year follow-up.

BACKGROUND: Inmates are a high-risk population for tuberculosis (TB) control efforts, including treatment for latent tuberculosis infection (LTBI). Completion of therapy after release has been poor. The goal of this study was to evaluate therapy completion and active disease over 5 years in a cohort of inmates. METHODS: The sample was from a completed randomized trial in 1998-1999 of education or incentive versus usual care to improve therapy completion after release from the San Francisco County Jail. Records from the jail, the County Tuberculosis Clinic, and the California TB Registry were used to measure therapy completion and development of active TB. Analyses were conducted in 2005. RESULTS: Of a total 527 inmates, 31.6% (n=176) completed therapy, of whom 59.7% (n=105) completed it in jail. Compared with the U.S.-born, foreign-born inmates residing in the United States for < or =5 years were less likely to complete the therapy (adjusted odds ratio [AOR]=0.49, 95% confidence interval [CI]=0.28-0.85), and those with more education were more likely to complete the therapy (AOR=1.06, 95% CI=1.01-1.12). Three subjects developed active TB in the 5 years of follow-up, resulting in an annual rate of 108 per 100,000. Compared with California rates, subjects were 59 times as likely to develop active TB (standardized morbidity ratio of 59.2, 95% CI=11.2-145.1). None had completed therapy, none were new immigrants, and two were known to be HIV-positive at diagnosis. CONCLUSIONS: Completion of therapy for LTBI is a challenge, but the active TB seen in this jail cohort emphasizes the importance of continued efforts to address TB risk in this population.