How can we maximise the benefits of smoke-free prisons? Decision analytic model to predict potential impacts on public health

Tobacco smoking prevalence remains high in disadvantaged populations such as people in prison. Smokefree prisons protect health, however around 90% of people who smoke pre-prison, relapse to smoking shortly after release. If people released from smokefree prisons maintain smoking abstinence this could benefit their health and finances. Knock-on effects of smoking relapse on families could also be avoided. Offering an intervention to reduce relapse to smoking on release has the potential to benefit released people and their families. This study assesses potential costs and outcomes for released people and their families, of introducing a smokefree prison policy and an intervention to reduce post-release smoking relapse. Based on the smoking/vaping status of released people we modelled the impact, on costs and outcomes, of four scenarios. We modelled scenarios which varied across two dimensions: (1) whether people were/were not permitted to vape in smokefree prisons, and (2) whether a smoking cessation intervention was offered/was not offered in smokefree prisons. The scenarios reflect different combinations of these factors. We estimated costs and outcomes (benefits) for released people, their partners and children over a lifetime. We included personal costs (vaping and smoking), healthcare and intervention costs, and outcomes included quality of life. For released people, results indicated that not permitting vaping in prison was less costly and more beneficial than when vaping was permitted. Offering a smoking cessation intervention to released people was less costly than not offering a smoking cessation intervention, irrespective of whether vaping was permitted or not. However, whilst offering a smoking cessation intervention was beneficial when vaping was permitted in prison, results are uncertain for the benefits of offering a smoking cessation intervention when vaping is not permitted in prison. Sensitivity analyses indicate uncertainty and show that changing the values for vaping prevalence and smoking relapse rates would change these results. For both partner and child (ren), costs were higher and quality of life lower for those living with released people who relapse to smoking compared to those who vape or neither smoke nor vape. Targeted support for smoking cessation interventions to improve health outcomes for people released from smokefree prison and their families can ultimately contribute to broader public health improvements and improve health in a priority group. There is a need for greater evidence in this area to inform future modelling, particularly on relapse to smoking on release and the long-term effects of vaping. Results indicate uncertainty about the overall value of permitting vaping in smokefree prisons; wider factors associated with not allowing vaping in prisons would need to be assessed in future work. Study findings enhance understanding of the potential cost-effectiveness of smokefree prison policy, highlight uncertainty in some model inputs, and can inform decisions about how value could be maximised.