Preliminary evidence that self-incentives uniquely promote smoking cessation in incarcerated populations: A pilot randomized controlled trial

Introduction: Smoking prevalence rates in prison are typically four times higher than the rates found within community-based settings, increasing premature mortality. Encouraging smokers to self-administer incentives contingent on abstinence (i.e., self-incentivize) are effective in community-based settings and have the potential to work in an incarcerated population. The present study aims to: gauge the feasibility of- and assess possible effects associated with- encouraging smokers to self-incentivize to increase smoking abstinence. Methods: The study recruited 25 smokers from a UK prison who were attending the prison-based stop smoking service. Participants were randomized to one of three conditions, either: (a) an active control, asked to form a plan to quit smoking (n=5); or to self-incentivize if they had not smoked at all by the end of (b) the week (n=7); or (c) the month (n=13). The study gave no specific options to participants in choosing a self-incentive for a number of reasons (e.g., to enable choice of a self-incentive of sufficient rewarding value within a restrictive setting). Participant eligibility included: (a) aged 18 years or over, (b) able to understand written English, (c) competent to provide informed consent, and (d) were tobacco smokers, smoking on a daily basis. Acceptance and completion assessed feasibility. Chi-square analysis assessed possible effect of the intervention. Results: Twenty five out of 27 smokers who were approached, welcomed the opportunity of further support and completed the baseline questionnaire. At 3-month follow-up, no participants (0/5) in the control condition abstained from smoking. In contrast, three (out of seven, d=1.12) participants who self-incentivized on a weekly basis abstained, and five (out of 13, d=0.83) participants who self-incentivized on a monthly basis abstained at the 3-month follow-up time point. Abstaining is defined as no cigarette smoked within at least the last 28-days via self-report and supported by biochemical verification from a sub-sample of participants (68%; 17/25). Conclusions: Use of self-incentives alongside behavioral support and cessation medication is feasible and can be delivered at low-cost. Furthermore, the promise seen in this trial suggests a fully powered randomized controlled trial has the potential to increase smoking abstinence in an incarcerated population.