Abstract

AIMS: To determine the effects of 1) a condom distribution program and 2) a condom distribution program combined with opt-out sexually transmitted infection (STI) screening on the transmission and prevalence of STIs in a prison system.

METHODS: Using data from an implementation evaluation of a state-wide prison condom program and parameter estimates from available literature, a deterministic model was developed to quantify the incidence and prevalence of sexually transmitted HIV, hepatitis B, chlamydia, syphilis and gonorrhoea across 14 Victorian prisons. The model included individual prison populations (by longer (>2 years) or shorter sentence lengths) and monthly prisoner transfers. For each STI, simulations were compared: without any intervention; with a condom distribution program; and with a combined condom and opt-out STI screening at prison reception intervention program.

RESULTS: Condoms reduced the annual incidence of syphilis by 99% (N = 66 averted cases); gonorrhoea by 98% (N = 113 cases); hepatitis B by 71% (N = 5 cases); chlamydia by 27% (N = 196 cases); and HIV by 50% (N = 2 cases every 10 years). Condom availability changed the in-prison epidemiology of gonorrhoea and syphilis from self-sustaining to levels unlikely to result in infection outbreaks; however, condoms did not reduce chlamydia prevalence below a self-sustaining level due to its high infectiousness, high prevalence and low detection rate. When combined with a screening intervention program, condoms reduced chlamydia prevalence further, but not below a self-sustaining level. The low prevalence of HIV and hepatitis B in Australian prisons meant the effects of condoms were predicted to be small.

CONCLUSION: Condoms are predicted to effectively reduce the incidence of STIs in prison and are predicted to control syphilis and gonorrhoea transmission, however even combined with a screening on arrival program may be insufficient to reduce chlamydia prevalence below self-sustaining levels. To control chlamydia transmission additional screening of the existing prison population would be required.

Link to publisher’s web site

The authors gratefully acknowledge the contribution to this work of the Victorian Operational Infrastructure Support Program. The authors wish to thank Justice Health, Victorian Department of Justice, for funding this modelling component as part of a broader evaluation of their condom implementation program. The authors also wish to thank Justice Health and Corrections Victoria for the provision of some of the operations data utilised in the modelling.