Modelling alternative strategies for delivering hepatitis B vaccine in prisons: the impact on the vaccination coverage of the injecting drug user population

Abstract

Since 2001 hepatitis B vaccination has been offered to prisoners on reception into prisons in England and Wales. However, short campaigns of vaccinating the entire population of individual prisons have achieved high vaccination coverage for limited periods, suggesting that short campaigns may be a preferable way of vaccinating prisoners. A model is used that describes the flow of prisoners through prisons stratified by injecting status to compare a range of vaccination scenarios that describe vaccination on prison reception or via regular short campaigns. Model results suggest that vaccinating on prison reception can capture a greater proportion of the injecting drug user (IDU) population than the comparable campaign scenarios (63% vs. 55.6% respectively). Vaccination on prison reception is also more efficient at capturing IDUs for vaccination than vaccination via a campaign, although vaccination via campaigns may have a role with some infections for overall control.

Fig. 1

Flow diagram of the turnover of prisoners in England and Wales [including injecting drug users (IDUs)]. The mutually exclusive compartments represent the different imprisonment and IDU states. Arrows represent the flow between the states. (X are individuals that have never been imprisoned, Y is the prison population, and Z are individuals previously imprisoned. N are those individuals that have never injected while I and P denote current and previous IDUs respectively). This structure is further stratified by age 15–74 years. The rates that individuals move between compartments may be age-dependent.

Fig. 2

Vaccination status of the injecting drug user (IDU) population over time under the continuous and pulse vaccination scenarios in which pulses are administered 3 months apart.

Fig. 3

Vaccination status of the injecting drug user (IDU) population in 2012 with variations in vaccination scenario. (For the 2-year pulses due to the oscillations of the vaccination coverage an average over 2011 and 2012 is taken.)

Fig. 4

(a) Number of doses required to implement each vaccination scenario from 2006 to 2012. (b) Percentage of doses (first, second, third doses) given to under-vaccinated injecting drug users (IDUs) in 2012 with variation in the vaccination scenario. (c) The age-specific percentage of the IDU population receiving ⩾2 doses of vaccine in 2012 when applying the 100% continuous and 100% 3-month pulse scenario. (d) Percentage of doses that are administered unnecessarily with variation in the vaccination scenario. Key (panels a, b, d): ■, Continuous; □, 3-month pulse; , 6-month pulse; , 1-year pulse; , 2-year pulse.