Post-implementation perspectives on smokefree prison policy: a qualitative study with staff and people in custody

Abstract Background A comprehensive smokefree prison policy (SFPP) was introduced in Scottish prisons from November 2018, reflecting concern about inequalities in occupational exposures to second-hand smoke (SHS) and tobacco-related harms among people in custody (PiC). We aimed to address a gap, whereby few studies have sought to understand SFPP from the perspectives of people living and working in prisons. Methods As part of a comprehensive evaluation, focus groups and interviews with staff (n = 99) and interviews with PiC (n = 23) were conducted 6–9 months post-implementation of SFPP in Scotland. Data were analysed using the framework approach. Results Our study found that new restrictions on smoking had been widely accepted by PiC, after a period of adjustment which was less troublesome than participants had anticipated. Benefits of the SFPP for the safety and comfort of staff and PiC who were no longer exposed to SHS, and additionally for the health of PiC who were now smoking-abstinent, were widely acknowledged. Drawbacks of the SFPP, such as difficulties managing without tobacco and use of alternatives (e.g. e-cigarettes and changes in use of illegal drugs), were also reported. Contraband tobacco was not reported to be a major problem following prisons becoming smokefree. Conclusions The findings strengthen evidence that SFPPs can be implemented without causing major disruption and highlight the need for removal of tobacco to be underpinned by careful planning, partnership working and ensuring the availability of support for smokers. Experiences from Scotland may be of interest and some comfort, internationally for jurisdictions considering smoke-free rules in prisons.


Introduction
T he WHO Framework Convention on Tobacco Control 1 requires countries to protect people from exposure to second-hand smoke (SHS) in work and indoor public spaces. When smokefree policies, were introduced in Scotland from March 2006, prisons had partial exemption in the legislation: people in custody (PiC) were permitted to smoke in designated rooms (their cell) and outdoors. In July 2017, the Scottish Prison Service (SPS) and Scottish Government announced their intention to strengthen smokefree policies in prisons, 2 in light of new evidence on SHS levels in Scotland's prisons. 3 The new comprehensive (indoor and outdoor) smokefree prison policy (SFPP) was implemented in November 2018.
Very few studies (see evidence from USA, 4,5 Australia 6 and Taiwan 7 ) have qualitatively explored the perspectives of people working or living in prisons with recently implemented SFPP, and none has comprehensively investigated these across a prison system.
Findings from these existing studies vary, potentially reflecting differing penal contexts and implementation strategies. Nonetheless, the published literature highlights possible challenges of prohibiting smoking in prison (e.g. contraband tobacco 4,5,7 and misuse of NRT 6 ) and factors which may facilitate (e.g. effective communication strategies, good smoking cessation/abstinence support 5,6 ) or impede, a more successful transition to SFPP. This study seeks to build on and enhance the very limited number of previous qualitative studies, by exploring experience of the recently implemented SFPP from the perspective of staff and PiC, providing evidence on opinions on SFPP, success factors and positive/negative consequences once the policy had embedded. The data were collected 6-8 months post-implementation of SFPP, as part of the most in-depth evaluation of SFPP to date [the Tobacco In Prisons study (TIPs)]. Previous papers from TIPs 8-10 and a complementary study of e-cigarette use in prison 11 report views of staff and PiC, prior to implementation of the SFPP using different participant samples. Research settings, sampling and recruitment SFPP was introduced in Scotland's prisons from November 30, 2018. It prohibits PiC from smoking tobacco in any areas of prisons; staff have been prohibited from smoking or using e-cigarettes within prison grounds from 2008. The transition to SFPP was underpinned by a wide-reaching implementation strategy, including enhancement of existing smoking cessation support (see 12 ) and new rules allowing PiC to purchase and use e-cigarettes in designated areas (see 11 ).

Methods
For this study, we analyze data from staff focus groups in Scotland's 14 'closed' prisons, involving 95 participants in total. A focus group conducted in Scotland's open prison was excluded from this analysis, since the issues staff raised were very distinct, as PiC spend some time in the wider community and have access to tobacco whilst outside the prison. Points of contacts in prisons were asked to arrange one focus group, ideally comprising up to eight staff in various work roles and a mix of staff who did or did not smoke and/or vape. In addition, four staff with leading roles in the implementation of SFPP at local (prison) level were interviewed, to capture their perspectives. The combined staff focus group and interview sample (99 participants) comprised: 75 men and 24 women, 83 never-or ex-smokers and 13 current smokers (3 did not report their smoking status) and 26 who reported ever using ecigarettes. Most participants were SPS staff (a few were healthcare staff); most (n ¼ 66) were prison officer grade and had worked in prisons for 11þ years (n ¼ 62).
Interviews with PiC were conducted in six (closed) prisons, selected to represent a range of prisons in terms of size and population (men/women, young people/adults, individuals who were untried/convicted and serving shorter/longer sentences). We chose in-depth interviews, rather than focus groups, with PiC for methodological, ethical and practical reasons (e.g. allowing PiC to speak freely on a potentially sensitive topic, without concern about the complex social and interpersonal dynamics among PiC). Through the staff points of contact, researchers asked that PiC with varied characteristics that might have a bearing on experiences of SFPP (i.e. sex, remand/convicted status and sentence length) were invited to participate. Of the 23 PiC interviewed (18 men, 5 women; 12 aged 40, 11 aged 41þ), all were convicted, 13 serving sentences of 4 years or less and 10 of 4þ years. All were former smokers; although this was not an explicit inclusion criterion it reflects the previously high rates of smoking in prisons prior to smokefree policy.

Data collection
Data collection took place 6-9 months post-SFPP. Focus groups with staff ranged in size (3-14 people). Two joint interviews were conducted at the request of the participants; the remaining interviews were one-to-one. Interviews/focus groups were conducted with only A.B. and/or K.H. present, in a room/area in the prison where participants could not be easily overheard.
Topic guides for interviews and focus groups largely covered similar topics for PiC and staff, informed by the study objectives, existing literature, research team discussions and input from staff within the prison service. They included: participant background; opinions of SFPP; perspectives on living/working in a smokefree prison (including successes/challenges and positive/negative consequences); compliance and enforcement of SFPP and lessons learned.

Data analysis and reporting
De-identified transcripts were thematically analysed using the framework approach, following a process described elsewhere. 11 A.B. led on developing a thematic framework based on close reading of transcripts, study objectives and existing literature. To synthesize and distil material prior to interpretation and abstraction, data were organized under themes and summarized (by A.B. and D.M.) and displayed in a grid format (row¼focus group/interview and colum-n¼theme) in NVivo 12. A.B. reviewed all summaries to check consistency and interpretations. A.B. led the detailed thematic analysis by comprehensively and systematically searching framework grids and reviewing data excerpts, to identify and compare perspectives and experiences of SFPP. Using an iterative process, different dimensions of the data were organized into themes and sub-themes, which were then structured to produce a coherent narrative. All authors agreed final themes based on reviewing transcripts/substantial involvement in data collection. Extracts, selected to evidence and illustrate key findings, are attributed to participants (staff/PiC) using a serial number (letter randomly allocated to each prison for this paper and participant number), an indication of smoking/vaping status (S ¼ smoker, ExS ¼ ex-smoker, NS ¼ never-smoker, V ¼ vaper, ExV ¼ ex-vaper and NV ¼ never-vaper) and, for staff, whether they worked in a prison officer (PO), managerial (MGR) or other role. Supporting extracts are presented in tables 1-4.

Results
We present findings on two key areas: factors contributing to successful implementation and benefits and challenges of SFPP.

Perspectives on policy implementation and success factors
Both staff and PiC indicated that, after a relatively brief adjustment period, the SFPP was widely accepted by PiC and had become the norm in prisons (table 1; Q1 and Q2). However, as before the ban, 8,9 SFPP remained more popular among staff than PiC post implementation. Popularity amongst staff was partly explained by the immediate impact of SFPP in reducing SHS levels in prisons (table 1; Q3). Concerns about restricting PiC's smoking choices and about the necessity and consequences of SFPP continued to be voiced in interviews with PiC (table 1; Q4 and Q5) and, to a lesser degree, by some staff. However, some PiC were, on balance, supportive of the SFPP because of perceived benefits for them personally and for non-smokers who were now protected from SHS (table 1; Q6).
The introduction of the SFPP was generally perceived to have been less troublesome than staff and PiC had anticipated (table 2; Q7 and Q8), and prior fears about the possibility of significant disorder (e.g. 'riots') in prisons had not materialized. Several factors which might have aided the relatively smooth implementation of SFPP were identified. First, the transition to SFPP was reportedly well managed by the SPS at local and national levels (table 2; Q9 and Q10). The decision to stop tobacco sales several weeks prior to the implementation date, and permitting local policies for the removal of tobacco, were highlighted as important aspects of the implementation strategy, since they increased the likelihood of PiC cutting down smoking in anticipation of SFPP.
Second, good communication and engagement with PiC and staff were perceived as important to the successful implementation of SFPP. 'Countdown' posters around prisons were generally said to have ensured widespread awareness of the impending SFPP (table 2; Q11), although it was acknowledged that some PiC had not taken notice of information. Hence, considerable efforts were also made to engage with PiC to understand their views, identify potential solutions to problems, signpost to cessation support and get feedback on issues such as e-cigarettes (table 2; Q12). However, some staff said they would have appreciated earlier communication about the detailed implementation strategy (e.g. introduction of rechargeable e-cigarettes) (table 2; Q13).
Third, collaboration with and input from a range of stakeholders was perceived by prison staff to have been instrumental to success, given the scale and complexity of the task of prohibiting smoking among PiC. This included partnership working across health and justice services (NHS and SPS), engagement with staff and PiC on SFPP and broad acceptance of SFPP among staff and PiC (table 2; Q14 and Q15).
Fourthly, ready availability of smoking abstinence/cessation support in prisons was considered essential for SFPP implementation preparation (table 2; Q16). There was discussion in some staff groups about how this support might need to evolve under smokefree rules. For example, one group discussed integrating support for nicotine addiction with other health promotion activities, to take a more holistic approach to improving PiC's health going forward.
Finally, the de-normalization of smoking in many contexts following the 2006 legislation prohibiting smoking in most public places in Scotland was perceived by prison staff to have aided implementation of SFPP (table 2; Q17).

Perceptions of benefits and challenges of SFPP
The perceived benefits and challenges of SFPP were discussed in relation to three main themes, as discussed below.

Elimination of SHS
The widespread elimination of SHS from prisons was viewed as a significant gain for the health of staff (and also PiC) (table 3; Q18). In some instances, staff discussed experiencing fewer symptoms, such as asthma, a sore throat and eye irritation, following SFPP. Staff, and some PiC, typically commented on improved sensory experiences, such as no longer smelling tobacco smoke in the air, improved appearance and cleanliness of the prison and their clothes no longer smelling of stale smoke (table 3; Q18 and Q19). Longserving staff reflected on how markedly SHS in prisons had reduced over several decades following successive tightening of smoking restrictions. Some expressed disappointment that progress to entirely smokefree prisons had not been quicker.

Smoking abstinence/cessation
PiC generally acknowledged general health benefits of stopping smoking, and some reported improvements in their own health (e.g. in fitness, breathing) following implementation of SFPP (table 3; Q20, Q21 and Q22). However, negative physiological and psychological impacts of SFPP were also raised by PiC and staff, particularly for new arrivals with insufficient funds to purchase (rechargeable) e-cigarettes to help them adapt to a smokefree environment. While these negative impacts reportedly reduced over time for some PiC, others continued to struggle with the SFPP. Some still sought effective strategies for managing nicotine dependence and coping with common problems such as poor mental health or low mood (table 3; Q23) or filling time in prison. A few PiC said that mood changes attributed to smoking abstinence contributed to tension and conflict in prison (table 3; Q24) and a few reported unwanted weight gain (table 3; Q25). The data also highlighted potential opportunities and challenges for extending the benefits of SFPP when abstinent smokers are released from prison. Some suggested the experience of health or financial benefits from not smoking in prison may increase people's motivations to give up smoking long-term, and strengthen beliefs that there is something to lose from smoking relapse (particularly for individuals on longer sentences) (table 3; Q26). There were some suggestions that use of NRT or e-cigarettes may help some individuals to avoid smoking relapse after leaving prison. Conversely, the association of smoking with pleasure, comfort and relaxation, a history of co-use of tobacco and cannabis and returning to environments where tobacco is available/smoked were cited as important potential barriers to remaining smokefree post-release (table 3; Q27 and Q28).

Use of alternatives to tobacco
Both participant groups spoke about the use of alternatives to tobacco, as discussed below.
E-cigarettes, NRT and illicit tobacco. It was reported that most former smokers had switched to vaping following the SFPP. Levels of support for e-cigarettes in prisons remained stronger in PiC than staff, who voiced more diverse views about the advantages and disadvantages of e-cigarettes in prisons. In both samples, benefits of e-cigarettes for implementation of SFPP (table 4; Q29, Q30 Q31 and Q32) and supporting abstinent smokers in the prison population were discussed.
However, staff reported that e-cigarettes had also brought challenges relating to: uncertainties about any health risks of exposures to e-cigarette vapour; misuse of e-cigarettes for drug taking and creation of extra problems for staff to manage, such as when PiC run out of e-liquids (table 4; Q33 and Q34). Other potential risks relating to e-cigarettes in prisons (raised by both staff and PiC) included concerns about continued nicotine addiction amongst PiC (table 4; Q35), user safety and cost. Some staff voiced doubts about whether e-cigarettes would be of net benefit in the long-term and worried that hard-won gains to health from SFPP may be undermined by widespread e-cigarette use.
Data from both staff and PiC suggested that use of illicit tobacco was not a significant problem within Scotland's (closed) prisons post-implementation (table 4; Q36 and Q37) (although there was reported to have been a period immediately post-implementation when stockpiled tobacco was in circulation). The data suggested that the scarcity of illicit tobacco may reflect: general acceptance of SFPP among PiC; availability of e-cigarettes; risks that illicit smoking will be detected and a potentially lower risk-return ratio for smuggling contraband tobacco compared with other items (e.g. illegal drugs) (table 4; Q38 and Q39). The data also suggest that misuse of NRT products (patches) in smokefree prisons is not a major concern (table 4; Q40).
Psychoactive substances/illegal drugs. SFPP was perceived by some staff and PiC to have contributed to changes in the use of 'new' psychoactive substances (NPS), which had already been identified as a problem within the prisons several years prior to the legislative change. 13 These included changes to the method of ingesting NPS (using repurposed e-cigarettes) and suggestions that some PiC may have taken NPS for pleasure/escapism in the absence of tobacco; as a cheaper alternative to vaping and/or as a replacement for previous consumption of smuggled cannabis (which had become increasingly difficult since the sale of tobacco, rolling papers and lighters stopped in prison) (table 4; Q41-Q43).
Both PiC and prison staff spoke about the adverse impacts of use of these unpredictable substances by PiC (table 4; Q44). Concerns about NPS use among PiC included risks for users of acute injury or death and impairment of cognitive functioning, with some participants describing people behaving like 'zombies' after continued use of NPS. For bystanders, particularly staff, concerns centred on risks from passive exposure to NPS or from assault or accidental injury by those under the influence.
By contrast, in some instances SFPP was said to have contributed to a decrease in other illegal drug use in prison, by reducing ease of access to materials such as lighters and rolling papers (table 4; Q45).

Discussion
Our findings, collected as part of a comprehensive evaluation of smokefree prisons, 3,8,10,11,14 suggest SFPP has been widely accepted by PiC and prison staff in Scotland, and the removal of tobacco had been less troublesome organizationally than either group expected. Benefits of the SFPP for the health, safety and comfort of staff following the wholescale reduction of SHS, verified in post-implementation measurements, 14 and the health of PiC no longer smoking were acknowledged. However, participants also reported difficulties of enforced smoking abstinence for certain groups (e.g. new admissions) and use of alternatives to tobacco in prisons (e.g. e-cigarettes, change in use of illegal drugs). The findings support our earlier studies 8,9 that showed that support for SFPP is higher overall among prison staff than PiC and that opinions and experiences of SFPP are complex; it is possible for individuals to be positive about some dimensions of smokefree rules, but negative about others, as similarly illustrated in our contemporaneous surveys of prison staff and PiC. 15 The findings contribute new knowledge about SFPP in several ways. First, they enhance understandings of SFPP, for example by highlighting ways in which some PiC may (re)frame smoking abstinence in prison as beneficial for themselves and others, while also confirming the need to be attendant to the physiological and psychological challenges of enforced smoking abstinence. Second, in contrast to other studies, 6,16 contraband tobacco and misuse of NRT were not reported to have been major problems in prisons in Scotland at the time data were collected. These findings may be explained by high levels of vaping among PiC in Scotland as an alternative to tobacco, in contrast to other jurisdictions which have reported problems with illicit tobacco markets 5,7,16,17 or NRT misuse. 6 Third, although there were a few suggestions from PiC that smokefree rules were contributing to tensions or conflict in prisons, a striking finding from the study is that the introduction of SFPP had been less troublesome organizationally than many PiC and staff expected, reportedly causing no significant disorder in any prisons. This corroborates reports of SFPPs in Australia and the USA, 6,17 and contradicts common media portrayals. 18 Finally, our study strengthens understanding of successful factors for creating smokefree prisons by being the first to comprehensively investigate the perspectives of prison staff and PiC across a prison system post-implementation. The findings provide deeper insight into the role of factors such as extensive planning work; comprehensive communication and engagement with stakeholders and supporting PiC to abstain/quit smoking in successfully implementing SFPP (see also 6,19 ). Scotland's experience also highlights the importance of collaborative working between health and justice services, and, we believe, the benefits of access to findings from ongoing, independent research to inform strategies (e.g. 3 ); all findings from TIPs were shared at the earliest opportunity with the multi-sector team responsible for smokefree implementation. The findings will be of interest to jurisdictions considering SFPP. Lessons from implementing SFPPs in Scotland can also support further public health achievements for PiC.
In relation to e-cigarettes in smokefree prisons, our novel findings are mixed and discussed in more detail elsewhere. 11,20 Perceived benefits for SFPPs were widely discussed by staff and PiC. However, many also voiced concerns about potential risks of e-cigarettes for users, staff (bystanders), and the prison system. These will be key issues for jurisdictions which allow the sale of e-cigarettes to weigh up when planning for SFPP. Strategies for minimizing risks from e-cigarettes are likely to be beneficial if the decision to sell e-cigarettes in prisons is taken in other countries, including learning from Scotland's novel guidance about ways to support PiC who wish to cut down or stop vaping. 11,21 These data were collected in the final phase of the most in-depth evaluation to date of the implementation of a SFPP, conducted across an entire prison estate at three points in time: pre-announcement of plans to implement SFPP, in the lead up to implementation and post-implementation. In this paper, we have included the perspectives of staff in every (closed) prison in Scotland and perspectives of PiC from six prisons which collectively house diverse populations. The size and composition of the samples enabled a range of views and experiences to be captured, and so expand understandings of smokefree prisons considerably. Another strength is that data were collected using well-established and robust methods, 6-9 months post-implementation when staff and PiC had been able to adjust to the SFPP and possibly had a clearer sense of its benefits and drawbacks. In relation to limitations, some perspectives (e.g. of people on remand) may be missing from this study, either because data were not collected from PiC in every prison in Scotland or due to self-   selection bias. Future studies could investigate the views of PiC who are never-smokers. This part of the TIPs study was not designed to quantify potential impacts of the smokefree policy; ongoing work is analysing prison and health service data on outcomes of interest in an economic analysis of the SFPP. The findings highlight the need for ongoing investment to maximize the long-term gains of SFPP. In future work we intend to explore how best to support PiC to remain smokefree post-release.

Conclusion
The findings substantially strengthen international evidence that SFPP can be implemented, and maintained, without major organizational disruption. Factors promoting success include: careful system-wide planning, engagement and communication strategies, collaboration across relevant organizations and services, and supporting PiC to abstain or quit smoking, including through evidence-based interventions.