Latent tuberculosis infection in a Malaysian prison: implications for a comprehensive integrated control program in prisons

Background: Prisons continue to fuel tuberculosis (TB) epidemics particularly in settings where access to TB screening and prevention services is limited. Malaysia is a middle-income country with a relatively high incarceration rate of 138 per 100,000 population. Despite national TB incidence rate remaining unchanged over the past ten years, data about TB in prisons and its contribution to the overall national rates does not exist. This survey was conducted to address the prevalence of latent TB infection (LTBI) in Malaysia's largest prison. Methods: From July to December 2010, all HIV-infected and a comparative group of HIV-uninfected prisoners housed separately in Kajang prison were asked to participate in the survey after explaining the study protocol. Subjects providing informed consent were interviewed using a structured questionnaire followed by the placement of tuberculin skin test (TST) with 2 TU of PPD RT-23 to subjects not being treated for active TB. TST was read after 48-72 hours and indurations of >= 5 mm and >= 10 mm were considered positive among HIV-infected and HIV-uninfected subjects, respectively. Additionally, HIV-infected inmates underwent phlebotomy for CD4 lymphocyte count assessment. A logistic regression model was explored to determine factors associated with TST positivity. Results: Overall, 286 subjects (138 HIV-infected and 148 HIV-uninfected) had complete data and TST results. The majority were men (95.1%), less than 40 years old (median age 36.0, SD 7.87), and Malaysians (93.3%). Most (82.5%) had been previously incarcerated and more than half (53.1%) reported sharing needles just prior to their incarceration. TST was positive in 88.8% (84.7% among HIV-infected and 92.5% among HIV-uninfected subjects) and was independently associated with being HIV-uninfected (AOR = 2.97, p = 0.01) and with frequent previous incarcerations (AOR = 1.22 for every one previous incarceration, p = 0.01) after adjusting for other potential confounding factors. Conclusions: The prevalence of LTBI was extraordinary high in this sample of Malaysian prisoners, regardless of their age or HIV status. This warrants further examination of the size of the problem of TB in other congregate settings and the establishment of an evidence-based TB control program in Malaysian prisons with integrated TB, HIV and substance abuse components.