Gough et al. raise concerns regarding our estimates of hepatitis C virus (HCV) incidence among continuously incarcerated persons, arguing that the inclusion of studies that do not account for the HCV seroconversion window period may result in overestimation of incidence rates. We accept that this is a valid point, but note that even when the studies in question are excluded from the summary estimate, HCV incidence among detainees with a history of injection drug use (IDU) remains many times higher than that for general detainee samples (6.6 per 100 person-years versus 0.4 per 100 person-years), albeit with overlapping confidence intervals.
Gough et al. note that in an earlier review of this topic, they were unable to estimate HCV incidence for detainees with a history of IDU because of a lack of data for this population. However, their review focused solely on U.S. jail and prison populations; our review was global and identified prison-based HCV incidence studies in Australia and the UK as well as the United States. Across these studies, we note the consistency with which HCV incidence in detainees who inject drugs exceeds that in general detainee samples. Furthermore, since the completion of our review, a Scottish study of continuously incarcerated persons who had been incarcerated for at least 75 days at baseline (accounting for the HCV antibody seroconversion window) has been published, estimating HCV incidence to be 0.6%-0.9% overall and 3.0%-4.3% in prisoners with a history of IDU. In those prisoners who reported injecting drugs in prison, HCV incidence was 14.2%-19.6%.
IDU is not the only potential route of HCV transmission in correctional settings. In-prison tattooing has been identified as a likely source of infection, although it may also be the case that detainees with a history of IDU are more likely than other detainees to obtain tattoos while incarcerated, complicating delineation of risk factors. Blood exposures from sharing of shaving equipment and physical assault have also been implicated in in-prison HCV seroconversion. Nevertheless, there is consistent international evidence that IDU occurs in prisons, and given the extraordinarily high baseline prevalence of HCV among detainees who inject drugs and the almost universal lack of sterile injecting paraphernalia in correctional settings, IDU while incarcerated is a biologically plausible source of HCV infection.
In light of the findings from the reanalysis, the consistency with which HCV prevalence is elevated among detainees with a history of IDU, compared with general detainees, recently reported findings, and the biological plausibility of the association, we remain confident in our conclusion that prisoners and other detainees who inject drugs are at particular risk of HCV infection. We endorse Gough et al.'s call for more evidence regarding HCV risk while incarcerated, but note that the uncertainty in the current evidence should not preclude the provision of stronger evidence-based prevention responses in correctional settings, including opioid substitution treatment and needle and syringe programs.
Sarah Larney, Ph.D.1,2
Nickolas D. Zaller, Ph.D.2,3
Louisa Degenhardt, Ph.D.1,4
1National Drug and Alcohol Research Center
University of New South Wales
2Alpert Medical School
3Division of Infectious Diseases
The Miriam Hospital
4Center for Health Policy, Programs and Economics
School of Population Health
University of Melbourne