transmission dynamics within prisons: a population-based genomic study.

BACKGROUND: One barrier to intervening in the global tuberculosis pandemic is that it is unknown whethertransmission largely occurs through repeated close exposures among few contacts or many shorter-term contacts. Identifying sources of transmission is particularly urgent in congregate settings like prisons with high incidence of infection. Our aim was to identify the type of contacts associated withtransmission risk within prisons.

METHODS: We conducted genomic surveillance in a prison system in Central West Brazil. We whole genome sequencedisolates and collected detailed incarceration histories. We modeled transmission linkages as a function of different types of prison exposures, using genomic clustering as a proxy for transmission and controlling for multiple pairwise comparisons.

FINDINGS: We collected detailed incarceration histories for 595 individuals, mean age 31 (ST-standard deviation 8.5) and 99% men, from whom we sequenced 550 high qualitygenomes. In a binomial model, a month-long increase in exposure to an individual with tuberculosis within a prison cell increased the odds of pairwise genomic clustering by 14% (odds ratio-OR: 1.14, 95% CI: 1.09-1.19) and a six-month increase in exposure doubled the odds of genomic clustering (OR: 2.24, 95% CI: 1.73-2.91). Most (83%; 494 of 595) individuals with tuberculosis had at least one potential day-long exposure in a prison cell to another individual with tuberculosis, and frequently many, with a median of 8 (interquartile range-IQR: 4-12) potential unique exposures to individuals in prison cells. Frequent movements by the prison system create a highly connected contact network: individuals with tuberculosis were transferred a median of 8 (IQR: 4-13) times in the 2 years before diagnosis.

INTERPRETATION: While documented cell-level exposures can explain a significant proportion oftransmission, most transmission links occur outside prison cells, either from other contacts in the same prison or from unreported or unsampled exposures. Our results support the urgent expansion of prison-wide mass screenings, tuberculosis preventive therapy, and structural interventions to reduce transmission risk in prisons and other congregate settings.

FUNDING: National Institutes of Health (NIAID: 5K01AI173385, R01AI100358, and R01AI149620).