Tuberculosis incidence in prisons: a systematic review

Abstract

Background: Transmission of tuberculosis (TB) in prisons has been reported worldwide to be much higher than that reported for the corresponding general population.

Methods and findings: A systematic review has been performed to assess the risk of incident latent tuberculosis infection (LTBI) and TB disease in prisons, as compared to the incidence in the corresponding local general population, and to estimate the fraction of TB in the general population attributable (PAF%) to transmission within prisons. Primary peer-reviewed studies have been searched to assess the incidence of LTBI and/or TB within prisons published until June 2010; both inmates and prison staff were considered. Studies, which were independently screened by two reviewers, were eligible for inclusion if they reported the incidence of LTBI and TB disease in prisons. Available data were collected from 23 studies out of 582 potentially relevant unique citations. Five studies from the US and one from Brazil were available to assess the incidence of LTBI in prisons, while 19 studies were available to assess the incidence of TB. The median estimated annual incidence rate ratio (IRR) for LTBI and TB were 26.4 (interquartile range [IQR]: 13.0-61.8) and 23.0 (IQR: 11.7-36.1), respectively. The median estimated fraction (PAF%) of tuberculosis in the general population attributable to the exposure in prisons for TB was 8.5% (IQR: 1.9%-17.9%) and 6.3% (IQR: 2.7%-17.2%) in high- and middle/low-income countries, respectively.

Conclusions: The very high IRR and the substantial population attributable fraction show that much better TB control in prisons could potentially protect prisoners and staff from within-prison spread of TB and would significantly reduce the national burden of TB. Future studies should measure the impact of the conditions in prisons on TB transmission and assess the population attributable risk of prison-to-community spread. Please see later in the article for the Editors' Summary.

Figure 1. Flow diagram for study selection.

Figure 2. Forest plot showing the study-specific estimates of the IRRs for LTBI in prisons as compared to corresponding general populations, by income area according to the World Bank classification.

Source: .

Figure 3. Forest plot showing the study-specific estimates of the IRR for tuberculosis in prisons as compared to the corresponding general populations, by income area according to the World Bank classification.

Source: . NA, not applicable.

Figure 4. Contour plot showing the relationship between the proportion of exposed population, IRR, and PAF%.

X-axis reports the proportion of the population in prison, Y-axis reports the PAF on a log scale calculated using the Levin's formula . The isoclines represent different levels of IRR. Klopf et al. reported TB incidence *before and **after implementation of a TB control program in New York State Department of Correctional Services and prisons personnel separately. Russkikh et al. reported TB incidence among prison personnel ^†during and ^‡following the socioeconomic crisis occurred in Russia in the late 1990s .