This well-thought out study by Richmond et al.  addresses smoking treatment in prisoners, a population with high smoking rates and a population that presents serious challenges to successful tobacco dependence treatment. The study was conducted in Australia, but the barriers to good health for prisoners are similar in other countries. As Richmond et al. note, prisoners suffer from lack of education, poor or non-existent job skills, and are disproportionately minorities. They also have a high rate of alcohol and drug abuse, and a high rate of mental health problems . It is to the authors' credit that they have completed such a well-designed and large-scale study in this population.
It is unfortunate that the authors were unable to show a statistically significant difference between active nortriptyline and placebo drug conditions. However, the investigators' rationale for using nortriptyline—low cost and once a day administration—is sound. The question arises as to whether active versus placebo differences would have been observed had a more effective drug been used. There are head-to-head comparisons of nortriptyline with bupropion , but not with varenicline. The comparisons of nortriptyline and bupropion did not find significant differences between the two drugs, although bupropion did produce somewhat higher abstinence rates than nortriptyline. Studies have also shown that varenicline is more effective than bupropion . One would infer from these findings that varenicline might have been a more effective drug to use than nortriptyline, and had it been feasible to use varenicline, active versus placebo differences might have emerged. However, concerns about varenicline's effects on suicide and homicide in this population probably preclude the hope of studying the drug in a prison population until the concerns are resolved.
Despite the lack of differences between active and placebo drug conditions, the abstinence rates reported in this study are good. The authors report continuous abstinence rates of 23.8% versus 16.4% at 3 months for active versus placebo conditions; 17.5% versus 12.3% for active versus placebo conditions at 6 months; and 11.7% versus 11.9% active versus placebo conditions at 12 months. As the authors note, these rates are not strikingly different from those found in the general population.
Targeting prison populations for tobacco dependence cessation would seem to be a worthwhile undertaking. However, many prisons have implemented total smoking bans. For example, as of 2010, approximately half of the prisons in the USA have banned smoking by prisoners and staff on the prison grounds ; it is very likely this number has increased since then. Ironically, these bans are a barrier to the treatment of tobacco dependence in prisons, as, presumably, a prisoner's admission of smoking could lead to penalties. And, as Eldrige and Cropsey  note in surveying US prisons, the culture of smoking in prisons is different than in the outside world. Smoking is normative and cigarettes are used as a means of exchange. Eldrige and Cropsey also note that despite the bans, health risks from smoking increase during incarceration, and that although some smokers do quit as a function of the bans, more start or increase smoking during incarceration. Because smoking is banned, estimates of actual rates are difficult, and those released from smoke-free facilities soon relapse .
The question becomes how efficacious tobacco dependence intervention can be implemented in the presence of prison smoking bans. There are at least two points in the incarceration process when interventions could be introduced despite smoking bans, and it would seem possible to adapt treatment models that have been used successfully in other settings for use at these points. The first point is immediately upon incarceration if the prisoner reports a history of smoking, when pharmacotherapy and behavioral counseling could be provided paralleling that provided to newly abstinence individuals in the outside world, with the goal of facilitating continued abstinence. The second is immediately upon or just before release. Recently, Prochaska et al.  developed and evaluated an intervention for increasing abstinence from cigarettes in psychiatric patients hospitalized on a nonsmoking inpatient unit. All patients were required to be abstinent from cigarettes during hospitalization. The intervention included a transtheoretical model-based computerized program and manual; on-unit counseling session; and nicotine replacement therapy (NRT) during hospitalization with access to NRT post-hospitalization. The usual care provided NRT during hospitalization with brief cessation advice. The intervention was evaluated in a clinical trial with positive results (J. Prochaska, personal communication, 30 January 2013). A similar model might be adaptable to the criminal justice system with ‘aftercare’ linked to appropriate medical or social service providers.
Declaration of interests
The author has received a material grant from Pfizer for one of her studies.