Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The impact of smoke-free policies on inpatient psychiatric units : an ethnographic study Grant, Lyle George 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2013_spring_grant_lyle.pdf [ 2.74MB ]
JSON: 24-1.0073609.json
JSON-LD: 24-1.0073609-ld.json
RDF/XML (Pretty): 24-1.0073609-rdf.xml
RDF/JSON: 24-1.0073609-rdf.json
Turtle: 24-1.0073609-turtle.txt
N-Triples: 24-1.0073609-rdf-ntriples.txt
Original Record: 24-1.0073609-source.json
Full Text

Full Text

THE IMPACT OF SMOKE-FREE POLICIES ON INPATIENT PSYCHIATRIC UNITS: AN ETHNOGRAPHIC STUDY  by  LYLE GEORGE GRANT  B.Comm., Dalhousie University, 1985 J.D., University of Alberta, 1988 B.S.N., University of British Columbia, 2004 M.S.N., University of British Columbia, 2007  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE STUDIES (Nursing)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)    March 2013  ? Lyle George Grant, 2013     ii Abstract  Smoke-free grounds policies (SFGPs) were introduced in acute psychiatric hospital settings to help improve health among patients, staff, and visitors. However, enacting these policies has been challenging. Recognizing that cultural norms around tobacco use may influence policy enactment, a qualitative ethnographic study was undertaken in Northern British Columbia, Canada to improve understandings about how SFGPs are affected by institutional cultures. Data included participant observation, document analysis, informal fieldwork discussions (n=11), and interviews with patients (n=20), healthcare professionals (n=19), and key informants (n=2) at two hospitals. Data were analyzed using iterative processes to inductively derive thematic findings and develop cultural understandings. Cultural factors supported some healthcare professionals in subverting and resisting the SFGP while advocating and caring for patients. Strong consultative leadership, including input and participation by those most directly responsible for policy implementation, offered the strongest indication that policy-maker intent could be implemented. This study highlights the actions taken and challenges faced by those implementing SFGPs in inpatient psychiatric settings. Consistency in implementing the SFGP across the organization was a significant challenge, influenced by local context, the nature of the policy, resource availability, and healthcare professional discretion under the policy. Patients responded to the SFGP in a variety of ways, but ultimately remained resigned to smoking and believed hospitals had a duty to accommodate them and their smoking. The centrality of smoking was rooted in personal beliefs sustained by both healthcare professionals and patients, and enforced through group norms. The study offers new evidence about the importance of local and cultural contexts to SFGP implementation and reports for the first time how rurality may influence SFGP implementation in psychiatric settings in Canada. Local contexts and cultural factors can be conceptualized in a socio-ecological model of intra-personal, inter-personal, institutional factors, and community/environment spheres of influence, to structure inquiry and analysis of SFGP implementation. It is suggested that policy-makers avoid oversimplified macro-level approaches to SFGP development and implementation in favour of more localized, simultaneous, top-down, bottom-up approaches, with accompanying     iii support for those most directly involved in implementation efforts to improve policy fidelity and any needed shift of cultural norms.      iv Preface Ethics approvals for this study were obtained from the following research ethics boards prior to commencement of data collection: ? University of British Columbia Behavioural Research Ethics Board, Certificate No. H09-00724 ? University of Northern British Columbia Research Ethics Board, Certificate No. E2009.0420.062 ? Northern Health Research Review Committee, Reference No. RRC-2009-0014        v Table of contents  Abstract .................................................................................................................................... ii Preface ..................................................................................................................................... iv Table of contents ..................................................................................................................... v List of tables............................................................................................................................ ix List of figures ........................................................................................................................... x Acknowledgements ................................................................................................................ xi Dedication ............................................................................................................................. xiii Chapter 1 - INTRODUCTION........................................................................................... 1 Background and context ....................................................................................................... 1 Problem statement ................................................................................................................. 4 Statement of purposes and research questions ...................................................................... 4 Rationale and significance .................................................................................................... 4 Definitions of key terminology used in this study ................................................................ 5 Presentation of the dissertation ............................................................................................. 5 Chapter 2 - LITERATURE REVIEW .............................................................................. 6 The culture of tobacco use on psychiatric units .................................................................... 7 Physiological effects of smoking related to mental health patients ...................................... 9 Addiction ........................................................................................................................ 10 Physiological interactions ............................................................................................... 11 Medication interactions .................................................................................................. 12 Use of smoke-free policies .................................................................................................. 13 Barriers to implementing smoke-free-policies in psychiatry .............................................. 15 Staff attitudes .................................................................................................................. 15 Human rights arguments ................................................................................................. 17 Ethics arguments ............................................................................................................. 18 Stigma and discrimination .................................................................................................. 20 Smoke-free policies and best practice patient care ............................................................. 20     vi The local policy environment ............................................................................................. 21 Chapter 3 - RESEARCH METHODS ............................................................................. 25 Rationale for ethnographic approach .................................................................................. 25 Data collection procedures .................................................................................................. 28 Site selection and entr?e ................................................................................................. 28 Fieldwork ........................................................................................................................ 29 Participant observations .................................................................................................. 30 Document collection ....................................................................................................... 32 Interviews ....................................................................................................................... 33 Research sample.................................................................................................................. 34 Recruitment procedures and consent .............................................................................. 34 Sample ............................................................................................................................ 35 The healthcare professional sample ........................................................................... 36 The patient sample...................................................................................................... 38 The key informant sample .......................................................................................... 38 Data analysis ....................................................................................................................... 40 Enhancing rigour ................................................................................................................. 45 Ethics... ............................................................................................................................... 47 A word about format ........................................................................................................... 47 Chapter 4 - FINDINGS: HEALTHCARE PROFESSIONALS? PERSPECTIVES ... 49 The Northern British Columbia context of Sites A and B .................................................. 49 The smoke-free grounds policy........................................................................................... 51 Policy implementation ........................................................................................................ 52 Site A ? Maintaining the status quo .................................................................................... 53 Locating Site A ............................................................................................................... 54 Policy as the focus of resistance ..................................................................................... 57 Patient advocacy through knowing best ......................................................................... 60 A culture of ignoring ...................................................................................................... 63 Tobacco as essential to patient relations ......................................................................... 66 Site B ? New practice challenges ........................................................................................ 68 Locating Site B ............................................................................................................... 68 Strong consultative leadership ........................................................................................ 71     vii Healthcare professionals? discretion in practice ............................................................. 76 Ambushed?The challenges of broad compliance ......................................................... 79 Scarcity of resources ....................................................................................................... 82 Conclusions ......................................................................................................................... 85 Chapter 5 - FINDINGS: PATIENTS? PERSPECTIVES .............................................. 87 Resigned to smoking ........................................................................................................... 87 Responding to power and control ....................................................................................... 93 Policy responses ................................................................................................................ 101 Conclusions ....................................................................................................................... 106 Chapter 6 - DISCUSSION AND RECOMMENDATIONS......................................... 108 Importance of context ....................................................................................................... 109 Intra-personal context: Individual beliefs and attitudes .................................................... 110 Inter-personal context: The influence of group norms ...................................................... 114 Inter-personal context: Leadership and consensus building ............................................. 116 Institutional factors: Organizational change ..................................................................... 121 Community: Locale-specific norms .................................................................................. 124 Limitations ........................................................................................................................ 127 Recommendations ............................................................................................................. 127 REFERENCES .................................................................................................................... 133 Appendices ........................................................................................................................... 149 Appendix A - Smoke-free Grounds Policy Document ..................................................... 150 Appendix B - Participant Observation Guide ................................................................... 152 Appendix C - Document Collection Guide ....................................................................... 154 Appendix D - Healthcare Professional Interview Guide ................................................... 156 Appendix E - Patient Interview Guide .............................................................................. 158 Appendix F - Key Informant Interview Guide .................................................................. 160 Appendix G - Key Informant Demographics Form .......................................................... 162 Appendix H - Healthcare Professional Consent Forms .................................................... 166 H.1 ? Healthcare Professional Informal Discussion Consent Form ....................... 166 H.2 ? Healthcare Professional Interview Consent Form ........................................ 170 Appendix I - Healthcare Professional Demographics Form ............................................. 174 Appendix J - Housekeeping Staff Email Invitation .......................................................... 178     viii Appendix K - Patient Invitations ...................................................................................... 180 Appendix L - Patient Consent ........................................................................................... 181 Appendix M - Patient Demographics Form ...................................................................... 185        ix List of tables Table 1 ? Healthcare Professionals? Demographic Information................................. 37 Table 2 ? Patient Demographic Information .............................................................. 39      x List of figures Figure 2.1 Northern Health Region Map .................................................................... 23 Figure 6.1 Spheres of Influence ................................................................................ 110      xi Acknowledgements First and foremost, I am most grateful for the cooperation, interest, and enthusiasm provided by the participants in this study. Improving the quality of care and life for those with mental illness is part of an ultimate goal that I believe participants in this study shared with me. I also appreciated the support that was often so quickly provided by key people at Northern Health. Amongst many of the supports required, they helped facilitate introductions to participants, access to sites, diagrams, histories, and troubleshooting where required. The results of this study are through the combined efforts of many. I acknowledge the untiring support of my supervisor, Dr. John Oliffe, who has contributed greatly to my development as a researcher, scholar, mentor, and teacher. Throughout my PhD dissertation, I felt privileged to work with Drs. John Oliffe, Joy L. Johnson, and Joan Bottorff as my supervisory committee. They provided guidance and support and encouraged a sense of excellence in scholarly inquiry that has shaped my own efforts toward excellence and meaningful inquiry. Any shortcomings in my dissertation are entirely my own.  I offer my sincere thanks to the many faculty members at UBC School of Nursing, and fellow nursing and PhD students who were and continue to be my instructors, mentors, and colleagues. They willingly provided feedback and supportive discussion about my PhD work that helped shape my development and thinking during my studies. To Gael Whelan, your support and pride, along with the lessons and advice you shared about what makes for an excellent and caring nurse, continue to profoundly and positively affect who I am as a nurse and the work I do.  I wish to acknowledge the assistance and contributions of Dr. Annette Schultz who, together with Dr. Vivian Ramsden, Nancy Snowball, and Margaret Green, initially worked with me on a study grant application related to exploring smoke-free grounds policies in psychiatric settings. Their ideas, enthusiasm, and stories helped shape my own ideas for studying the phenomenon I explored in the current study. Dr. Schultz contributed significantly to the development of the demographic and document collection tools used in my study as well as some of the questions used during participant interviews, in anticipation     xii that some of the data collected would be comparable to that collected in a separate study where I was a co-applicant.  I recognize and thank my spouse, Wanderley, for steadfast and continued support throughout my studies. This support proved the essential element to successful completion of my PhD work.  Finally, I am grateful for the substantial funding support I received during my PhD study. In particular, I acknowledge the University of Northern British Columbia, Northern Health, and the Interdisciplinary Capacity Enhancement (ICE) project at University of Waterloo for funding the direct research expenses related to this dissertation work. Other major sources of funding for my PhD studies included: UBC ? Four Year Fellowship for PhD Students UBC ? PhD Tuition Fee Award  Pacific Century Scholarship, Province of British Columbia  UBC ? Catalyst Paper Corporation Fellowship Award  UBC ? Janet Gormick Memorial Graduate Scholarship in Nursing  Registered Nurses Foundation of BC ? Helen Margaret King Memorial Bursary Registered Nurses Foundation of BC ? Elgin & Vivian Lockridge Bursary        xiii Dedication  I dedicate this work to the many who have supported me in developing as a nurse and to those living with mental illnesses who continue to endure struggle. My hope is for improvement.      1  In this study I explored the enactment of a smoke-free grounds policy (SFGP) at two inpatient psychiatric settings in Northern British Columbia (BC) to produce descriptions and analysis of the experiences and perceptions of inpatients and healthcare professionals (HCPs) in these settings. The purpose of this ethnographic study was to improve our understandings of how the implementation of SFGPs is affected by institutional cultures. The findings advance knowledge and understandings to guide policy-makers and HCPs with regard to developing, implementing, and evaluating smoke-free policies in inpatient psychiatric settings.  This chapter begins with an overview of the background and context that frames the study. This is followed by a problem statement, study purpose, and specific research questions. This chapter concludes with an outline of the rationale and significance of this study, along with definitions of key terms used.  Background and context Smoking is a major cause of morbidity and a leading cause of death (Health Canada, 2005; Makomaski Illing & Kaiserman, 2004; Prochaska, Fletcher, Hall, & Hall, 2006; U.S. Department of Health and Human Services, 2004; U.S. Department of Health and Human Services, 2010). Over the last 50 years public policy in Canada and other Western countries has encouraged smoking cessation, and in many cases this has resulted in decreased smoking rates and reduced second-hand smoke in the general population (U.S. Department of Health and Human Services, 2006; U.S. Department of Health and Human Services, 2010). That said, some subgroups continue to have higher than average smoking rates. Among those are people who experience severe and persistent mental illness (SPMI). Estimates of prevalence of smoking in people with SPMI range from 32?88%, and those with more severe mental illness tend to be more likely to smoke and smoke more heavily (i.e., consuming more than 20 cigarettes a day) (Dalack, Healy, & Meador-Woodruff, 1998; Davidson et al., 2001; Hughes, Hatsukami, Mitchell, & Dahlgren, 1986; Lawrence, Mitrou, & Zubrick, 2009; Schroeder & Morris, 2010; Shetty, Alex, & Bloye, 2010). People with mental illness Chapter 1 -  INTRODUCTION     2 purchase between 44.3% and 46.3% of all the U.S. tobacco sold, and collectively that subgroup accounts for 40.6% of all smokers in the U.S. (Dome, Lazary, Kalapos, & Rihmer, 2010; Grant, Hasin, Chou, Stinson, & Dawson, 2004; Lasser et al., 2000). These statistics are likely similar in Canada. The persistently high prevalence rate of smoking among the mentally ill is complex and multifaceted, and tobacco control policies can be integral to influencing tobacco use and exposure to second-hand smoke among people with SPMI. The current study investigated the implementation of smoke-free policies within two psychiatric inpatient settings to better understand how those policies play out from staff and patients? perspectives.  Second-hand smoke is estimated to be responsible for serious diseases that include heart disease and lung cancer in otherwise healthy non-smokers (Ontario Tobacco Research Unit, 2001; U.S. Department of Health and Human Services, 2006). Second-hand smoke also increases occupational risk, fire risk, and health risk to staff (Rudnick, 2002). Smoke-free buildings are key to eliminating harmful exposure to second-hand smoke (U.S. Department of Health and Human Services, 2006). Policy-makers have become increasingly concerned about second-hand smoke exposure, and numerous tobacco control policies have enforced smoking bans around public buildings (Ontario Tobacco Research Unit, 2007). To reduce exposure to second-hand smoke, health authorities in Canada extended smoke-free buildings policies by targeting smoke-free grounds, as well as bolstering stringent tobacco-free grounds policies (Cowan & Langley, 2004; Kunyk, Els, Predy, & Haase, 2007; Parle, Parker, & Steeves, 2005). The Northern Health region of BC continues to have persistently high smoking rates. Whilst BC has a smoking prevalence rate of around 18%, the Northern BC rate of 26% remains higher than in any other health region in the province (BC Stats, 2007). Subsumed within these statistics are smoking rates of 42% among people 20 to 24 years old, a rate more than double that for any other provincial health region. The Northern Health region also has the highest proportion of Aboriginal persons (13%) in BC (Northern Health, 2009) and this subpopulation group has an estimated 36% smoking rate (BC Stats, 2007). Smoke-free building policies and SFGPs have often been differentially and inconsistently applied to psychiatric inpatient settings (Ratschen, Britton, & McNeill, 2008;     3 Smoke-Free (Exemptions and Vehicles) Regulations, 2007; Vancouver Coastal Health, 2008; Williams, 2008), and the experiences of implementing smoke-free policies within these environments are poorly understood. The presence of tobacco smoking in psychiatric settings is long established?a centrepiece in the history of mental institutions?and only recently has more systematic and critical investigation begun (Lawn & Pols, 2005). The idea of imposing smoking bans in psychiatric settings is a recent phenomenon (Lawn & Pols, 2005). Rationales for differentiated policy approaches in psychiatric inpatient settings where those with SPMI reside or seek treatment have not been fully articulated or necessarily supported by empirical evidence. The differential application of policy may be attributable to the unique cultures that typically embrace and affirm tobacco use within psychiatric settings (Reilly, Murphy, & Alderton, 2006). These cultures entrench tobacco to include its use as commodity, currency, therapy, and instrument to change or govern behaviour (Grant, 2007; Lawn & Condon, 2006). Moreover, smoking is shown to modify the effects of medication commonly taken by psychiatric patients (Desai, Seabolt, & Jann, 2001; Dome et al., 2010; Kroon, 2007; Williams, Gandhi, & Benowitz, 2010; Ziedonis, Williams, & Smelson, 2003), and conversely, some medications commonly administered in psychiatric settings affect smoking behaviours (Barr, Procyshyn, Hui, Johnson, & Honer, 2008; Keltner & Folks, 2005; Matthews, Wilson, & Mitchell, 2011; McEvoy, Freudenreich, & Wilson, 1999; Procyshyn, Ihsan, & Thompson, 2001; Procyshyn, Tse, Sin, & Flynn, 2002; Williams et al., 2010). Tobacco and medication interactions are reflected in both the use and embodied experiences of smoking among those with SPMI. Furthermore, despite recent changes to smoking policies in inpatient psychiatric settings, there are predictions for, and some anecdotal evidence of, inconsistent enforcement. These differential approaches to tobacco policy implementation and enforcement within psychiatric settings have been influenced by human rights arguments (Ratschen, McNeill, Doody, & Britton, 2008), ethical concerns (Campion, McNeill, & Checinski, 2006; Gray, 2004; Head, 2003; Kunyk et al., 2007; Lawn & Condon, 2006), suggestions that such policies have the potential to harm patients (Lawn & Pols, 2005), and staff feeling unprepared to provide humane care where smoking cessation is not the choice of the patient (Wolfenden, Campbell, Wiggers, Walsh, & Bailey, 2008; Ziedonis & Williams, 2003).      4 The impact of SFGPs has not been systematically investigated within psychiatric settings in Northern BC or extensively elsewhere, and without such exploration our understandings about SFGPs are limited particularly in regard to staff and patients? perspectives. These are vital insights for making predictions about the overall effectiveness of the policies. This qualitative study was designed to explicitly address this knowledge gap by exploring the enactment of SFGPs. Moreover, by ?being there? and observing the cultures that can affirm and disaffirm smoking within inpatient psychiatric settings, and by interviewing the end-users and potential ?enforcers? of smoke-free policies, the findings derived from this study afford important contextual insights.  Problem statement SFGPs are often introduced without consideration of their unintended consequences. If tobacco control and smoke-free polices are to be fully actualized, policy-makers must understand how their implementations shape and are shaped by institutional cultures.   Statement of purposes and research questions  The purpose of this ethnographic study was to improve our understandings of how the implementation of SFGPs is affected by institutional cultures. The overarching research question guiding the current study was: How do SFGPs impact inpatient psychiatric units in Northern BC? To address this overarching question, two research questions guided the study: 1. What are the SFGP-related experiences of HCPs in inpatient psychiatric units, and how do cultural norms and institutional structures shape these experiences? 2. What are the SFGP-related experiences of patients in inpatient psychiatric units, and how do cultural norms and institutional structures shape these experiences? Rationale and significance The rationale for this study originated from my interest in how SFGP enactment is shaped by cultural aspects of tobacco use among patients experiencing mental illness, amid anecdotal accounts of recent implementation challenges for smoke-free policies. Challenging     5 longstanding ties between smoking cultures and inpatient psychiatric units is ?essential in improving the health of both patients and staff? (Reilly et al., 2006, p. 277). Improved understandings about how tobacco use and mental illness interface in response to new smoking policies provides valuable information about how best to work with this patient subgroup and these treatment locales to advance smoke-free policies. It also guides policy-makers and HCPs in developing, implementing, and evaluating smoke-free policies in inpatient psychiatric settings. Definitions of key terminology used in this study Healthcare professional (HCP) ? includes any person providing direct care services to participant inpatients of acute psychiatric units, including nurses, recreational therapists, and physicians. Inpatient psychiatric unit ? refers to a unit or ward location within an acute care hospital dedicated to providing psychiatric or mental health services to persons who are admitted for a planned stay of at least one night.  Presentation of the dissertation This dissertation is presented in a series of chapters. The second chapter provides a review of relevant literature in highlighting the knowledge gap addressed by the current study. Study methodology is presented in the third chapter, and findings are presented in Chapters 4 and 5. Chapter 4 includes a description of the study setting and introduces the two data collection sites; it then answers the first research question focusing on HCP perspectives. The sites are labelled as Sites A and B to help preserve institutional and participant confidentiality. Chapter 5 addresses the second research question, presenting findings related to patient perspectives. In Chapter 6, findings are discussed in the context of other research amid implications, followed by limitations and recommendations.      6 The purpose of this ethnographic study was to improve our understandings of how the implementation of smoke-free grounds policies (SFGPs) is affected by institutional cultures. In this chapter a review of relevant literature is presented in order to locate the current study and point to the knowledge gap it addresses. A brief outline of important physiological effects of tobacco use in relation to severe mental illness and use of psychotropic agents is provided, as are implications of neurobiological theories of nicotine addiction. The use of smoke-free policies to promote health is also detailed and contrasted with mental illness treatment facilities and general hospital settings. Barriers to implementing smoke-free policies in psychiatric settings and notions of stigma and discrimination are outlined to locate the current study within the experiences reported by others, and then smoke-free policies are considered within best practices for patient care to locate policy intent. Finally, as important background to the study, the local policy environment of Northern BC is described, and considerations about rural and remote psychiatric patients and services are reviewed for the purposes of defining and describing the locale of the current study.  In locating and selecting the reviewed literature the following databases were searched: PsycINFO, Embase, PubMed, Cochrane Systematic Reviews, Academic Search Premier, CINAHL, ProQuest, ACP Journal Club, BIOSIS, ERIC, and Web of Science. Various keywords, subject headings, and MeSH terms were applied systematically, adding single additional search terms one at a time, until search results numbered approximately 200; then titles were reviewed for relevance to the research questions. Search terms included: smoking, mental disorder, mental illness, hospital, policy, healthcare personnel, and health personnel. After selecting titles, abstracts were reviewed for relevancy, and full articles were read. Searches were made to describe the presence of tobacco use in inpatient psychiatric units, the physiological effects of smoking in mental health patients, and the use of smoke-free policies in acute care facilities that included inpatient psychiatry units. Article exclusion criteria included those published before 1993, not in English, not related to adult populations, not related to acute care hospital smoking policies, related to residential or forensic psychiatric care facilities only, and that primarily focused on smoking cessation interventions.  Chapter 2 -  LITERATURE REVIEW      7 The culture of tobacco use on psychiatric units  Smoking activity has been described as a cultural norm in psychiatric settings, and some attribute the high rates of smoking amongst those with severe and persistent mental illness (SPMI) as primarily influenced by cultural factors (Crockford, Kerfoot, & Currie, 2009; Voci et al., 2010). Reilly, Murphy, and Alderton (2006) described smoking as ?endemic within mental health facilities worldwide, with an entrenched smoking culture within these settings, among both staff and patients? (p. 276). Many clinicians believe smoking to be inextricably linked to mental illness (Prochaska, 2011). The presence of tobacco smoking in psychiatric settings is long established and central to the history of mental institutions, and it has only recently been more systematically and critically investigated (Lawn & Pols, 2005). The idea of imposing smoking bans in psychiatric settings is a recent phenomenon (Lawn & Pols, 2005), and the implementation of these bans has begun in various Canadian psychiatric settings over the last few years (Kunyk et al., 2007; Schultz, Bartmanovich, et al., 2010; Schultz, James, et al., 2010; Schultz, Finegan, Nykiforuk, & Kvern, 2011; Voci et al., 2010).  Measured differences between staff attitudes toward smoke-free policies on general inpatient units compared to those on psychiatric inpatient units confirms the uniqueness of psychiatric units. Data collected in a U.K.-based survey study indicated that 1 in 10 staff on general inpatient units opposed smoking bans, whereas 1 in 3 on inpatient psychiatric units opposed such bans (McNally et al., 2006). Studies revealing divergent levels of compliance with non-smoking policies between hospitals with psychiatric services versus those without also affirmed the need to better understand the smoking practices among inpatient psychiatric patients (Joseph, Knapp, Nichol, & Pirie, 1995). Similarly, findings from a U.S. survey study on smoking policy compliance conducted during hospital accreditation processes revealed that hospitals without psychiatric or substance abuse units were more likely to have stricter non-smoking policies than those hospitals that did (Longo et al., 1998; Williams, Hafner, et al., 2009). Collectively, these studies illustrate differences in policy implementation between psychiatric inpatient units and other hospital units that raise questions about what underpins these prevailing trends.      8 Qualitative studies have tended to focus on the role smoking plays in the lives of those with SPMI and affirm the uniqueness of the circumstances and experiences of smoking to that subpopulation. York?s (1997) phenomenological study revealed smoking as a central activity of those with SPMI that formed part of a person?s identity. A study by Lawn (2004) linked cigarettes and smoking to the mental illness institutional culture, and described an enculturation of smoking reinforcement that manifested as using cigarettes as a form of currency for economic, social, and political exchange. Those who work in psychiatric institutions have also witnessed a ?rich smoking reinforcement history? and a ?system that condoned smoking by staff and patients and accepted the clinical use of cigarettes to assist patients with their mental illness management? (Lawn & Condon, 2006, p. 113). The use of cigarettes by those with SPMI has been positioned as helping manage psychiatric symptoms and side effects from medications, and to create a sense of normalcy, each of which operates to tie smoking to personal identity (Grant, 2007; Lawn, Pols, & Barber, 2002; Prochaska, 2011; Solty, Crockford, White, & Currie, 2009). Cigarettes can operate as a black market currency and commodity, and as a source of influence over others (Grant, 2007; Kagan, Kigli-Shemesh, Tabak, Abramowitz, & Margolin, 2004). Tobacco can also be a source of friction on inpatient psychiatric units that requires management (Rich & Knowlden, 2002; Voci et al., 2010). Psychiatric hospitalizations are often involuntary and unexpected for those with SPMI, and too strict a policy may create underground smoking activities that put others at risk. Where patients on psychiatric inpatient units are detained involuntarily under mental health legislation, human rights violations have been suggested where smoking bans are imposed (Kagan et al., 2004). This line of reasoning is consistent with legislative and regulatory provisions that permit incarcerated persons to smoke in the buildings in which they are confined (Explanatory Notes to Health Act, 2006). This suggests that in settings where individual liberties are restricted, there is shared concern about curtailing other human rights. Human rights violation is a central argument for resistance to smoke-free policies in mental illness treatment settings, and by big tobacco companies (Katz, 2005). In a study using grounded theory methods, Solway (2009) interviewed leaders of smoking cessation programs to explore barriers to addressing the cultural role of tobacco use in psychiatric settings. Study participants reported that psychiatric patients felt further marginalized if     9 denied opportunities to adopt healthy practices through treatments offered others that promoted smoke-free lifestyles. Staff members were seen as role models for smoking behaviours for those with SPMI. Based on the findings, Solway provided several recommendations. First, reframe failure to address tobacco use with those with SPMI as discriminatory. Second, find alternatives to smoking that are pleasurable, promote socialization, and are easily incorporated into mental health settings. Third, frame smoking cessation as exercising one?s choice to more closely align with systems of care that increase focus on recovery-based, person-centred care and treatment. Fourth, reduce smoking rates among staff to address a cultural change by severing the tobacco connection between staff and clients. Best smoke-free policies were suggested as those that were sensitive to staff, offered easy access to cessation programs at work, and acknowledged staff for personal quitting attempts or for helping clients quit.  Secured psychiatric units have been recognized as particularly complex social environments with unique dynamics and interplay between social actors (Campion et al., 2008). Smoking bans in these settings potentially affect patients, staff, and visitors in unique ways. In particular, Campion and colleagues highlighted a lack of research within psychiatric settings that adequately reviewed the cultural, structural, political, and environmental contexts in which smoking bans are implemented and that reviewed whether variation in implementation strategies is required in response to localized circumstances and contexts. They suggested adequate interrogation and understanding of the nuances of the social context in which policy is applied as critical to successful smoke-free policy implementation. The current study addresses this knowledge gap by examining the enactment of a SFGP through the lens of culture. Locating findings contextually also affords insights about how best to mobilize smoke-free policies to advance the well-being of psychiatric patients and staff.  Physiological effects of smoking related to mental health patients Understanding the embodied experience of smoking is important to understanding the meaning of actions and interactions exhibited by the study participants. A brief review of the literature demonstrates how smoking for those with SPMI might differ from those in other subpopulations.      10 Addiction Cigarettes contain over 4,000 chemicals of which the full effects on mental illness, addiction, and psychotropic agents are not fully known (Dome et al., 2010; Ziedonis et al., 2003). Addiction to smoking has complex neurobiology and may be based on any one or number of its constituent components or chemicals, although nicotine is the most examined (Dome et al., 2010). Nicotine stimulates the release of dopamine and other brain neurotransmitters, like ?-aminobutryric acid (GABA), glutamate, norepinephrine, acetylcholine, and serotonin (Benowitz, 2008; Dome et al., 2010; Evans & Drobes, 2008; Williams et al., 2005; Wing, Wass, Soh, & George, 2012). These same neurotransmitters are often considered in the diagnosis, etiology, and treatment of mental illness and are targeted for modulation in drug therapies used in mental health. This may help explain increased smoking activity in those with mental illness and smoking as a form of self-medication, as well as support understanding of a neurobiological model of nicotine addiction that links new understandings of psychiatric disorders involving nicotinic receptors. Neurobiology theories of sensitization-homeostasis help explain how addiction can occur after just one cigarette, why individuals continue to smoke additional cigarettes after reporting previously unsatisfactory experiences of smoking, and why some individuals can smoke only a few cigarettes a day over long periods of time without withdrawal symptoms following quitting (Gardner et al., 2009). Complex arrays and composition of nicotinic subtype receptors and the plasticity of neural networks within the human brain may explain diversity in individual experiences of dependence, withdrawal, and addiction. Nicotine binds to nicotinic receptors in the brain to contribute to a reward system that increases dependency (Aubin, Rollema, Svensson, & Winterer, 2012; Dome et al., 2010; Wing et al., 2012; Ziedonis & Williams, 2003) and may help account for the higher incidence of smoking and higher prevalence of heavy smokers in this population group. The neurobiology of nicotine addiction may include genetic components that influence susceptibility to addiction, and addiction behaviours and experiences that help explain diversity in individual experiences and perceptions of tobacco (Aubin et al., 2012; Benowitz, 2008; Gardner et al., 2009; Novak et al., 2010). Biological and physiological investigations indicate that individuals with SPMI may be predisposed to initiating and maintaining smoking behaviours; therefore vulnerability theories and social     11 and environmental determinants can all play a key role in improving understandings of the comorbidity of smoking and SPMI (Morisano, Bacher, Audrain-McGovern, & George, 2009; Moss et al., 2010; Wing et al., 2012). Understanding nicotine addiction from a neurobiological perspective supports understandings that the addiction may have biological components that can be targets of treatment and may also counter assumptions that the addiction is solely habit or socially based.  Physiological interactions Effects of nicotine may create pleasure, arousal, learning, memory enhancement, mood modulation, and reduction in anxiety and tension through activation of a series of neurotransmitters (Aubin et al., 2012). Nicotine withdrawal creates a state of decreased dopamine, leading to a state of malaise and inability to experience pleasure, termed hedonic deregulation and/or depression (Aubin et al., 2012; Benowitz, 2008). Withdrawal symptoms may include irritability, impatience, restlessness, insomnia, anxiety, problems getting along with others, concentration difficulty, increased hunger and eating, constipation, and craving for tobacco (Gardner et al., 2009). Non-nicotine chemicals in tobacco can enhance the availability of dopamine, and nicotine increases dopamine release. Dopamine is clinically significant in numerous mental disorders (Keltner & Folks, 2005) and plays an important role in the subjective experiences of illness in those with SPMI. The effects of increased dopamine may account for an increase in positive symptoms with accompanying decrease in negative symptoms in those with schizophrenia (Ziedonis & Williams, 2003). Nicotine is also thought to improve cognition through actions that reduce signal-to-noise ratios in the prefrontal cortex, improve sensory gating, and otherwise modulate cognitive processes that improve attention and working tasks (Aubin et al., 2012; Dome et al., 2010; Leonard et al., 2001). The modulating effects associated with tobacco support theories that tobacco consumption is used for self-medication by those with SPMI. Finally, biological evidence that nicotinic receptors may have abnormal composition or operation in those with schizophrenia, attention-deficit hyperactivity disorder, Alzheimer?s disease, and Tourette?s syndrome supports different embodied experiences of smoking and dependence in those with SPMI (Dome et al., 2010; Leonard et al., 2001; Ziedonis & Williams, 2003). Understandings of how physiological and biological mechanisms in those with SPMI are affected by tobacco     12 use are emergent, but there is evidence to support embodied experiences that differ from those without SPMI. Medication interactions A select review of the pharmacokinetic and pharmacodynamic properties of concomitant use of tobacco and psychotropic agents is important because it informs understanding of what might affect participant experiences, interpretations, and actions. Nicotine and most psychotropic agents are metabolized in the liver by the cytochrome P-450 (CYP-450) system, which partially explains the interaction between nicotine and medications (Dome et al., 2010; Urichuk, Prior, Dursun, & Baker, 2008; Williams et al., 2010). The enzyme substrate 1A2 has been implicated in the metabolism of common psychotropic agents, including clozapine, haloperidol, olanzapine, phenothiazine, selective serotonin reuptake inhibitors (SSRIs), venlafaxine, and nefazodone (Keltner & Folks, 2005). The tar component of tobacco has been revealed as a potent inducer of the 1A2 substrate of CYP-450 creating an increased metabolism of many antipsychotics, antidepressants, and antianxiety medications (Kroon, 2007; Ziedonis & Williams, 2003). The polycyclic aromatic hydrocarbons in cigarette smoke induce CYP substrates 1A1, 1A2, and 2E1, while carbon monoxide and cadmium present in cigarette smoke may inhibit CYP enzymes generally (Desai et al., 2001). Evidence of the effects of smoking on benzodiazepines and their CNS effects is mixed (Desai et al., 2001). Clearance rates of the antipsychotic drug olanzapine have been shown to be up to 55% faster in smokers than in non-smokers or past smokers (Bigos et al., 2008; Desai et al., 2001). From a clinical perspective, smokers using psychotropic agents who are suddenly faced with tobacco abstinence may experience higher serum levels of medications and thus increased side effects from these medications without dose adjustment (Kroon, 2007; Ziedonis & Williams, 2003). Even with dose adjustment, patients may experience a new period of dose titration based on previous smoking routines, now changed.  Medications used for psychiatric purposes also affect tobacco use (Matthews et al., 2011). Clozapine can reduce smoking compared to those treated with conventional antipsychotics (McEvoy et al., 1999), those treated with depot neuroleptics (Procyshyn, Ihsan, & Thompson, 2001), and those treated with risperidone alone (Procyshyn et al., 2002).     13 Bupropion is a drug used for smoking cessation, but it is also commonly used for mental illness treatment (Keltner & Folks, 2005). Chlorpromazine has been reported by smokers with schizophrenia to increase their motivation to smoke as drug doses increase (Barr et al., 2008). Carbamazepine is suspected to increase nicotine metabolism that may lead those on this drug to smoke more (Williams et al., 2010). These drugs are provided as examples of how dosage and drug changes might impact smoking behaviours and reported motivations for smoking and reinforces the linkages between psychotropic agents used by those with SPMI and changes to smoking behaviours.  Use of smoke-free policies The literature was reviewed to gain understandings about the use of smoke-free policies and relevant policy environments of the SFGP. In this section, relevant literature relating to smoke-free policies is presented that reflects smoke-free policy use in health promotion, in hospitals generally, and in mental illness treatment facilities. This review helped position the SFGP in relation to other relevant smoke-free policy initiatives. Early tobacco control efforts tended to focus on smoking cessation and the individual, affording a ?one person at a time? approach. Later tobacco control approaches focused on the broader, community level and collaborative interventions that targeted identifiable smoker groups using media advocacy and policy instruments, including legislation and taxation to influence behaviour change (National Cancer Institute, 2007). Increased public interest in protecting individuals from known and suspected harmful effects of second-hand smoke resulted in a number of policy initiatives to regulate smoking in public or confined spaces (Pierce & Le?n, 2008; U.S. Department of Health and Human Services, 2006). These policy initiatives attempt to establish smoke-free areas through enactment of various forms of legislation, regulations, and/or organizational policies.  Systems approaches can be key to effecting desired health promotion changes in tobacco use at both the epidemiological and personal health levels (National Cancer Institute, 2007). The current study is one response to examining tobacco control from more complex perspectives to enhance understandings about how cultures shape the enactment of tobacco control policies. How smoke-free policies have been used in health promotion, in hospitals generally, and in mental illness treatment facilities follows.     14 While various smoke-free policies were primarily aimed at protecting people from exposure to second-hand smoke, also recognized were important opportunities to promote the physical health of those with SPMI and reduce health inequities within this subpopulation (Jochelson, 2006). In a study conducted in 1989, when smoking was allowed on inpatient psychiatric units, it was estimated that nurses spent about 15 minutes of their time each shift managing cigarette use (Resnick & Bosworth, 1989). For units that manage patient smoking through the use of smoking breaks, it was estimated that up to four hours of a staff member?s time could be redirected from tobacco management activities to offering cessation interventions to improve health outcomes (Prochaska et al., 2006).  The experiences of implementing smoke-free policies within mental illness facilities are poorly understood. Smoke-free policies in U.S. hospitals have been heavily influenced by the Joint Commission on Accreditation of Healthcare Organizations and the tobacco control standards, announced in 1991, to move all accredited hospitals to smoke-free by 1994 (Longo et al., 1998). Smoking bans at hospitals enjoy strong support in the U.S., but some for-profit, and federally owned hospitals are less likely to have smoke-free policies in place (U.S. Department of Health and Human Services, 2006; Williams, Hafner, et al., 2009). Designated smoking rooms are ineffective at adequately protecting people from the harmful effects of second-hand smoke (U.S. Department of Health and Human Services, 2006). Goals underlying SFGPs include reducing exposure to second-hand smoke on hospital buildings and grounds and provision of a health message discouraging tobacco use (Cowan & Langley, 2004; Fox & Shovein, 1993; Parle, Parker, & Steeves, 2005). The experiences of smoke-free policy development and implementation in hospitals generally can be contrasted with that of mental health settings to support identifying different cultures of tobacco use between the two settings. The implementation of smoke-free policies in mental illness treatment facilities has often been delayed, postponed, or otherwise differentially applied when compared to policies affecting other healthcare facilities (Prochaska, 2011; Schultz et al., 2011). This is a well-documented phenomenon in many Western countries. For example, in England, where a single legislative act has wide-reaching effects, the Health Act 2006 (c. 28) provided a one-year exemption to mental illness treatment facilities before requiring smoke-free status, and     15 exempted long-term stay facilities. In the U.S., the move to develop smoke-free psychiatric facilities is incomplete and often delayed or complicated by political action, lawsuits, and stalled policy initiatives (Williams, 2008). In New South Wales, Australia, a survey of mental health inpatient units found a 98% presence of smoking bans on indoor inpatient spaces, but only 34% of the units had SFGPs, with many continuing to permit smoking on verandas, courtyards, unit perimeters, and grounds (Wye et al., 2009). In Canada, there is no single legislative act that regulates smoke-free spaces, and various pieces of provincial and municipal legislation must be consulted. In BC, provincial legislation provides some regulation of smoke-free spaces in hospitals, but permits smoking in designated smoking areas (Tobacco Control Regulation, BC Reg. 232/ 2007). Policy initiatives to provide smoke-free spaces in inpatient psychiatric units have largely been at the discretion of regional health authorities within the province, resulting in disparate application, and generally excluded private healthcare facilities. As one of the leading health authorities in BC, Vancouver Coastal Health initially provided exceptions to its SFGPs to including mental health/addictions services, residential facilities, and palliative care services (Vancouver Coastal Health, 2008). At the largest health authority in BC, Fraser Health, newly introduced smoke-free policies were stringent, but delayed implementation one year for mental illness facilities, and residential mental health facilities did not impose the stricter SFGP and continued to permit designated outdoor smoking areas (Fraser Health, 2008a; Fraser Health, 2008b). In the Northern Health region, the site of the current study, policies provided for smoke-free grounds at all facilities it primarily funds by the year 2008, but provided timing and implementation exceptions for Mental Health and Addiction Services (see Appendix A). Barriers to implementing smoke-free-policies in psychiatry Implementing smoke-free policies in inpatient psychiatric settings has been challenging for a variety of reasons. Challenges include barriers to implementation as a result of 1) staff attitudes, 2) human rights arguments, and 3) ethics arguments. These barriers are discussed next. Staff attitudes Studies have indicated that staff in inpatient psychiatric settings are likely to oppose smoking bans. Prochaska (2011) suggested five myths healthcare professionals (HCPs) may     16 believe that contribute to this resistance and wide acceptance of tobacco use by those with mental illness: 1) tobacco is necessary for self-medication, 2) those with mental illness are uninterested in quitting smoking, 3) those with mental illness cannot quit smoking, 4) recovery from mental illness is confounded by removing smoking as an important coping mechanism, and 5) smoking is not a priority issue for those with severe mental illness. A Swiss study indicated that 87% of the staff rejected the idea of a smoking ban (Etter & Etter, 2007) and, in a Canadian study, close to 40% opposed a smoking ban (Voci et al., 2010). U.K.-based studies showed that a majority of staff in psychiatric hospitals opposed total smoking bans, and those who smoked were more permissive in their attitude toward smoking on inpatient units than were non-smokers (McNally et al., 2006; Stubbs, Haw, & Garner, 2004). Support for smoking bans in psychiatric settings may be stronger in HCPs who are non-smokers than in those who are smokers (Dougherty et al., 2002; Voci et al., 2010), and HCPs who currently or formerly smoked may be reluctant to provide tobacco dependence treatments to patients (Sarna, Bialous, Wells, & Kotlerman, 2009). HCPs with lower education were also less likely to offer tobacco dependence treatments (Ratschen, Britton, Doody, Leonardi-Bee, & McNeill, 2009; Sarna et al., 2009). Nurses were more permissive of smoking than were physician groups (Stubbs et al., 2004). Nurses in the U.K. described anticipated smoking bans on inpatient psychiatric units as especially challenging, and anticipated staff overload, logistic and safety issues, and inadequate treatment of tobacco withdrawal (Snow, 2006). Staff in inpatient psychiatric settings often supported smoking in designated areas, allowing staff to smoke with patients, based on the belief that it enhanced therapeutic relationships (Stubbs et al., 2004). However, more recent studies demonstrated a shift to increasing staff support for smoking bans (O'Mahony & Rahmani, 2004; Praveen, Kudlur, Hanabe, & Egbewunmi, 2009; Wye et al., 2010), indicating that support may increase subsequent to smoking ban implementation (el Guebaly, Cathcart, Currie, Brown, & Gloster, 2002; Hempel et al., 2002; Lawn & Pols, 2005; Voci et al., 2010).  One project to reduce tobacco use on a psychiatric unit described the most significant barriers as ?staff resistance, habit, and the leaden hand of administrative procedure? (Rich & Knowlden, 2002, "Barriers & Difficulties? section, para 1). In inpatient psychiatric settings, managers believed these settings faced particular challenges arising from high smoking     17 prevalence, related safety risks, adverse effects on patient?HCP relationship, and potential interactions with antipsychotic medication (Ratschen, Britton, & McNeill, 2009). Staff resistance included concerns over patient rights violations, anticipated violence, assault, relapse, exacerbation of symptoms, suicide, involuntary discharge, death, and other disasters (Jochelson & Majrowski, 2006; Rich & Knowlden, 2002; Voci et al., 2010).  Even though there are some negative consequences to smoking bans in psychiatric illness facilities, there is some evidence to indicate that such bans do not typically increase levels or incidents of aggression, discharge against medical advice, or use of as-needed medication, and may serve to reduce adverse events (de Nesnera, Folks, & Rauter, 2012; Hollen et al., 2010; Lawn & Pols, 2005; Shetty et al., 2010). Generally, staff anticipated more problems with policies that ban smoking than actually materialized (Lawn & Pols, 2005; Voci et al., 2010). A review article by Lawn and Pols (2005) detailed various unintended negative results of smoking bans and illustrated the mixed-results of various studies. Generally, however, the literature suggests smoking bans produce beneficial outcomes (Moss et al., 2010). It has been argued that the most important elements influencing the implementation of tobacco bans are attitude and culture (de Nesnera et al., 2012; Rich & Knowlden, 2002). Key recommendations for successful policy implementation and avoidance of unforeseen problems have included consistent, coordinated, full administrative support for the application of smoking bans, strong leadership to champion the cause, and supporting and continuing education that assists HCPs in enforcement and treatment measures (de Nesnera et al., 2012; Lawn & Pols, 2005; Moss et al., 2010; Voci et al., 2010).  Human rights arguments Human rights arguments tend to take two forms. First, non-smokers have the right to clean air and to be protected from harmful substances, the only solution for which is a total smoking ban (Ontario Tobacco Research Unit, 2001; U.S. Department of Health and Human Services, 2006). Second, individuals have the right to choose to smoke or assume health risks in their private lives that might include smoking (Appelbaum, 1995; Ratschen, McNeill, et al., 2008). Psychiatric HCPs who smoke are more likely to endorse these arguments, favouring patient rights to smoke (Dwyer, Bradshaw, & Happell, 2009). Legal arguments have also been advanced purporting inequitable application of laws or smoking ban     18 exemptions between those in locked psychiatric inpatient units or residences and those in prisons (Appelbaum, 1995; Ratschen, McNeill, et al., 2008). Patients with mental illness have advocated on multiple fronts for their ?right to smoke? (Hirshbein, 2010). Smoking restrictions have generally been upheld through judicial processes, but these types of arguments are likely to continue to be relevant in smoke-free policy discussions in mental illness treatment facilities. Ethics arguments Public health policies, including tobacco control policies, are ethical if they do no harm (Gray, 2004). Philpot (1994) suggested that smoking bans imply utilitarian ethical approaches to health care that place the good of the many before those of the individual, amid reminders to readers that a more humanist approach to understanding health might pay attention to the individual experience of health. An emphasis on an individual experience of health may include recognizing individual benefits of smoking rather than invoking unexpected cessation and withdrawal. Nurses generally advocate for patients?a position mandated as a professional standard?and few patients who are smokers advocate for smoking bans. Furthermore, Philpot has suggested that enforcing or policing a non-smoking policy is likely to fall within the duties of nursing staff, and this may undermine individualized holistic care and therapeutic nurse?patient relationships to ?paradoxically, reduce a nurse?s effectiveness as health promoter? (p. 310). Internal conflict can emerge as nurses attempt to reconcile policy implementation with individualistic, humanistic, and patient advocacy practices, a tension that likely contributes to staff resistance to smoke-free policies. The advocacy role of HCPs working with vulnerable persons with SPMI can be especially pronounced. As Head (2003) suggests of palliative patients: To coerce smokers who happen to be in hospital with an unrelated condition into accepting smoke-free behaviour as a condition of their care may be questionable. When patients have no prospect of benefit from smoking cessation, and enforced abstention aggravates their existing distress, they are being managed unethically. (p. 104) While Head?s plea is to afford pleasure to dying patients, similar pleas resonate in the care of inpatient psychiatric patients with arguments to preserve rights for self-determination (Lawn & Condon, 2006). On inpatient psychiatric units, nurses have been reported to evaluate     19 patients? mental illness concerns as priority over concerns about the longer term health effects of smoking (Lawn & Condon, 2006). Furthermore, nurses perceived smoking cessation interventions as community based rather than acute care based (Lawn & Condon, 2006). Tobacco abstinence is increasingly understood as a way to improve psychiatric symptoms in patients and not to harm them, so neglecting tobacco as a treatment priority raises ethical concerns (Cole, Trigoboff, Demler, & Opler, 2010; Hitsman, Moss, Montoya, & George, 2009). Ethical considerations can also appeal to employers? and health authorities? obligations to provide healthy and safe environments to non-smoking staff and patients at mental illness treatment facilities (Campion et al., 2006). These arguments typically rely on utilitarian principles of ethics, advancing legally structured arguments of fiduciary obligations owed by employers and health authorities to others. In mental illness treatment facilities, some staff may experience ethical conflicts between a public health agenda that relies on the utilitarian principle of doing the most good, and responding to the individual needs of people who may be distressed, non-comprehending, and in crisis (Campion et al., 2008). Adding to this dilemma are HCP perceptions of a ?duty to care,? mandating assistance aimed at improving health in patients who smoke (Ratschen, Britton, Doody, & McNeill, 2009). It is the involuntary nature of patient detention, absent or impaired capacity to comprehend, and the nature of the health crisis that potentially distinguishes the reality experienced by patients on psychiatric units from non-psychiatric acute care units. A closely related argument in opposition to the implementation of smoke-free policies relates to staff concerns that they cannot adequately provide ethical care, either because they lack the education and competencies to manage the immediate needs of patients in tobacco withdrawal or because they lack the necessary policy and organization supports to provide adequate care and implementation of the policy (Wye et al., 2010). It has been suggested that a number of measures can support a smoke-free policy and alleviate many of these concerns. Included are 1) providing adequate medication orders for nursing staff to use in managing withdrawal symptoms, 2) improving integration of tobacco reduction and treatment procedures in patient care, 3) ongoing educational support to staff, 4) adequate protection and safeguards to patients who must leave the hospital property to smoke, and 5)     20 adequate security staff or staff supports for managing safety and enforcement issues (Kunyk et al., 2007). The need for patients and staff to leave hospital property to facilitate smoking exposes individuals to inclement, sometimes dangerous weather, thereby increasing risk of personal injury, and increasing the hospital?s and staff?s exposure to legal liability. These dangers can influence staff and, without adequate education, policy, and organizational supports, increase staff resistance to smoke-free policies.  Stigma and discrimination Stigma and discrimination is commonly experienced by those with SPMI and has been linked to patient experiences with health care (Corrigan, 2004). Stigma perpetuates stereotyping, discrimination, and prejudice rooted in dominant cultural beliefs that are harmful to the social opportunities and self-esteem of those with SPMI (Corrigan, 2004). Stigma has been conceptualized and understood in a variety of ways and its contingency on differential access to social, economic, or political power may help explain why those with mental illness, or those associated with them, are positioned vis-?-vis others and how separation, status loss, and discrimination may unfold in various contexts (Link & Phelan, 2001). Worldviews common in Western countries have also been suggested as broad influences that reinforce stigma associated with mental illness (R?sch, Todd, Bodenhausen & Corrigan, 2010). In the mental health field, Corrigan has characterized public stigma as what a na?ve public does to a stigmatized group in differentiating that those members of a stigmatized group can also internalize stigma as self-stigma. Others have referred to these distinctions as external versus internal stigma and highlighted how internalized stigma may manifest as embarrassment about illness, limiting self-disclosure and help-seeking (Lee, Kochman, & Sikkema, 2002). Understandings of stigma and discrimination were necessary to understanding the experiences and perceptions of the participants in the current study. Smoke-free policies and best practice patient care  Humane enforcement of smoke-free hospital policies requires an examination of best practice recommendations for patients who smoke and helps place in context the experiences and perceptions of HCPs in the current study. Implementing a smoke-free policy, particularly as this might affect involuntarily admitted patients, requires concomitant provision of care     21 that addresses nicotine withdrawal to remain ethical. Additionally, if the sudden abstinence from tobacco is left unaddressed, it can compromise psychiatric care. Best practice guidelines for treating and supporting tobacco abstinence have common characteristics. Primary goals include identifying those who smoke, providing advice to quit, offering scientifically proven tobacco use and dependence treatments, and enhancing motivation (Fiore et al., 2008; Hitsman et al., 2009). Care for nicotine withdrawal should include: 1) an assessment of smoking status and recording it in the patient record, 2) provision of support for inpatient abstinence, 3) assessment for and provision of nicotine replacement therapy, 4) assessment and adjustment of currently prescribed psychotropic agents as may be affected by the withdrawal of nicotine, 5) monitoring for and support and medication assistance with the management of withdrawal symptoms, 6) support and direction to resources to help with cessation including outpatient supports, and 7) treatment and discharge plans that address tobacco use (American Psychiatric Association, 2006; Irish Health Promoting Hospitals (HPH) Network, 2006; New South Wales Department of Health, 2008; Wye et al., 2009). While best practices for supporting cessation efforts on inpatient psychiatric units are not fully developed, knowledge is advancing on how best to tailor treatment of tobacco dependence among those with SPMI (Hitsman et al., 2009).  The local policy environment The written SFGP at the centre of this study (please see Appendix A) was supplemented with a binder of materials titled ?Smoke Free Grounds Policy Toolkit.? This Toolkit was available to all employees of the health authority and comprised key elements of the implementation strategy to be applied across the organization. Inpatient psychiatric units were classed under Mental Health and Addictions groups with Special Consideration to the set policies and included unique implementation strategies (please see Appendix A). A section of the Toolkit binder was specifically reserved for written guidelines related to these Special Considerations, but the anticipated guidelines never actually eventuated for the specific context of inpatient psychiatric units. In the absence of written guidelines, unit-specific policies and implementation strategies were primarily communicated through series of staff meetings leading up to introduction of the policy. Minutes of the staff meetings     22 available to staff on the sites were reviewed, but did not reveal detailed written accounts of the unit-specific policies relating to the SFGP.  To locate the policy context of the current study and to detail who implements policy in mental illness treatment facilities, a brief review of the health management administrative structures in British Columbia and Canada is provided. Canadian regional health authorities are administrative branches of provincial governments responsible for governing, planning, and coordinating the delivery of public health services within pre-determined regional areas (Ontario Hospital Association, 2002). The Northern Health authority services the needs of those living in a northern geographic region of BC (see Figure 2.1). Services are delivered through approximately 4,000 full-time equivalent positions across the region, and this includes 24 acute care facilities with 544 acute care hospital beds, 14 long-term care facilities with 900 residential care beds, plus various public health units and offices providing specialized health services (Northern Health, 2009).  Northern BC?s towns and settlements are largely rural or remote, and its cities are small by Canadian standards. Within BC, rural zones have fewer healthcare professionals per capita, and individuals have additional challenges accessing services, including travel time (Ryan-Nicholls & Haggarty, 2007). Healthcare service delivery, policy development, and implementation also differ in rural and remote locations from those in urban areas. For example, in rural settings, policy interpretation is more variable than in urban settings and, in comparison to urban practitioners, rural practitioners often lack the amenities and resources, are required to be more autonomous in their practices, but have fewer educational options, and cannot always separate work from community membership (MacLeod et al., 2008; Ryan-Nicholls, & Haggarty, 2007). Rural and remote are terms without universal definition (Bollman & Clemenson, 2008; Pitblado, 2005). In examining rural HCP practices, some studies suggest that factors of isolation, access to amenities, socio-demographic characteristics of the communities in which HCPs live and work, availability of healthcare resources, and the character of HCP practices may be important to those living in rural and remote settings (Kulig et al., 2008; MacLeod et al., 2008).       23  No published studies were located that focused on rural perspectives about smoke-free policies on inpatient psychiatric units. Rural locations are often unique and poorly understood environments for enactment of policy. Rural issues in the delivery, access to, and use of mental health services are also known to be notably different (Hunter, 2006; Ryan-Nicholls & Haggarty, 2007). The nature and extent to which some of these issues may affect enactment of SFGPs at inpatient psychiatric settings has not been systematically studied. Existing literature affirms the importance of tobacco control policy and health promotion initiatives, the predominance of tobacco use within mental health care settings, and the special character of inpatient psychiatric units. Reported barriers to smoke-free policies and best practice in caring for those with SPMI who smoke locate the current study within broader contexts. Little is reported in the literature that includes patient accounts of experiences with SFGPs at inpatient psychiatric settings. The literature also lacks research that contains detailed accounts and systematic reports about how institutional cultures affect Figure 2.1 Northern Health Region Map  (Supplied by Northern Health, 2009. Used with permission)     24 the implementation of SFGPs in inpatient psychiatric settings, particularly those in more rural or northern settings, so this study addresses these gaps to improve understandings of how SFGPs are affected by institutional cultures.        25 This chapter includes a description of the research methods used and discussion of the rationale for the research design. Within this chapter I also describe the data collection procedures, the research sample, approaches to data analysis, rigour, and ethics procedures. A short section about the format of the dissertation is also included.                Rationale for ethnographic approach An ethnographic approach was well suited to the current study, and its focus on describing human interactions derived from studying people in their everyday contexts, to better understand implementing a smoke-free grounds policy (SFGP) (Emerson, Fretz & Shaw, 1995). Qualitative research is similarly grounded in examining human phenomena in naturalistic settings to describe and improve understandings about the contexts and social settings in which people act and how they view their own and other?s actions (Hammersley & Atkinson, 1995). The research questions posed for the current study also enlist naturalist approaches to examine two inpatient psychiatric units in Northern British Columbia. In this regard, the SFGP was viewed a product of, and its implementation affected by, a variety of social structures. Naturalist approaches are undergirded by knowledge that understanding peoples? behaviours are accessed through understanding the meanings that guide these behaviours (Hammersley & Atkinson, 1995). Policy implementation was therefore viewed as manifest through the behaviours of individuals affecting and affected by the policy. Ethnographic methods are also suited because cultures of tobacco draw upon norms about how participants embody and speak of shared beliefs and meanings, tacit understandings of acceptable behaviour, and of simultaneously inhabiting multiple cultures. Ethnography in focussing on describing cultures enhanced understandings about the participants and social settings in the current study (Spradley, 1979).  There is no universal definition of culture, and the conceptualization of culture is a social construction that may shift over time. For the purposes of my research, I adopted a view of culture as shared, acquired knowledge that people use to generate and interpret social behaviour (Spradley, 1979; Spradley & McCurdy, 1972). Cultures can be described through inferences derived from what people say and do, and the artefacts they use (Spradley, 1979), Chapter 3 -  RESEARCH METHODS     26 and ethnography allows a way of seeing that permits glimpses of culture without moving to full definitions of culture (Wolcott, 2008). I applied understandings of culture that recognized culture as dynamic and changing, and participants? experiences of culture were dependent on contexts including social, political, locale and structures (Lynam, Browne, Reimer Kirkham, & Anderson, 2007). Multiple cultures were viewed as operating contemporaneously whereby  individuals participated in these multiple cultures, both shaping and being shaped by them in ways that provided a matrix of possible selves (LeCompte, 2001); some of these potential selves may have been observed by, or revealed to me, and others not. Through an ethnographic lens, individuals may be seen as ?historically and socially situated entities engaged in constructing their own realities through interaction with others in the social, political, and cultural environments? (LeCompte, pp. 291-292). Charon (2007) described culture as representing the stability of a group, with such stability having dimensions that are enduring, dynamic, negotiated, and influenced by changes around the social players. In that respect participants were viewed as social players in multiple co-existing cultures, influencing and influenced by an array of contexts, and as a researcher, I was afforded  glimpses of these cultures on which to interpret cultural milieus. Any resultant description or explanation of culture contained in this study is therefore acknowledged as partial, contestable, temporal, and emergent (Clifford & Marcus, 1986).  Organizational or institutional cultures formed an additional way to understand the multiple cultures operating at one time within the study environment and form part of the puzzle that make up the organization implementing the SFGP (Jung, 2009). Like definitions of culture, organizational cultures are understood to pertain ?to the multiple aspects of what is shared among people within the same organisation: for example, beliefs, values, norms of behaviour, routines, traditions, sense-making, et al.? (Parmelli, et al., 2011, p. 2). Dynamic and emergent, I accepted that organizational cultures emerge and develop based on the interactions of individuals within organizations that contain ?mosaics of realities, understandings, and meanings? rather than being of a uniform nature (Dan???man, 2010, p. 203).  In focusing on what people do, a cultural lens or ethnographic way of seeing focuses upon shared concepts, beliefs, principles of action, and organization similar to those in     27 anthropological traditions (Wolcott, 2008). Questions about what an individual must know to behave acceptably as a member of a group help to reveal the norms and values of the dominant culture (Wolcott, 2008). Accepting culture as an orienting concept, the data collection and analysis in the current study focused on examining conventions, customs, practices, shared knowledge, and beliefs as motivating or guiding what people do, rather than finding focus in trying to define people within a prescribed culture (Wolcott, 2008). In the current study, the influences of cultural norms on enactment of the SFGP were explored to reveal how cultural norms might be transgressed or reinforced. Culture may be viewed as part of that which characterizes a social group. These groups have shared meaning systems that individuals use to organize behaviour, interpret themselves and others, and make sense of the world (Spradley, 1980; Wolcott, 2008). When viewed in this way, individuals may be socially grouped in more than one group within the same setting, and cultures may be examined at various grouping levels. For example, patients participating in the current study are members of a group who exhibit and participate in interactions with each other to produce cultural identities as psychiatric patients; however, they are also social actors within the larger unit setting, interacting with staff, and therefore part of a culture group that is the psychiatric unit. Cultural understandings of the hospital in which groups reside can also be derived. In the current study, social groupings of greatest interest to answering the research questions were derived from the data. Ethnographic approaches suggest potential differences in ?emic? (cultural insider) and ?etic? (cultural outsider) perspectives. Important information about the SFGP was uncovered by identifying and contemplating differences in perspectives, determining who was viewed as an insider or an outsider within social groupings. In the current study, patients and HCPs constituted distinct social groupings. Ethnography provided a methodological fit for this inquiry because of its focus on discerning difference as a means to uncover cultural elements that influence SFGP enactment (Wolcott, 2008). Understanding the difference between emic and etic views of SFGP provided a useful tool for analysis and to inform policy-makers in pragmatic ways.       28 The overarching research question guiding the current study was: How do SFGPs impact inpatient psychiatric units in Northern BC? To address this overarching question two research questions guided the study: 1. What are the SFGP-related experiences of HCPs in inpatient psychiatric units, and how do cultural norms and institutional structures shape these experiences? 2. What are the SFGP-related experiences of patients in inpatient psychiatric units, and how do cultural norms and institutional structures shape these experiences? Data collection procedures  Data collection was designed to capture the experiences of participants related to the SFGP. Two inpatient psychiatric units were selected as research sites, and the research sample was drawn from HCPs, managers, and staff working there, and from patients currently on or recently discharged from the units. Data collection commenced approximately one year post policy implementation and concluded nearly two-and-a-half years post policy implementation. Data collection during this time enabled me to ask about participants? experiences after implementation, anticipating that participants recall specific events and circumstances in sharing their current perspectives. Site selection and entr?e  Within the Northern BC Health region, two hospital-based adult, inpatient psychiatric units were intentionally selected. Each site afforded sampling diversity in terms of patient age, primary diagnosis, ethnicity, and rurality. The Northern Health region is predominantly rural and occupies approximately two-thirds of the most northerly portion of BC (Northern Health, 2009). The region has approximately 300,000 inhabitants, and Site A was located in the largest urban centre, with about 71,000 inhabitants (11.4% of whom are Aboriginal peoples) (Statistics Canada, 2006). Site B, approximately 575 kilometres west of Site A, is in a city with approximately 11,300 inhabitants (21% Aboriginal peoples) (Statistics Canada, 2006). The city in which Site B is located is the more rural of the two when considering factors of nearness to population centres over 50,000, access to services, inhabitants? perceptions of rurality, and remoteness (MacLeod et al., 2008; Pitblado, 2005).     29 The two sites were selected to ensure that adequate sampling could be achieved to complete the study and to offer both urban and rural perspectives. The two sites offered opportunity to compare the findings in distilling diversity and patterns applicable to Northern BC.  I gained entry to the research sites and accessed potential participants through connections with the individuals in the local health authority, some key individuals within Mental Health and Addiction Services, and the health authority?s Tobacco Reduction Office. These connections did not, however, extend to familiarity with the two sites included in the current study, so my entr?e to the unit was first as an outsider with some social knowledge about nursing, health care delivery, and Northern BC. Four data collection methods were used: fieldwork, participant observation, document collection, and interviews. I conducted all data collection myself, with the exception of four hours of participant observation and 5.5 hours of field observations that were conducted by a trained research assistant in an effort to compare interpretations around what was observed. Examining the research assistant?s observation notes also helped me examine my own potential bias to the hospital setting as a nurse and allowed me to reflect upon the details I was observing and recording to increase my understanding of the effect I had on the data. I attended unit meetings to discuss the study and invite participation, and attended one regional team leader meeting to solicit assistance with study recruitment. Additionally, assistance was requested from unit support staff and the unit team leaders for purposes of encouraging participation and communicating with staff. The data collection methods are discussed in turn below. Fieldwork Fieldwork included mapping the hospital grounds and adjacent public lands and examining the hospital grounds for evidence of where smoking activities occurred and for signage and materials related to the SFGP. Similarly, the psychiatric units were mapped and physical characteristics recorded. Digital photographs were taken and reviewed to complete field observations and to assist me in recalling and describing the settings. I then developed detailed narrative descriptions of each of the sites.      30 Participant observations Participant observations, a defining feature of ethnography that assisted in illuminating and understanding data from other sources, were key to witnessing and interpreting firsthand the SFGP implementation (Spradley, 1980; Wolcott, 2008). Participant observations focused on tobacco-related interactions within and between HCPs and patients. A nursing station location was selected at each site for participant observations to reduce the potential for paranoia in patients and to permit patients the opportunity to refuse consent to be observed by moving away from me during participant observation periods.  HCPs were advised in team meetings, memos, and through personal contact with me about scheduled observation and reminded when the participant observations were about to occur. HCPs were told that the observations related only to SFGP enactment. Assent for participant observation was obtained by asking HCPs individually or in a group meeting for confirmation that they consented to being observed and having the observer make notes about these observations. Participants were advised that they could refuse assent and would be excluded from the observations and researcher notes without consequence to or effect on their work or employment. No HCP refused assent.  Assent for observation of patients was obtained around mealtimes or patient community meetings on the units, when the largest attendance of patients could be accessed as a group. At these times, I introduced myself, explained the study?s purpose, and sought patient assent to participate (or my research assistant performed these steps). Patients were advised that they could participate or refuse to participate without consequence to their care or treatment on the unit. They could refuse by making this known to me, my research assistant, or their nurse, or by moving their conversations away from the location where my research assistant or I was observing. My research assistant and I confined ourselves to the nursing station during participant observation periods so that patients and HCPs had the opportunity to move away from the nursing station to signal that their interactions were not to be observed. Patients were generally eager to participate and asked questions about the study during the assent procedure. On one occasion at each site, a patient requested not to be observed, a request that was honoured by ceasing participant observations and deleting existing related field notes.     31  A participant observation guide (Appendix B) was used. Early observation periods focused on the unit routines. In addition to tobacco related actions and interactions I focused on how staff were dressed and routinely behaved toward each other as well as who came and went from the unit. As time lapsed and I became more familiar with these orienting observations, and as participants developed a comfort level with my presence, I was able to focus on tobacco related interactions, increasing the detail of what was observed. In each fieldwork session I recorded start and finish details in observing the staff and patients present on the unit. Jottings regarding tobacco interactions focused on writing key phrases of what was said, observing speech patterns and intonations, and non-verbal actions. I developed interpretations over time and repeated exposure to the environments and checking for consistency about what I saw at different times. Notes of events were made during observation periods and then augmented with interpretations in a separate column or in brackets to distinguish descriptive observations from my own preliminary impressions (Spradley, 1980). Capturing my impressions in this manner at or near the time of observation prompted me to ask what contributed to these impressions to help explain the actions and motivations of participants and thus helped apply an ethnographic lens to the observation data. Participant observation occurred over separate blocks of time spanning up to two weeks at each of the sites. Blocking the observation periods in this fashion helped ensure HCPs and patients became habituated to my presence and increased the validity and reliability of the data (Fetterman, 2010; Wolcott, 2008). I noticed how staff marginalized me less as I spend more time on the unit and how their tone in speaking with me moved from reactive and suspicious toward indifference and some acceptance of my presence. Over the course of the study I reflected upon previous observations and my interpretations to increase data validity. Observation periods were generally limited to about four hours to ensure that the observer remained attentive. In recognizing that I could not observe or record everything going on I focussed on specificities (i.e., actions, behaviours, objects, speech) to further develop  my awareness and interpretations of the data (Spradley, 1980). Observation periods were conducted at various times of the day?morning, afternoon, and evening?and included all days of the week, to best provide an overview of unit routines and capture related     32 smoking activities. At Site A, participant and field observations were conducted for 37.5 hours and 12.5 hours respectively, between July 6 and 29, 2009. At Site B, participant and field observations were conducted for 32 hours and 7.5 hours respectively, between August 10 and 19, 2009. I also attended the units for HCP and patient interviews between July 2009 and December 2010 and was able to observe the general activity of the unit and engage in informal conversations with HCPs, which augmented the more focused observation periods.  Participant observation data were collected at both sites early in the data collection process as a means to orient and sensitize me to possible similarities and differences, which I explored further in subsequent data collection. Interview data were collected in sequence from one site at a time to assist with the contemporaneous analysis. Data collection extended beyond the anticipated collection time for several reasons. Unsuccessful attempts to attract HCP participants at Site A in the Fall of 2009, and during the December and January holiday periods resulted in delays for conducting individual interviews at Site A. This delay also afforded separation time between the participant observations and interview data collection at Site B, which, in hindsight, benefited my incremental analysis and enabled me to further refine my individual interview guide.  Participant observation and attendances on the unit also engaged me in spontaneous and informal discussions with HCPs on subjects related to the study. To capture and use the data obtained from these conversations, written consent was obtained from each participant. This allowed for the collection of a broad range of perceptual, contextual, and experiential data from HCPs. In total, 16 HCPs provided data through these informal conversations, adding the perspectives of 11 HCP participants to the study. These informal discussions also permitted me to observe interactions between HCPs, to further describe cultural contexts. For example, focused observations were made about how HCPs interacted with each other and co-constructed meanings through conversations and group action. The less formal setting of these conversations allowed more natural interactions between HCPs to be observed, which in turn informed interpretations of HCP participant data. Document collection Document collection included retrieving blank copies of patient care documents (admission template forms, care maps, referral forms, etc.), written unit tobacco policies     33 directed to staff, tobacco-related signage, other documented tobacco polices related to the unit, and tobacco-related resource materials. Appendix C outlines the guidelines used for document collection. These documents helped contextualize actions and processes that occurred on the units during policy enactment, and made visible official organizational values and attitudes. Document contents were analyzed considering the intended audiences and how organizational and social structures influence the tone, language, and mechanisms for communication. Assistance in locating the documents was negotiated with clerical staff, HCPs, and team leaders at each of the sites, and included examining predominantly hard copy print materials on the unit that did not contain patient or employee-specific data. Interviews HCP interviews focused on personal and professional experiences, observations, perspectives, and impressions of the SFGP and commenced by asking HCPs to first ?Think about what a typical day on the unit looks like in relation to tobacco use and smoking, and then tell me about that day.? Patient interviews focused on their experiences and observations of tobacco use during their hospital stay and their impressions of the SFGP. Interviews commenced with an open-ended request: ?Tell me about how smoking or tobacco has been a part of your life so far.? To help ensure the voices of participants were heard, open-ended questions were used with additional probative questions to increase richness and depth of data (Milne & Oberle, 2005). These types of questions and prompts helped me garner what participants saw as important and were revealing of dominant cultural norms from various perspectives. Appendix D, the HCP focus group interview guide, and Appendix E, the individual patient interview guide, contained questions prompting an account of their experiences around handling tobacco, viewing tobacco on the unit, any noted changes on the unit following the SFGP implementation date, examples of conflict around tobacco, what help was offered patients around smoking, and what they would they tell a new staff member or patient about smoking and tobacco on the unit. Two other interviews conducted were considered key informant interviews and focused on the individual?s experiences and impressions of the SFGP, beginning with an open-ended question asking them to recount what they remember about how the SFGP came into being. Appendix F, the key informant interview guide, contained questions centred on their experiences around SFGP     34 implementation, changes observed/experienced, and unexpected outcomes recalled. A demographic questionnaire (Appendix G) was also collected. Field notes completed immediately following interviews provided a record about participant actions or behaviours observed during interviews. All interviews were conducted by me. This allowed the opportunity to increase the depth of the data by extending questioning on matters raised by others as additional participants were interviewed and permitted me to check analytical approaches and analyses. Research sample Recruitment procedures and consent Prior to data collection, HCPs were introduced to my study, and a broad invitation to participate was issued during one of their regular employee meetings. HCPs also became aware of my study and my desire to include them in interviews during my regular attendance for field and participant observations at the beginning of the data collection period. HCPs often entered into discussions with me about the study and provided voluntary written consent (Appendix H.1) to permit inclusion of these discussions in the study data. Additionally, a day or two ahead of my expected attendance on the unit the administrative support person on the unit or the unit supervisor reminded HCPs of my attendance and invitation to participate in individual interviews. HCPs participating in interviews provided advance written consent (Appendix H.2) and completed a demographic information form (Appendix I).  Key informants were those regarded as non HCPs who had direct knowledge and experiences related to the implementation of the SFGP. While I had hoped to include first-level managers, housekeeping staff working on the unit, and security staff, housekeeping and security personnel were targeted for recruitment. Housekeeping staff and security guards were sent invitations (Appendix J) by email or through their work supervisor and received consent of the supervisor to attend an interview during work hours without pay deduction. Follow-up contact was made by telephone or email. However, no participants were recruited through these means. Initial contact with patients to discuss the study was sought during hospitalization, in two ways. First, HCPs were provided with patient invitations (Appendix K) and asked to     35 provide these to voluntary patients who they believed were potential participants. Second, HCPs were asked by me to identify and approach eligible patients to obtain permission for me to discuss their potential participation. Most patients were recruited through HCPs? efforts to locate eligible patients and introduce them to me. Written consent was obtained prior to the interview (Appendix L). Participants completed a demographic information form (Appendix M). Sample Sampling of HCPs and patients was derived from those who responded to invitations to participate and included self-declared smokers and non-smokers. All participants were able to provide informed consent and spoke English. Additionally, patients who were certified (involuntary patients) under the Mental Health Act (R.S.B.C. 1996, c. 288) were not eligible to participate. Patient availability and status were obtained immediately prior to a proposed interview from either the patient?s doctor or primary nurse, to avoid schedule conflicts and to screen against any potential harm to the patient. All patient interviews were conducted in person. One patient interview was conducted in a private office at a community recreation facility for people with mental illness, two at a community mental health office, and the balance at the respective study sites. Patient interviews ranged from 20 minutes to one hour, with most lasting between 30 and 45 minutes. HCP interviews occurred primarily at the workplace, and the employer permitted HCPs time during their scheduled work day to attend an interview for up to one hour. With the exception of one telephone interview, individual interviews with HCPs were conducted in person and ranged between 45 minutes and one hour.  Convenience and snowball sampling were used to identify and invite key informants for participation in individual telephone or in-person interviews. Key informants included those individuals not involved in direct patient care but who worked on the sites or were part of the management team and decision makers involved in the SFGP development, adoption, or implementation at the sites.  Sample sizes were not predetermined and were based on decisions to ensure that participants reflected diversity, and that the data collected had depth and breadth. Data were analyzed contemporaneously with their collection, and sampling continued until data were     36 repeating and representative coverage of emergent themes were saturated (Sandelowski, 1995b). Sampling remained flexible and evolved over time as data were collected and analyzed (Milne & Oberle, 2005). One example of this flexibility occurred in expanding recruitment of patients at Site B to include patients who had been recently discharged. The Site B unit had only 10 inpatient beds, which limited the number of voluntary patients available to participate. Augmenting the sampling to include recently discharged patients and interviewing them with respect to their time in hospital added valuable patient perspectives from Site B.  Some difficulties were experienced in achieving participation from HCPs at Site A, which delayed the collection of interview data. The challenges resolved with prolonged contact with staff and supervisory staff and assurances that the data were not evaluative of individual staff members or staffing teams. With the reorganization changes that had occurred at Site A during data collection for the current study, establishing relationships with HCPs was key. All Site A staff were interviewed during their regular work time. Patients, HCPs, and key informants who agreed to participate in an individual interview or focus group interviews were offered a $20-cash honorarium acknowledging their contribution to the study. Consent was voluntary and could be withdrawn at any time. The healthcare professional sample Individual interviews were conducted with 19 HCPs, including team leaders from each site and a manager from one. Informal discussions were also collected in field note data. Demographic data were collected from all participants who agreed to an individual interview, and these details are presented in Table 1.      37 Table 1 ?Healthcare Professionals? Demographic Information    Site A Site B Individual interviews  8 11 Informal discussions  8 3 N (females) =  16 (12) 14 (9)     Current smokers  2 1 Former smokers 2 4 Non-smokers 4 6 N (females) = 8 (7) 11 (7)    Age range: 32?65 years Mean age (years) 47.6 49.3 Reported Years? Experience in Psychiatry (n=19) In current position ? 4 years 5 In current position > 4 years 14 Mean years  14.3 Currently employed full time 15    Stated Education of Nurses (n=14) Nursing diploma 5 Diploma in psychiatric nursing 7 Baccalaureate degree (1 in nursing) 2   Stated Education Non-nurse HCPs (n=4) Master?s degree or equivalent 2 Post-secondary diploma or certificate 2    Self-identified Race  White/Caucasian  16 White/East-Indian 1 Black 1 Other race (not specified) 1           38 The patient sample In-depth interviews were conducted with 20 patients. Table 2 provides demographic information collected from interview participants. At the time of their interview, patients? length of stay on the units ranged from 1 to 30 days (M = 10 days). Table 2 also provides some breakdown of an open-ended, self-reported reason for hospitalization. Depression was the most commonly identified underlying diagnosis among participants. The key informant sample Individual interviews were conducted with one staff member from each site who provided administrative and other support to the unit and had direct contact with patients but was not involved in direct patient care. Both of these participants were female, non-smokers.      39 Table 2 ? Patient Demographic Information    Site A Site B Current smokers  7 9 Non-smokers 3 1  N (females) = 10 (3) 10 (8)    Age range: 20?71 years   Mean age (years) 39.0 44.6  Stated Reason for Hospitalization** Depression 6 Medication change/issues 5 Mood disorder (not depression) 4 Schizophrenia 3 Suicide attempt 3 Anxiety disorder 2 Addiction 1 Not provided 1   Living Arrangements & Income Source Living independently 14 Living in family home 4 Not provided 2   Social assistance payments 13 Employment income 5 No income 1 Not provided 1  Stated Education Less than high school 5 High school 9 Some post-secondary 6  Self-identified Race** White/Caucasian 15 Aboriginal/First Nations  1 White/Aboriginal 2 Aboriginal/M?tis 1 7-race mix 1 Note: not all data fields were completed by participants ** allows for multiple answers      40 Data analysis Analysis of the data occurred contemporaneously with data collection. Interviews were digitally recorded and transcribed, and transcripts were checked against the original audio recording to ensure accuracy. Field notes and observation data were collected in notebooks and subsequently transcribed to electronic documents. Documents collected at the sites were scanned into electronic PDF files. Electronic documents were uploaded into NVivo? and my computer to aid in organization and retrieval of the data. Data stored electronically were tagged or identified with descriptive codes to maintain a marker of origin for subsequent analyses. Various methods of analyzing the data were applied to enhance the analysis and included analytical note-taking of systematic approaches applied to the data, thematic organization, concept mapping, memo writing of thematic impressions during the analysis, journaling of personal impressions and reflexive thinking, and writings of various iterations of the findings sections. I began my analysis recognizing that ethnographic methods formally start with analytic notes and memoranda, but are informally embodied in the ethnographer?s ideas and hunches (Hammersley & Atkinson, 1995).  Initial approaches to the data included reading each interview to develop an appreciation for the essential features without feeling the need to move forward analytically (Sandelowski, 1995a). Coding was also applied to help locate structures, routines, beliefs, values, and patterns of behaviour that would help reveal understandings of social structures. For example, early in the data analysis I line coded several transcripts to identify text that referred to cultural concepts identifying sub-categories regarding beliefs about smoking, addiction, mental health, psychiatry, and quitting. I also selected and coded large sections of transcripts, participant observation notes, and field notes with the research questions in mind, for example, ?cultural norms.? These processes helped fracture the data to aid pattern recognition. They also helped me recognize insider and outsider views of what was going on by reflecting upon what I saw as potentially significant or important to the participants. Also, reflective and interpretive memos were written on the first six HCP interviews; these helped establish pattern recognition and included notes on what people were doing and saying that would help uncover beliefs and social structures. These initial processes allowed me to begin formulating general impressions about what participants believed or observed had changed     41 since the SFGP was introduced and how they perceived the purpose of the SFGP. It also helped sensitize my interpretations of the data. Analysis also involved reviewing large sections of data while applying iterative processes of moving back and forth from distance to immersion in the data, looking at the individual case and a sense of the whole. As part of the analysis, ongoing comparisons between blocks of the data were deployed to draw out similarities and differences in perspectives. Notes were made in a notebook to help track emergent themes and collapse smaller themes into larger thematic categories using processes of comparison and contrast. Wolcott (2008) cautions against over-reliance on comparative analysis in the ethnographic method and, with this in mind, comparison was made between HCP accounts and patient accounts, both within and between sites, to help reveal distinct cultural norms, attitudes, and beliefs. This type of comparative analysis helped shape how findings were organized and presented in the findings chapters that follow. Thoughts about what was going on, storyline cues, and key points or important facts were noted in the border of hard copy transcripts, in a separate notebook, or on file cards as the analysis advanced. I returned to the data regularly to check thematic development and often did keyword searches across interview transcripts to locate discussions about beliefs, values, events, assumptions, and feelings (Seale & Ankiss, 2012; Spradley, 1980). For example, words like I feel or I felt were searched, and then larger portions of the text read around the word hits to help identify underlying beliefs or actions that created emotional responses in the participants. This and other keyword searches were not used as a form of a systematic content analysis, but as iterative processes that helped reveal tacit values or beliefs in deriving thematic findings and cultural understandings grounded in the data. Large sections of text around individual words were reviewed to ensure I had ascribed correct meaning based on the broader context, and this too formed part of the iterative processes of moving from detail in the data to the whole dataset, in a back and forth manner. Various forms of memo writing were used throughout to help develop and record ideas about what was happening as supported in the data and to permit reflexive thinking about the data and analysis. The research questions provided reference to guide the analysis, as did culturally oriented questions like: What are people doing? What are they trying to accomplish? How do     42 [they] talk about, act, characterize, and understand what is going on? What assumptions are being made? What symbols are in use and what is their meaning? (Emerson, Fretz, & Shaw, 1995; Spradley, 1979). I remained immersed in the data, reading and re-reading the data during all phases of the study. Cultural inferences reported in the findings were based on analyses of what people were saying and doing (how they were acting), and the artifacts they were using (Spradley, 1979) as revealed through and triangulated by observations, the documents collected, the informal discussions, and formal interviews. I remained aware of how I influenced the data and of possible alternative interpretations of the data to help advance analyses whilst acknowledging myself as a research instrument engaged in a research ?dialogue? with the participants (Anderson, Adey, & Bevan, 2010). Reflexivity helped maintain this awareness. Reflexivity refers to the self-conscious activity of me reflecting back on myself as co-constructing the data to raise my consciousness to the biases, values, and experiences that I bring to the study (Creswell, 2007). Reflexivity acknowledges that all ethnographic data are value-laden (Hammersley, 1992). Fieldwork was dialectical between researcher and participants, and I both affected and was affected by the phenomena under study (Anderson, 1991). Field notes clearly demarcated my own thoughts about what was going on within the observation data. This type of clear demarcation in the data set also offered me discrete points on which to be reflexive. To discover and explicate my own effects on the data, I engaged in ongoing reflexive journal notes and field notes to aid this process (Thorne, Kirkham, & MacDonald-Emes, 1997). Self-reflexive questions included asking myself why I selected to record various data during participant observations, how I felt during an interview or while conducting participant observations, and what that may indicate about what was going on between me as a researcher and the participant(s), or what might influence my own data-filtering processes.  My orientation to data collection and analyses was shaped by my own socio-historical location, including the values and beliefs that these locations have cultivated in me (Creswell, 2007; Hammersley & Atkinson, 1995). I was conscious of my own approaches insofar as I am largely grounded in philosophy of pragmatism and realism, and that I applied this philosophy to understandings of what others did and how they ascribed meaning to events and objects. This aligned with Mead?s (1934) pragmatic philosophic stance.     43 Understanding my philosophical stance or worldview helped me ask reflexive questions during analyses, including considering how personal experiences as a mental health nurse, an educator, a former lawyer, an ex-smoker, and an outsider influenced how and what I observed and how I interpreted the data. I also thought about the fact that I lived in Northern BC, so had some insider knowledge, and of my position as a researcher new to these particular hospital units. I examined the data to recognize insider and outsider views. Alternating consideration of these views would help reveal differences between my own cultures and the one under study to assist with translation. Furthermore, I contemplated how I influenced the social structures and human behaviours I had observed and recorded. This included thinking about how participants might be selecting, framing, and reframing details and language in my presence and in the presence of others. Another researcher approaching the data might bring a different background to produce findings of a different nature. My studies and experience with law and my background as a mental health nurse influenced what, as well as how I interpret the data and position the findings in terms of a commitment to equity and justice. Although I deployed various techniques for awareness, reflexivity, and reflection on insider and outsider stances in co-constructing the data and findings, some biases prevail. For example, I initially positioned Site B overall findings as ?a battle,? based on what I heard from participants, but as the depth of my analysis increased, I became attuned to my own biases amid moving toward more tempered positions informed by understanding the challenges HCPs and patients faced. I acknowledge that analysis combines perspectives of myself and participants and that ?truth? in the data is more likely to be discovered by looking at the wide field of data and interrogating my own inferences for their limits, assumptions, and possible alternate points of view (Hammersley & Atkinson, 1995). I also self-interrogated for what type of assumptions I might be making that were affecting analyses, including whether the analysis was developing along insider or outsider perspectives and whether or not personal experiences clouded a view of the data. I consciously attempted to step back from the data and consider alternate views or assumptions to test the logic and approach of the analysis.  Witnessing how people grouped themselves in shared thought, speech, practices, and meanings related to the SFGP is what Yanow (2000) called grouping into ?interpretative     44 communities? (p. 10). Pal (2006) defined these groupings as ?policy communities? (p. 380). I examined the data for the emergence of these interpretive communities and as potential units of analysis. Thinking about interpretative communities helped interpretations by focusing on the actions of individuals and groups within social structures to build understandings of why people were acting as they were or why they adopted certain positions in relation to each other. Reflecting upon interpretive communities during data collection and analysis allowed me to analyze how people may socially align with those who hold similar perspectives, goals, and interpretations. Concept mapping was used to periodically check and advance my analyses. This technique helped me map and distill thematic and cultural understandings. Initial ideas were also advanced beyond descriptive events to higher levels of abstraction conceptually advancing theory-based understandings of the data. Thematic categories were depicted in rough visual representations?a map?of thematic clusters (Jackson & Trochim, 2002). This allowed me to visualize and piece together and fit developing ideas and analyses clustering and connecting themes to ensure a cohesive whole grounded in available data. Index cards were also periodically used to compare and sort pile emerging thematic ideas. Concept analyses formed part of the informal and iterative processes to check developing themes as correctly labelled or identified, consistency in the data across multiple sources, and hierarchy in relation to other emerging thematic ideas. The concept map also allowed me to iteratively locate and arrange ideas about cultural norms. A collection of notes, memos, diagrams, and notations on documents, transcripts, and field notes provide a record of developing ideas leading to the findings outlined in the findings chapters that follow. Iterative processes included reading large sections of data, reflexivity, conceptual mapping, and constantly returning to the data to identify patterns and understanding of the whole to confirm or disaffirm hunches and developing ideas as the analysis proceeded. These iterative processes continued through writing of the findings and included applying refinement as the writing continued, increasing both context and level of analysis.     45 Enhancing rigour Trustworthiness provides a broad assessment of rigour in qualitative work (Lincoln & Guba, 1985; Onwuegbuzie & Leech, 2007; Speziale & Carpenter, 2011). In this section, methods I adopted to enhance rigour in the current study are discussed. Credibility was assisted through sampling participants who directly experienced SFGP implementation, and sampling continued until there was repetition in the data of thematic categories and a diversity of perceptions and experiences were collected and represented in the findings (Milne & Oberle, 2005). Interviews sought insider perspectives through open-ended and probative interview questions so that participants could tell and discuss their own stories (Milne & Oberle, 2005). Rich, thick descriptions were sought during the data collection process and are further developed and presented in the findings section of the current study. Two sites were selected to enhance diversity in the data that helped identify similarities and differences, perspectives important to ethnographic approaches (Wolcott, 2008).  Triangulation was deployed by using multiple methods of data collection (documents, fieldwork, interview, informal conversation, and observation) (Onwuegbuzie & Leech, 2007) and by collecting perspectives from patients, HCPs, and key informants across two sites. These perspectives permitted increased depth and diversity in the data and identification of the data or themes that prevailed. Interviews, informal conversation data, and participant observations were used to triangulate findings presented in subsequent chapters herein. Document data were discussed with interviewees and reviewed, and helped identify organizational structures and values in the findings. Prolonged engagement can be viewed as part of establishing credibility (Speziale & Carpenter, 2011). Sufficient time at the sites helped reduce any bias created by my presence in the social setting (Morse & Field, 1995) and helped establish and support authenticity and confirmability by allowing patterns in participant behaviour to be observed and interpreted. Data were collected over 16 months. Repeating patterns in the data helped me decide when to conclude participant observations (Fetterman, 2010). Prolonged engagement not only enhanced the trustworthiness of the data and findings, but also helped to reveal changes in the phenomena of interest over time. In the current study, in particular, data collection was     46 started then unexpectedly stalled, resulting in collection over a period ranging from one to nearly two years following implementation of the SFGP and offered longitudinal data to aid important findings.  Dependability and criticality were bolstered by providing rationale and justification for research decisions as part of ensuring the study?s overall integrity (Milne & Oberle, 2005). Reflexivity and my efforts to be reflexive were discussed above, but remain part of enhancing rigour by increasing the dependability of the data. Confirmability was assisted through maintaining an audit trail that helped illustrate the evidence and thought processes that led to the findings (Speziale & Carpenter, 2011). The focus of the audit trail was to maintain an organization of all data and a written record of all transitions of the data so that the findings could be traced to their development and are supported by the data. NVivo? software assisted in maintaining some of this audit trail and regular and periodic saved electronic backups created a substantial record that was supplemented by handwritten notes and iterations of work in progress. Audio recordings, transcripts, and source documents collected during data collection, plus a log of decisions related to the conduct of the research, are retained as part of the research record. Field notes were written, audio recorded, and transcribed to an electronic and typed format for storage and analysis. Observational field notes were expanded by me as soon as possible after notations were made, to enhance accuracy (Emerson et al., 1995). Data were maintained in a secure environment, and all data stored on computers were password protected. Data will be maintained in hard copy and electronic form until at least the year 2015, pursuant to relevant institutional ethics agreements. A transcriptionist used my notation choices to guide preparation of typed transcripts. A systematic means of including some of the conversational aspects of an interview was adopted to include notation of pauses, inaudible portions, and speech patterns, including ?hmms,? laughter, and other emotional content from the audio recording, which served to increase the richness of the data and improve the context of interpreting the interview (MacLean, Meyer, & Estable, 2004). Field notes helped document the setting in which interviews occurred, as well as observed actions and behaviours of the participants, to increase accuracy of understanding and interpretation of interview data.      47 Rigour was reinforced by returning to the data frequently and by asking others to read and challenge my findings as I was developing the final manuscript. This provided a form of peer review or peer debriefing (Onwuegbuzie & Leech, 2007) that helped reduce or reveal various types of potential bias and permitted adequate attention to elements of enhancing rigour. Ethics Ethics approval was obtained from the University of British Columbia, the University of Northern British Columbia, and Northern Health before data collection commenced. One amendment to ethics approval was made to include recently discharged patients at Site B to ensure that there was adequate patient representation. To protect the anonymity and confidentiality of participants, some potentially identifying details were excluded from the presentation of specific data and the findings. While detailed depictions of participants are often included in ethnographic studies, these details were purposefully omitted from the current study. Identifying details would risk identity disclosures because the research was conducted in small communities and because admissions to inpatient psychiatric units tend to be repeat patients who are well known to HCPs and members of their community. A word about format Participant quotes are linked to a participant type (i.e., HCP or patient), number, and the site from which the participant was recruited, and this is indicated in square brackets. For example the notation ?[HCP25/A]? provides information that specific excerpt(s) are drawn from a HCP, participant number 25 from Site A. Patient interview data are indicated with the prefix ?PT.? No direct quotations from key informants were cited in the study findings. Specific examples or references to data derived from field notes, documents, participant conversation, and participant observation (PO) are indicated as such in square brackets. Word choices offered in the data from participants or documents are enclosed within double quotation marks within the running text to allow the reader to identify direct quotes. Lengthy quotations from participants are indented and italicized to identify direct quotations. Within quotations, square bracketed text indicates text that has been altered from the original to     48 improve readability, and text within parentheses indicates accompanying action of the speaker. Long quotes from documents are presented in a single line framed box.     49 As we speak today, it is a constant fight. [HCP participant conversation]  The study findings are presented in Chapters 4 and 5. Chapter 4 focuses on healthcare professionals? (HCP) perspectives in addressing the research question, What are the smoke-free grounds policy (SFGP)-related experiences of HCPs in inpatient psychiatric units, and how do cultural norms and institutional structures shape these experiences? In order to contextualize the findings, background information is presented first to describe specificities about Sites A and B, and the SFGP and its implementation. Site A findings are described in the theme maintaining the status quo and the results pertaining to Site B are shared in the theme new practice challenges. Sub-themes are described within each section to support and illustrate the overarching major themes. Site A findings are presented first, followed by Site B findings, which also contain some comparisons and contrasts between the two sites.  The Northern British Columbia context of Sites A and B Long settled by Aboriginal peoples, Northern British Columbia (BC), with its large expanses of mountain ranges, pacific coastline, farmland, and boreal forests, is home to a diversity of people and communities. As in the past, today the people of Northern BC maintain connection to its natural geography. Historically the region was influenced by the European fur trade of the 1800s, the Klondike Gold Rush of 1897, the development of the Alaska Highway to support Second World War efforts, and the commerce associated with natural resource development. Geographically occupying the most northerly two-thirds of the province at about 600,000 square km, Northern BC is home to just under 300,000 people, about 6% of the population of BC (BC Stats, 2009b). Most of Northern BC is sparsely populated, with one-third of the population living in communities of less than 3,000 people (BC Stats, 2009b). Major settlements are located primarily along two major highways; the Yellowhead Highway, running east-west, and the Alaska Highway, running north-south. The largest communities remain chiefly dependent on harvesting natural resources for Chapter 4 -  FINDINGS: HEALTHCARE PROFESSIONALS? PERSPECTIVES     50 commercial use, with a few towns having strong agricultural sectors. Research Site A is within the largest municipality in Northern BC, which has about 71,000 inhabitants, and is the hub of transportation, commerce, and healthcare for Northern BC. It supports a more diversified economy than other northern communities and also includes a small tourism industry and several post-secondary education institutions. Research Site B is located in the fourth largest settlement in Northern BC, with 11,300 inhabitants, approximately 575 km west of Site A. The Site B region is heavily dependent on forestry, mining, and fishing.  The two areas of Northern BC involved in the current study share some similarities that are demographically distinct from the rest of BC. First, heavy economic dependence on male-dominated industries of forestry and mining means that males outnumber females. Median ages (37.3 years (Site A); 38.2 years (Site B)) are lower than the provincial median of 40.8, and fertility rates and child-to-adult ratios are higher than the provincial rates (Statistics Canada, 2007a; Statistics Canada, 2007b). Downturns in the forestry industry have resulted in declining population since the late 1990s (BC Stats, 2011). Northern BC lacks the ethnic diversity of the rest of BC, with only 4.2% of its population identified as a visible minority (BC at 24.8%), but it contains the largest proportion of Aboriginal people (27.5% versus 3.8% for BC) (BC Stats, 2009b). Aboriginal populations are known to have poorer health outcomes than other Canadians and higher smoking rates (Frohlich, Ross, & Richmond, 2006). Other social determinants of health indicate that those in Northern BC on average utilize income assistance more frequently (6.0% versus 4.5% for BC), and host fewer low income family units, although the average income is slightly lower than for the rest of BC (BC Stats, 2009b). On a ranking of median family incomes, five of the top10 municipalities in BC were located in Northern BC, and Northern BC municipalities occupied the top three positions (BC Stats, 2009a; BC Stats, 2009b; BC Stats, 2011). Northern BC has lower proportions of people with post-secondary education credentials, lower completion rates of high school, and lower school grades than the BC average. Serious crime and drug crime rates are some of the highest in BC, as are mortality rates attributed to suicide, homicide, and accidents. Northern BC has more deaths and hospital days attributable to alcohol and tobacco use than any other BC health authority region.      51 A vast geography and sparse population distribution mean most specialized health services are located in the Site A city, and travel to here from other Northern BC locations can be lengthy at any time, and especially difficult and dangerous in the winter. Site B is contained in a 40-bed regional hospital, where specialist services are limited. At Site B, a psychiatrist travels from Vancouver or Kelowna once a week to consult with local primary care physicians. Psychiatrists will periodically make fly-in rounds to remote communities in the region, and telephone consultation with non-resident specialists is often the means of follow-up after initial assessment for those living outside the city. Many people with severe and persistent mental illness (SPMI) migrate to northern cities containing the regional hospital in order to regularly access basic and specialized mental health care services. Mental health specialist services have limited availability, and access is often an issue for over two-thirds of those who live in Northern BC (BC Stats, 2009b; Hunter, 2006). Both Sites A and B psychiatric units respond to local and regional community needs to accommodate individuals requiring admission to hospital for psychiatric treatment. Site A can normally accommodate 20 patients within the main hospital, whereas Site B can accommodate up to 10 patients in a building adjoined to the main hospital. The sites, though different in many respects, implemented the same SFGP, the specificities of which are outlined in what follows. The smoke-free grounds policy  Policy documents can be examined for both their effect and cultural underpinnings, as policy represents an organizational choice of action that reflects the values of an organization (Pal, 2006). The SFGP is a policy of the Northern Health Authority (the jurisdictional organization for healthcare provision in Northern BC). The policy prohibits tobacco use on all of the Northern Health Authority?s owned or operated facilities or grounds, with a few exceptions. When the policy was developed, Mental Health and Addiction Services were specifically offered exception status and opportunity to develop individual implementation strategies. The formal SFGP document is appended (please see Appendix A). The SFGP document and supporting implementation documents speak to the health authority providing ?preventative and supportive measures to assist [persons] . . . in adapting to a totally smoke-free environment,? of assisting ?communities,? advertising its smoke-free status, and minimizing smoke exposure to its employees. The policy language reflects valuing the health     52 of individuals, commitment to being a good corporate citizen and community focused, being equitable, and being supportive of all persons. For the policy to be effective, implementation must garner compliance. The policy framed procedures for ?enforcing policy compliance? suggest they be delivered in a ?respectful, consistent, reasonable, and sensitive manner? (Appendix A). This re-emphasized organizational valuing of individuals, equity, caring, community membership, and courtesy. The SFGP implementation plan for psychiatric units relieved HCPs of almost all responsibility for managing or handling tobacco even though this was not specifically outlined in official policy documentation. This change followed new practices of confiscating tobacco on admission and the organization?s policy rationale to ?protect patients, clients, staff volunteers and the general public from the harmful effects of smoking? and assume ?responsibility for health promotion and disease/injury prevention . . . and practice that supports the achievement of high health standards? (see Appendix A). Despite the policy goal to reduce the presence of tobacco on inpatient psychiatric units and achieve ?a tobacco-free environment? under this strategy, tobacco remained central to much of the activity at both sites. Sites A and B developed divergent implementation approaches reflected by differences in site-specific tobacco management. Policy implementation  One day you could smoke from 7a.m. to 11 p.m.; the next day you could not. [PO field notes/Site B]  They were called ?stay alive? days, marking the specific dates that the health authority rolled out the implementation of a 100% smoke-free policy on its grounds across Northern BC. The effect was visible and somewhat dramatic, moving all smoking off hospital grounds to the periphery of the property. Now everyone could see the smokers. Newly installed ?smoking prohibited? signs were in place, and posters adorned prominent locations. Letters telling of the upcoming policy implementation date had gone out to neighbours adjacent to the hospital grounds, municipal councils, regional districts, Aboriginal and other community organizations, contractors, Members of the Legislative Assembly, and home care clients. The health authority as a responsible community member spread the news, setting the example for health improvement, while also attending to     53 necessary public relations and media interest. A series of presentations had been made to staff and workers at the hospital sites. The message was clear and the preparation completed. Declining the opportunity to postpone implementation, the psychiatry Sites A and B in the current study selected May 1 and April 2, 2008, respectively, as their ?stay alive? dates, with Site B intentionally avoiding April 1 for fear of having the action viewed as an April Fool?s Day joke.  Sites A and B were to implement the same SFGP, and formal site leaders had agreed to adopt similar site-specific implementation strategies. Tobacco and tobacco-related items would be handed in or confiscated from patients on admission to the unit and held as contraband until the patient was discharged, at which time the items would be returned. In contrast to previous practices, tobacco and tobacco products were no longer handled by HCPs for patient use or consumption. Previously known ?smoke breaks? were renamed ?Fresh Air Breaks? (FABs), and patients were prohibited from talking about smoking activities during their stay.  For HCPs, the SFGP could easily emerge as a source of tension and conflict. Often characterized by HCPs as an ongoing battle or effort, understanding what was going on at each of the sites was contingent on uncovering cultural norms and the structures, beliefs, and attitudes that guided HCP practices. Site-specific findings are described next, beginning with Site A and four sub-themes comprising the major theme of maintaining the status quo. Site A ? Maintaining the status quo  Asked to consider what changed with the new policy, a HCP explained that she did not view the SFGP as having created a smoke-free environment but believed that it just moved the smokers to the sidewalk [HCP12/A]. This signalled that most pre-SFGP practices and routines were maintained and thus the status quo upheld. Maintaining the status quo comprised four sub-themes: 1) policy as the focus of resistance, 2) patient advocacy through knowing best, 3) a culture of ignoring, and 4) tobacco as essential to patient relations. These sub-themes all supported maintaining the status quo and are detailed in the next section, following a description of the unit itself.      54 Locating Site A A description of the physical aspects of the Site A psychiatric unit and its relationship to the hospital is provided to orient readers and contextualize the related findings. The unit was located in the older of two adjoined buildings comprising the 220-bed hospital. The hospital is centrally located and is the only one in the city. Connected by a central atrium, it merges a long, 1950s-built, five-storey building running north-south, with a newer, shorter building running parallel to the east. The hospital was marked by large, stainless steel signs near the roadway. When I first approached the hospital to begin my research, I noticed a small gathering of smokers near the main driveway. They were grouped around a large concrete garbage container with a built-in ashtray pan over the top. As I walked nearer the hospital, I noticed a group of patients smoking at the corner of the building, some sitting on a concrete prominence that forms a bench. I entered into a three-storey atrium with large, almost enormous, wooden beams that reminded me of the importance of the forestry industry to the community.   Hallways had a fabricated Venetian flooring of grey-beige hues. Walls were muted beige and overhead fluorescent lighting ran down the centre of the hallway ceiling, interspersed with white 2-by-4 ceiling tiles. As I walked toward the elevators on the ground floor, most of the corridors off the main hallway were absent of human traffic. Twin, aging elevators housed large black buttons with stainless steel framed doors and off-white Arborite panelling ? nothing modern here. The general feel was one of a clean and sturdy, but old institution.  On the third floor, the elevator doors opened and I arrived on the psychiatric unit to find myself enclosed in an unexpected confined space. When I exited the elevator, I felt watched. Directly across from the elevator was a wall covered with an array of assembled jigsaw puzzles that had been hung, unframed?a tapestry of cheap recreational labour posing as d?cor. There was also an assortment of ruggedly built, wood-framed couches and chairs and a public payphone area in this small foyer. As I moved to the left, I was dwarfed by a raised floor nursing station. This structure was no doubt constructed to increase the safety of staff from violent patients. The counters were high and wide to help prevent patients from jumping into the nursing station. Nurses in street clothes occupied and milled about in the     55 nursing station and implicitly questioned my arrival as I exited the elevator. The nursing staff appeared reserved, and I suspected that guests and new arrivals to the unit immediately felt uneasy for a variety of reasons, not the least of which was the ?us and them? division created by the distinctly separate and strategically positioned designated staff space?the nursing station.  Located at one end of the unit, the nursing station was a hub of activity where patients roaming the hallways could be seen by the HCPs perched within. Finished with a light birch wood exterior, it was sectioned into several distinct functional areas. There was a front reception area and a back office area behind an interior wall lined with glass. The back office could be locked off with the closing of a solid wood door. At one end of the nursing station was a glassed-in corner. Later I discovered that this section was where one HCP can sit and view the video monitor transmitting images from the Psychiatric Intensive Care Unit (PICU) contained within the larger unit. The PICU consisted of four seclusion rooms for the most ill patients. This portion of the nursing station contained a small battery of technological devices, including four flat-screen monitors for viewing the inside of the seclusion rooms, a public address system, an intercom system, and camera control devices. Most of the surveillance-related technologies on the unit appeared dated but in working order.  Adjacent to the glassed-in portion, and at the centre of the nursing station, was a wide, curved transaction counter about 8 feet in length, where patients and HCPs could interact. Behind the counter were two work stations signified by the presence of computer monitors. My observations reflected how the physical structures of the site contributed to the power differential between HCPs and patients. Patients interacting with HCPs at the nursing station stand below the HCPs. The patient, often with arms crossed on the transaction table to hold their chin off the counter emphasizes the position of the HCP speaking down as well as across a physical barrier that separates them from patients who approach them at the nursing station. [PO field notes July 13, 2009] The station was bounded at the other end by a low, built-in desk where the unit clerk sat and worked. This area was walled off from a large, adjacent patient common room by a wood panelled pony wall of approximately 4 feet, topped with two feet of glass paneling, adding to the overall glassed-in feel of the nursing station.      56 The dated and obsolete technology in the patient areas included non-functioning computers and a video cassette player. Unframed, glued puzzles on the walls, and the d?cor, suggested little patient activity occurred here, and perhaps reflected a focus on biomedical treatments and limited human resources for facilitating recreational activities. Boundaries marked divisions and imposed segregation such that spaces and facilities did not invite patients to be actively involved or interactive with each other or HCPs. For a patient, getting off the unit may be a welcome diversion, and participant observations indicated a certain excitement and buzz of activity when the hourly scheduled FABs approached. [PO field notes July 20, 2009]. The unit was a ?locked unit,? meaning that staff must release doors or call elevators to allow people to leave. When a door was unlocked or the elevator called, a loud tone sounded to draw attention to the fact that someone had unlocked an exit door. A HCP drew parallels between the unit and prison environments, and by implying and accepting similarities, indicated norms around the level of control HCPs had over patients. Amid describing the locked unit as prison-like, a HCP mentioned that nicotine replacement therapy is absent in prisons, but available here. This comparison between the unit and prisons suggested both similarities and differences. The similarities are in the locked space, control over inmates and patients, and in the power vested in HCPs and prison officials. The difference highlighted is nicotine replacement therapy availability, implying some increased level of humanity of care at the hospital versus the correctional facility. [PO field notes July 20, 2009] Generally, the unit seemed a sterile conversion of an older hospital ward that clustered activity and observation around a dominant nursing station. On the whole, the physical space and amenities of the unit supported an environment focused on HCP safety, monitoring of patient activity, and creating spaces for structured interactions between HCPs and patients. The space reinforced norms about roles, activities, power imbalances, and the unit as a place where people are managed rather than bolstered to their potential. The space reflected that HCPs led patients and that they, too, were organizationally accountable in the space. Implementation of the SFGP was intended to both build upon and integrate with existing structures influencing HCP and patient interactions between and within these groups.      57 Policy as the focus of resistance Implementation of the SFGP was met with resistance and defence as HCPs fought to maintain the status quo by keeping unit routines and practices around tobacco use the same as prior to the policy implementation. Recent staff cuts, demotions, and organizational and management restructures created vulnerabilities that may have fuelled HCPs to react in this way. HCPs were strongly bound by group norms and processes in adopting a unified stance to regain some control over their workplace. Within an environment where divisive worker?management ?us and them? elements had already emerged, the SFGP also provoked and exacerbated many HCPs? concerns. In building HCP group unity, peer pressure was exerted as a means of ensuring solidarity in resisting the SFGP-induced changes and potential conflict with patients they believed would flow from its implementation. Post-SFGP implementation activity around tobacco changed, in that HCPs handled cigarettes for patients only in exceptional circumstances. Prior to the policy change, HCPs held tobacco for patients and dispensed a single cigarette hourly for patient consumption off the unit, outside. While embracing fewer dispensing duties, HCPs were pessimistic about the efficacy of the SFGP and were resistant based on changes that might result in patient dissatisfaction and behaviour change. Participant observations revealed how patients and HCPs settled into a routine for hourly FABs, essentially consistent with pre-SFGP routines [PO field notes July 15, 2009]. As part of the SFGP implementation, short, off-unit breaks predominantly used to facilitate smoking opportunities for patients were renamed ?fresh air breaks? (FABs). A nurse, a non-smoker, explained: I was kind of curious to see how this was all going to play out. But . . . it doesn?t feel all that different, really, in so many ways; it doesn?t even feel like there is a policy. Because people are still going down for their hourly break . . . the only difference is we don?t hold onto the cigarettes anymore. People still say I?m going down for a cigarette break. [HCP12/A] This HCP confirmed that little had changed and, with a wry cynical tone, she communicated doubt that the policy had been or would be fully implemented. She revealed how HCP group norms mustered a sense of collective control over the policy implementation, in terms of both what staff did and did not do to ensure its efficiency, including resistance by retaining the label of cigarette break. The policy also challenged many HCPs? personal beliefs about the     58 role of tobacco in patients? lives, and the legitimacy of strict policy enforcement was contested. A HCP and smoker indicated how the high levels of ?vigilance? when the policy was first introduced were resisted and created resentment toward and dissatisfaction with management. She suggested that these tensions had eased with a less hard-line approach by management:  We weren?t allowed to say the word cigarette, we weren?t allowed to say the word smoke . . . we had a lot of restrictions, and so it seemed more difficult. Things have relaxed slightly . . . management maybe isn?t quite on us if we happen to say somebody?s gone for a cigarette break, like we?re supposed to say fresh air break. If we slip, we?re not . . . chastised . . . I think management has kind of softened a little bit. [HCP11/A]  Management included the HCPs? floor supervisor and more senior and executive managers, a delineation implicit to the aforementioned participant?s reference to her immediate work group as ?we? in separating the management and workers. In demarcating roles between HCPs and management, the potential for conflict or resistance was recognized amongst HCPs favouring the more relaxed approach to implementing the SFGP.   Many HCPs also framed the SFGP as belonging to someone else, handed down by management, who had little insight into the clinical practice challenges of implementing the policy. Problematizing the policy implementation permitted HCPs to resist the policy without targeting co-workers. HCPs were unforgiving, negatively judging the policy and its creators. That said, ultimately HCPs were aware that their performance as employees could be judged in relation to their compliance with the policy and management?s direction. They devised ways to circumvent management by acting in subterfuge to create some ambiguity about the specificities for how the policy was to be implemented while reserving the right to triage individual patient circumstances as trumping the policy parameters. A HCP, a non-smoker, spoke of how the HCPs typically stayed under the radar, hoping their passive defiance would go unnoticed: Our immediate supervisor . . . was unaware that we had adopted this ?don?t ask, don?t tell? policy that was the most common approach that worked for the staff. . . . For a period of at least six months, she was unaware that we were operating under the radar. [When she discovered this] she reacted quite surprised and quite frustrated . . . not because we were being insubordinate, but that from the way the world works for her, she?s a very policy-oriented     59 person. Both for our own protection and for a seamless psychiatric program in general . . . she didn?t like that inconsistency. [HCP42/A] In this example, though intended to unify and regulate institutional practices, the policy made visible divisiveness among staff and management subgroups. The supervisor?s managerial ideal to mandate front-line unity to mobilize the SFGP verbatim was quashed, and by subverting the policy, HCPs used their clinical practice and direct patient care to control the SFGP implementation. Belief in the power of acting as a group, commitment, and valuing their cause as just and right were important to these HCPs? norming processes, and group cohesiveness was a protective factor in deciding the extent to which they would support the policy implementation. The unit manager brought to my attention a management notice regarding the importance of policy to organizational unity. Annual employee performance reviews also required evaluation of how well individual employees implemented organizational policies, and she highlighted how the SFGP was an example against which employee actions would be assessed. This former smoker, manager had evaluated other out-of-province facilities? practices and experiences around SFGPs and believed staff divisiveness or defiance would jeopardize the policy implementation. A strong proponent of cohesive HCP teams, she recounted addressing staff in regards to SFGPs elsewhere: We have learned [from other?s] mistakes, so don?t do the same mistakes . . . they are wonderful because all the staff are together. The only thing that [made a difference between facilities] was that staff was divided and that would make or break policy implementation. . . . We were very successful at the beginning because all the staff worked together. [HCP37/A] Unity was claimed as a strong cultural value on the unit. However, this manager believed front-line HCPs were acting in unity to implement the policy in a manner consistent with her initial directions, when in fact they were subverting the policy and her direction. Staff unity garnered privacy of practice and a collective autonomy or will discordant with management?s direction.  Initial policy interpretations requiring patients to abstain from smoking while hospitalized and requiring tobacco to be seized and held as contraband were immediately resisted by HCPs. Few HCPs recognized the policy as health promoting and instead focused on requirements for reducing the visibility of cigarettes and the requirement to abandon     60 ?smoke break? terminology. Questioning the credibility and contesting the power of policy-makers was central to HCPs? resistance. A HCP, a former smoker, illustrated the perceived detachment of policy-makers from front-line workers: Well, the policy, I tell you what the problem is . . . for something to work, we all [need to] work as a unit. You cannot rely on the lowest people with the lowest area of power to do the difficult job while the top ones who are the ones with the power and authority ignore it [the policy] completely. [HCP37/A] The importance of group norms and behaviours is underscored by this HCP. She also revealed how her own actions were formulated as resistance to policy-makers? power and the marginalization that might invoke in her and/or her workgroup. She perceived policy-makers as uninterested in SFGP implementation and was resistant to implementing a policy that she or her workgroup had not agreed to in the first place. Policy-makers and other leaders were expected to lead and support difficult front-line work. Where leaders assumed HCP compliance and/or did not fully assess the policy implementation, they created opportunity for workers to resist the new policy and draw solace from the inertia of their pre-policy practices. Patient advocacy through knowing best Most HCPs adopted a patient advocacy position through knowing best. This supported the status quo and informed, in part, HCPs? resistance to the SFGP. Grounded in assumptions that the SFGP was about smoking cessation, HCPs preferred knowledge, acquired through their practice experiences, that patients were uninterested and/or unable to quit. This extended to patient advocacy in various ways. In speaking about how she first came to learn that the SFGP was to involve tobacco confiscation, a HCP who smoked detailed an individual rights-based argument in advocating for patients and resisting the policy: I remember asking, How are we going to do that? Like how? One of the big things was that we were told that when people come in, anybody came in, we had to take their cigarettes away. Then if they went on pass, say they went on a weekend pass, we weren?t to give it back. And that I had trouble with that because that?s someone?s personal belongings and what right do I have to keep that away. [HCP 11/A]     61  Although she was resistant to the confiscation provisions of the SFGP, she did not view pre-SFGP practices of holding all patients? cigarettes and dispensing them at the HCPs? discretion as problematic. This pointed to underlying beliefs about the importance of tobacco to patients. Longstanding viewpoints linking tobacco use to mental illness symptom control and coping mechanisms were understood as valid and legitimate knowledge. In illustrating the link to symptom control, a HCP, a non-smoker, highlighted differences between patients on the psychiatric unit and those in regular hospital units: My understanding is . . . nicotine itself does affect brain chemistry . . . with our mental health patients with their serotonin and whatnot and it actually does help to alleviate some of the symptoms. So . . . there?s a difference. [HCP12/A] Drawing on linkages between smoking and patient-centred mental illness management enabled some HCPs to advocate for a status quo around smoking, one that would accept that patients might need to self-medicate with nicotine and other substances drawn from smoking. A HCP, a non-smoker, summated that her past experiences led her to believe that patients do not want to quit, which in turn eventually led her to not ask them about quitting:  I used to [ask if they were interested in quitting], but none of them ever say that . . . and just at the time, they?re usually in such an acute situation that it?s kind of not one of our foremost thoughts . . . . I don?t hear others mentioning it either to them. . . . When I?ve asked them, you know, if they?re interested in cutting back on their cigarettes . . . they never are. The odd time they?ll say they want to cut back some but . . . then as the day progresses I see them keep coming up for inhalers, several of them and gum, so it doesn?t seem like they?re trying to really cut back, to me. [HCP10/A] Similar experiences were shared by other HCPs, whereby the differences between what some patients say and do in regards to smoking cessation supported the practice of not mentioning quitting, let alone advocating for tobacco reduction among this patient population. As illustrated in the previous quote, patients? use of nicotine replacement therapies (NRTs) was viewed as attempts to boost nicotine levels, and this provided HCPs with further indication that patients were uninterested in smoking cessation. Here, too, is illustrated the commonly held belief among HCPs that the hospitalization (the acute situation) was the wrong time to address smoking reduction as a priority. Determining care and treatment priorities for     62 patients became part of HCPs? professional practices of knowing best that led to patient advocacy in which resistance to tobacco reduction was located and the status quo promoted. HCPs? propensity to hold cigarettes for patients they viewed as vulnerable demonstrated paternalistic approaches of knowing best. Vulnerable patients were those assessed as being unable to manage their own tobacco, those who lacked judgment or cognitive ability to regulate their own tobacco use or the gifting of it, or who risked having their tobacco stolen by other patients. HCPs were concerned for patient welfare and suggested that the provisions of the new policy that prohibited HCPs from holding and dispensing tobacco to patients abolished means, options, and support for the most vulnerable patients. Paternalism was the norm for expressing caring and was illustrated by a HCP, a non-smoker: There?s still the odd patient who has to keep their cigarettes up at the front because maybe they?re too sick to be running around with a whole pack or you know, they?re not careful enough to look after the whole pack and they can get stolen out of their room and that sort of thing like that. [HCP10/A] Holding cigarettes for vulnerable patients was also part of avoiding the conflict and upset that comes from a patient who has unexpectedly or too quickly run out of cigarettes. HCPs positioned themselves to continue to help regulate consumption of tobacco for the most vulnerable through maintaining a practice of holding and distributing cigarettes, part of a pre-SFGP practice. But this view was not universally held and, for one HCP, a non-smoker, the best solution was to help patients problem-solve their behaviours and recognize the consequences of their actions, thus preserving and advocating for patient autonomy: I?ll say, Steve, I heard you?ve been handing out cigarettes . . . is this an issue for you? Just see where they?re coming from and then if they say, ?well, yeah, because now I don?t have any cigarettes? . . . I would say ?what do you think you could do differently next time?? Or I?d get them to figure it out. I wouldn?t necessarily tell them what to do. . . . Ultimately it?s their choice, if that?s what they want to do with their cigarettes . . . and then just support them if that?s the decision they make. . . . I?m kind of . . . trying to get them to figure it out really, instead of me implanting my own thoughts and ideas. [HCP12/A]       63 She went on to acknowledge her struggle between personal views valuing the autonomy to smoke and health, which she understood as compromising some aspects of well-being: To me, all this stuff is about control. . . . I do agree, people shouldn?t smoke, but I also believe in autonomy as well. Like people should have a say in what they choose to do. And sometimes people don?t always make the best decisions. But that?s not my decision to make. . . . I just feel things can become a real power struggle with people, like even on the hour kind of breaks. You know, someone wants to go, I mean I go along with it because it?s kind of a rule here and . . . I don?t want to make it difficult for someone . . . so I do follow the rules, but I?m not necessarily in agreement with them. I just feel like sometimes, it?s demeaning . . . the whole demographics: I?m up at this nurses? station looking down and you know. . . . I?m thinking to myself, okay, you?re feeling powerless to your mental illness, you?re struggling with that. Then you?re told when to eat, when to, like there?s all these rules that you have to abide by. . . . It?s probably my own internal kind of struggle. [HCP12/A] Group norms could influence individual HCPs? practices, without resolving such tensions. Participant observations confirmed that HCPs generally stayed behind the raised nursing station when talking to patients; however this HCP was observed walking outside of the nursing station to meet patients at the floor level, an action that ran counter to those norms, and that she used as a means to reduce HCP?patient power imbalances. Her focus on patients? rights to smoke is similar to the positions taken by big tobacco, which side-steps connections between illness and smoking by arguing that consumers have the right to choose (Katz, 2005). She and many HCPs favoured a human rights?based approach, effectively advocating for patients to have the choice to smoke rather than a policy that requires HCPs to confiscate tobacco and prohibit smoking. A culture of ignoring A culture of ignoring helped support resistance to the SFGP and sustain the status quo. ?Turning a blind eye,? ?what I don?t see, I don?t know,? ?don?t ask, don?t tell? were HCP responses when asked about the SFGP and current tobacco use by patients. Arguably, ignoring patients at times can have a therapeutic purpose in helping patients regulate their behaviours, impulses, and social skills. However, participant observations also revealed these practices as garnering other, perhaps unintended outcomes: It is 9:30 a.m. The nurses do not seem to be closely monitoring patient FABs. This group of nurses also seem rather uninterested in the regular activity of     64 patients occurring on the unit this morning. All nurses seem focused on their own routines. These routines require a large volume of solitary work at the computer along with some other forms of paperwork and charting. They often keep their heads down at their paperwork or eyes fixed on a computer screen, ignoring patients that come up to the nursing station or stand quietly, or softly speak to get the nurse?s attention. The delayed response times range from 10 -60 seconds, but few patients receive an immediate response from a nurse when they walk up to the nursing station. I notice, however, that staff routinely apologized for any wait patients experience in getting the elevator called after they sign out on the whiteboard, but an apology for having to wait to have a question answered at the nursing station is not offered. Why do they apologize for the wait to leave when someone is going to smoke (FAB), but not when they seek attention for other issues at the nurses? station? [PO field notes July 24, 2009]  Similarly, some HCPs ignored me and visitors entering the unit via the elevator. On arrival of a newcomer, HCPs behind the nursing station glanced the person?s way without acknowledgement before focusing elsewhere. HCPs? body language messaged to onlookers that they were unavailable, perhaps in hopes that someone else would respond. This highlights hierarchical structures that delineate boundaries and roles. Ignoring others conveyed messages of ?do not interrupt me,? ?I am busy,? and/or ?I do not see this as part of my job.? These observations also provide an example of where I needed to be reflexive to reduce the influence of my own biases upon the data and analysis. HCPs at times similarly ignored me, prompting me to reflect upon my insider and outsider status with HCPs. My initial reactions to being ignored were quite reactive ? which likely influenced my objectivity in data collection and analyses. However, with repeated exposure to the environment and these cultural norms I was able to interrogate more fully and objectively how the culture of ignoring was connected to the SFGP.  Work routines supported and contributed to a culture of ignoring around the SFGP. I observed the importance of routine to nursing staff and how they unofficially claimed work spaces in front of selected computers and segregated themselves and their duties from others working on the unit. Non-nursing HCPs were transient, routinely coming and going from the unit with little or no interaction with the nurses. On busy days, nurses withdrew and patients noticed and reduced their contact with HCPs [PO field notes July 13, and 24, 2009]. Most staff had worked together for years, so group norms around expected behaviours, like limited     65 interactions with others working on the unit and conceptualizations of team within front-line HCPs, were well established and supported the status quo. ?Don?t ask, don?t tell? approaches to tobacco regulation reduced patient?HCP conflicts around SFGP compliance and also helped HCPs reconcile individual actions with values and beliefs around tobacco use by psychiatric patients. For example, the site-specific rule under the SFGP was to confiscate tobacco and tobacco-related items on patient admission and when otherwise visible on the unit; however, post-admission confiscation was not routine. Ignoring became a practice of avoiding this aspect of policy implementation, as a HCP explained:  I don?t know who smokes now that we have ?fresh air? breaks. I do not know if they smoke and I do not want to know . . . what they do on their break is not my business. . . . If I see them with cigarettes I have to take them away because it is the new policy. I have a conflict of interest if I ask them if they smoke, so I don?t want to know.? [HCP participant conversation July 15, 2009/A].  Ignoring was prevalent and during participant observations; when I asked HCPs how many patients on the unit were smokers, they were always uncertain and unable to estimate how many patients smoked. At best, these inquiries yielded rough estimates, with HCPs claiming a lack of knowledge as an artefact of not handing out cigarettes. For example, one HCP claimed it is ?difficult for me to know,? and made disclaimers such as ?I only know my own patients? [PO field notes July 13?26]. HCPs seemed resistant to completing the tobacco assessment admission information for new patients, and their lack of knowledge about who smoked further illustrated their allegiance to a ?don?t ask, don?t tell? approach. Assessing tobacco use and initiating treatment for tobacco withdrawal was also positioned by some HCPs as physicians? work. A HCP, a non-smoker, asserted: I?ll be honest, I know there?s that assessment page, that we?re supposed to ask people if they are smokers . . . how much do you smoke and blah, blah, blah and then the doctor can look at that and determine what they should be ordering for this patient. I don?t routinely do that . . . personally I didn?t feel like that was that crucial to be doing. I mean if they really needed nicotine replacement . . . it?s going to be ordered one way or another anyway. So that?s been a bit of a hindrance, or a bit of a pain in the butt. [HCP13/A]     66 In BC, a registered nurse may administer NRT without a physician?s order. The hospital?s nursing practice guidelines were that nurses complete the tobacco assessment form and physicians write NRT orders following a clinical guideline algorithm for dosing; that said, nurses could administer NRT without a physician?s order in cases of urgent need. At Site A, nurses did not normally complete the assessment form nor administer NRT without a physician?s order, thus ignoring their scope of practice in regards to NRT amid side-stepping any assessment of patient smoking or intervention toward reducing or quitting. A culture of ignoring also extended to HCPs? perceived segregation from the larger hospital community. HCPs regularly responded that they did not know what went on in the rest of the hospital, did not see what was happening on the hospital grounds, and could not speculate about how others in the hospital approached the SFGP. Some of these reports were incredulous or willful in isolating the unit from the rest of the wards and hospital organization. This position may have been derived from transference of stigma associated with mental illness. Through ignoring, HCPs resisted changes that came with the SFGP to support the status quo, and in their ignoring roles, HCPs also strengthened group bonds and SFGP resistance through collective action. Tobacco as essential to patient relations HCPs saw tobacco as important to establishing and maintaining therapeutic rapport with patients and to managing patient behaviours, especially unruly ones. This contributed to HCP resistance to the SFGP and desire to maintain the status quo. The power of tobacco addiction was instrumentally used by HCPs to influence behaviours in patients through a reward system that included discretionary granting of FABs. If a patient was a smoker, HCPs, by granting or withholding hourly FABs, could influence behaviours. The use of FABs emerged as a bargaining tool, and a HCP, a non-smoker, offered the following insights about this approach: I?m not in agreement with people smoking [and] recognize it?s unhealthy and . . . I?d prefer if they didn?t because in some it?s for their health in general. However, for some people, especially our chronic schizophrenics, I sort of feel like it?s the only thing they have to hold onto sometimes. They look forward to that cigarette break and so I find it gets incorporated into our behavioural contracts . . . it?s worded fresh air breaks, but really I know it?s a cigarette break. [HCP12/A]     67 This participant illustrated the continued importance of tobacco to console patients and aid behaviour change, implying she thoughtfully considered the social benefits and medical risks of cigarette use among patients with specific ailments. In the end she favoured tobacco use for patients with severe mental illnesses, and in doing so revealed a paucity of other resources to build patient rapport and reward toward influencing patient behaviours. The same HCP was nonetheless reflective about the contradictions accompanying her approach to sustaining relations with patients who smoked: I just get caught up in the everyday stuff and I just, it?s awful, maybe I don?t see it as a priority and that?s wrong of me because it is unhealthy for people, you know. And I mean, you know, just because someone has chronic schizophrenia doesn?t mean they have to have lung cancer. Maybe I should be more proactive regarding it. . . . It hasn?t been a real priority with me. [HCP12/A]  Evident in these and many HCP interviews was dissonance and tensions about the need to intervene regarding patients? smoking and a desire to manage patients? mental state. In addition, while HCPs? personal beliefs tended to determine priorities for patient care, essentially ?knowing best,? the idea of knowingly transgressing professional practices sat awkwardly with some participants.  Many HCPs suggested that controlling tobacco also forced them to adopt surveillance roles with patients. A HCP, a former smoker, when asked to reflect upon a typical day regarding tobacco and tobacco management, lamented policing and parent roles, which, she asserted, accompanied the new policy:  It can be frustrating. We feel like policemen. We have to have them lined up and they sign on the board and they go off every hour and then ?no, you?re not due for another 15 minutes.? And then they?re hounding us, you know, and you feel like a policeman many a time. . . a cranky one some days. And you basically give them a privilege for?it?s almost like for good behaviour or like treating children for like, you know, good behaviour, really. You get this privilege if you perform this way. Which I suppose is wrong, but that?s the way it goes here. [HCP14/A] Frustrated, this HCP weighed personal and/or professional beliefs in ultimately accepting the existing group norms. The influence of these norms is made known through her continued reference to ?we? and complicity in giving way to the ?will of the dominant group.? Though     68 critical of the policy, this HCP implied the status quo of HCPs? practices around patients? tobacco compromised her ideal nursing practice. After all, tobacco-related health concerns about the patient, while giving way to controlling patient behaviours, raised questions about who benefited most from patients continuing to smoke?the patient or the HCP. The SFGP brought to the forefront longstanding tobacco practices and the use of tobacco to maintain patient relations and, in doing so, revealed some tensions in maintaining the status quo. That said, group dynamics supported resistance to the SFGP implementation, but some HCPs began to question their own practices in light of the policy. Yet, overall, these tensions did not give rise to the wholesale changes that would accompany fully implementing the SFGP. Site B ? New practice challenges  HCPs at Site B consistently referred to practice challenges when asked about their experiences with implementing the SFGP, and this often led to making practice choices about how and what to enforce under the policy. An array of policy-induced tensions occurred between the patients and HCPs, within the HCP team, and between the unit and staff members in other areas of the hospital. There were high expectations of compliance with the SFGP, and this often meant HCPs needed to regulate their practice to ensure they supported, and, where necessary, implemented the specificities of the policy. Amid competing work demands and fatigue it was evident that HCPs were focused on avoiding conflict by thoughtfully choosing their actions. Under the overarching theme of new practice challenges, which referred to HCPs? desire to resolve practice challenges arising from implementing the SFGP, four sub-themes are used to detail the results: 1) strong, consultative leadership, 2) healthcare professionals? discretion in practice, 3) ambushed?the challenges of broad compliance, and 4) scarcity of resources. As in the previous section, following is a description of Site B, offered as a means to locate the findings detailed thereafter.  Locating Site B Located in a quiet, residential area of the city, surrounded by forests on two sides, the 40-bed, 1959-built hospital had an air of nostalgia about it. As I approached the hospital doors, it was as if I had gone back in time to a quieter and perhaps simpler place. However, the corporate branding emblazoned on the ?new? hospital sign above the original name plate     69 signaled a blurring of old and new. Entering through the front doors into a small waiting area and hospital auxiliary gift shop, time stood still again. I walked toward the elevator but then decided to take the stairs down a dimly lit, deserted, concrete stairwell before emerging in a corridor flanked on one side by an adjoining hallway to the hospital food services. The hallway was conspicuously empty, with closed doors and a few bulletin boards with some nutrition information about cereal products. I continued to walk, past a small, barren cafeteria with its empty food counters, and noticed a few kitchen staff drinking coffee. I turned left at the end of the hallway and was met with two closed entrance doors to the psychiatric unit.  The psychiatric unit was a locked unit, allowing visitors to enter the unit without formal permission, but requiring staff assistance to unlock the exit doors to leave. The double-door entrance had one door reserved for entering, the other for exiting, and each housed a small window of wire-enmeshed glass at eye level. The windows were covered by signs allowing only small slits for peering into or out of the unit. The windows were no doubt intended to prevent entry/exit collisions; however, the strategic location of the signs was likely intended to provide privacy for individuals on the inside. The doors had various bold-lettered signs that provided visitors with a variety of information and imperatives: all visitors must report to the nursing station, please do not bring children onto unit, video surveillance is in use in this area, please make sure door closes behind you. An octagonal sign gave warning that violence and foul language would not be tolerated and were subject to prosecution; a no-smoking sign and a posting detailing the SFGP were also prominently displayed at eye level. Many of the signs had a small ?thank you? in italicized font in the lower left hand corner, as if to apologize for the tone and confirm that caring and compassionate people worked there.  The door?s push-bar opening mechanism had most of its chrome-like electroplating rubbed off, leaving the underlying brass-coloured metal exposed. Many had passed through these doors. The doors themselves were metal, painted turquoise to match the frame and perhaps intended to lift the bland tones of the hospital?s colour scheme. Upon entry, visitors faced a long, empty hallway that extended the length of the unit, with a turn to the right about 10 feet in, and another 30?40 feet ahead, at the end of the hall. Initially, the unit appeared cold and empty amid a maze of unfamiliar hallways. The ward was separate from the hospital     70 and appeared sterile. The dull physical spaces conveyed a cold, segregated space in which patients with mental illness were housed.  The entire unit lay to the right of the entrance hallway, except for a recreation area on the left. After entering the unit, I turned right and walked alongside an open-concept patient common area that included a kitchen, dining area, and television area, where chairs and couches had been lined up to separate it from the larger space, creating a common meeting area with an assortment of comfortable chairs, coffee tables, and a few plants. Another hallway to the left was flanked by the glass-enclosed nursing station on one side. The pattern of the unit then became more discernible to me. It was two large squares of hallways with bedrooms along the south and common-use rooms along the north. The nursing station, seclusion rooms, and other utility rooms occupied the centre squares of the unit.  The glassed-in nursing station clearly separated its HCP occupants from the rest of the unit. A transaction counter ran the full length of its longest side, with glass windows extending from the countertop level to the ceiling on three sides. One area of the countertop had an opening, with a sliding glass window roughly 4 feet wide, and doors with glass windows set at either end of the room. It looked as if staff could batten down this room if need be. The door closest to a sliding glass window was ajar. Desktop computer screens and keyboards marked staff workstations and were located about 5 feet away from the open window. Talking to staff inside the nursing station while remaining outside positioned at the window felt awkward. It required stretching over the countertop surface and turning my head to the left to see the person seated at the computer and to avoid having my voice reverberate back at me off the glass surfaces. Standing at the open door seemed to remedy this, but I needed to speak up to ensure that my part of the conversation reached the person seated at a computer workstation ? who was 10 to 16 feet away. While raising or projecting the voice made it easier to communicate with staff seated within, the distance, glass barrier, and enclosed spaces all clearly signaled an ?us and them? division.  Staff tended to dress in street clothes, so making distinctions between patients, visitors, and staff members could be challenging. Some staff wore identification badges, and visitors such as me, were quickly and politely greeted when approaching the nursing station, a practice that created, at least for me, a sense of welcome.      71 Strong consultative leadership  Strong consultative leadership from those in management positions at or near the unit level was a distinguishing feature at Site B; here, formal leadership governed and ultimately subtly overrode any HCP resistance. Though some resistance to the SFGP was evident, related debate between HCPs was open, a practice encouraged and facilitated by the managers. Managers stated that they encouraged discussion and debate as opportunities to model responses to those who voiced resistance to the policy. Accordingly, formal leaders led front-line HCPs to develop strategies to counter resistance in their own and others? practices. Managers clearly communicated their vision for ensuring policy implementation. One manager, who had been promoted through the nursing ranks at the hospital, was instrumental in developing the SFGP for the governing health authority. The HCP team members also respected and were loyal to this manager, so the policy was regarded as having been developed with the input of the unit, and ensuring implementation was part and parcel of HCPs? respect for the manager. In this regard, there was no question about whether the policy would be implemented, but rather the HCPs? debates tended to centre on how that might best be achieved. A HCP, a manager and former smoker, illustrated the early momentum around successful policy implementation: There was a honeymoon period . . . when there was so much attention and many, many kudos for the unit . . . the manager[at the time] ensured that everybody was aware of how important a role they had played in the implementation and really shone a nice light on everyone. That was all really nice for the start-up and I should?ve planned to keep it going. [The question is] how to deal with things that are, inevitably, no longer interesting? [HCP30/B] Also illustrated here was how formal leaders lobbied to own the processes associated with implementing the SFGP and encouraged co-ownership with front-line HCPs. A manager said of the implementation successes garnered through obtaining HCP support backed by strong team leadership: Mental Health Services really led the way at the [Site B] hospital in adopting the policy and today are leaders in the policy, whereas other areas of the hospital still have issues with tobacco use. [Participant conversations, August 5, 2009]      72 To further gain HCP support for the SFGP, it was presented as a professional practice imperative of ?respecting patients? by promoting care equal to that provided to others in addressing the whole of an individual?s healthcare needs, including smoking. A former smoker, HCP manager, and strong supporter of the SFGP spoke of the subtleties and related challenges:  One of the challenging things for me was how to deliver that message to staff around . . . [seeing or knowing] somebody is going out to smoke while they?re on a pass, but not seeing tobacco, so there?s no breaches on the unit and . . . that?s okay, but not to go any further. . . . We had to pull back a few times to stop it becoming a nudge, nudge, wink, wink, because then you actually lose the policy. . . . At the end of the day our goal wasn?t to get them to quit smoking, it was to remove tobacco and also not to expose staff to having to deal with tobacco. [HCP40/B] This participant highlighted how management shaped HCPs? understandings about the policy goals. He made an important distinction about minimizing harm for all, but did not focus on the health of the patients who smoked. Another HCP manager and former smoker was asked about turning a blind eye to smoking as long as it was off the unit. Drawing on past personal experiences and dissatisfaction of indifferent leadership and governance, she countered the suggestion, quipping:  That?s saying I don?t care what you really do around this. . . . As you know, we?re not forcing them to quit, but when it comes to messaging this policy and how it?s incorporated in treatment. We?re saying it?s important that you?re aware of this whole thing . . . how it affects your physical health, your mental health, what it does to you financially. And also what it does to your metabolism in terms of the medication that you take. . . . We have to make it part [of care] and not turn around and pretend it?s not there. . . . If we say we don?t care, we?re kind of signalling, I don?t know. . . . I think that?s indicative of an indifference and I?ve never liked that. I?d rather have the full-on arguments about it than indifference. [HCP 30/B] This manager?s reflexivity laid out her beliefs about HCPs? responsibility to implement the policy as a means of embodying their professional and moral obligations to provide care to, and protect patients. This positioning also aligned with healthcare treatment goals and values of providing psychiatric care, a stance that fuelled her desire to enact the policy. Ultimately,     73 however, her approach to the policy was motivated by personal beliefs and experiences, and perhaps management obligations toward policy implementation.  A manager from Site B, a former smoker, explained the importance of leadership in guiding the cultural norms necessary for effective SFGP enforcement and nuance to garner effective policy enactment: The subtle nuances was probably the hardest thing because a lot of psych staff can work that way but if somebody really needs a policy to follow then it gets harder . . . [whether HCPs should seek out tobacco from someone who is known to smoke, but the tobacco has not been seen] . . . is where I?m conflicted [because I already hear] ?I?m not the smoking police.? And they?re not the smoking police . . . [but] it?s ultimately about reducing the tobacco on the unit and reducing the impact on patients and staff. So you remove the tobacco when people arrive and really set that standard that you know you?re not allowed to have any smoking materials. If people are then able to control their smoking and . . . you can never put that down in a policy, ?that I?m not going to search you if you behave sensibly.? And [guidance] has to come from the team leader. [HCP40/B] The strength and willingness to consult with the formal leadership team on implementation of the SFGP potentiated the possibility that it would be successfully implemented. Nevertheless, there were implementation challenges. A HCP, who smoked, in speaking of other HCPs at the hospital, explained: It?s us and them [other hospital HCPs]. They?re the good guys because they are letting patients go smoke and we?re these rotten horrible people that are stopping these [patients] from doing this . . . down here [people get frustrated and start questioning the policy] . . . and then emails go around . . . about ?haven?t you done this?? and it?s almost like everyone?s pointing fingers. [HCP24/B] Though the psychiatry unit managers sought to depart from community norms around tobacco use and disrupt cultures accepting of smoking among those with mental illness, the lack of unity in supporting the SFGP policy within the hospital began to infiltrate, challenging the unity they had initially enjoyed on the unit. A HCP, a former smoker, further highlighted the difficulties: When you have such a discrepancy in enforcement and buy-in and interpretation of [the SFGP], you have a divided unit on some very important     74 philosophies around patient care and how we message this care to the patients. [HCP30/B] Lack of management consensus shifted the responsibility for decision-making about patient smoking to HCPs. Managers recognized that their team needed affirmation of their efforts to date, as well as some bolstering to lobby them to sustain their commitment to the SFGP. The managers and HCPs understood the negative impact of outside forces and high work demands, and a memo from one of the managers to the HCPs, while acknowledging the challenges, appealed to the HCPs? collective to stay consistent in working to implement the policy: I would urge you all to be consistent regardless of individual preferences relating to the policy. . . . I am proud of the efforts and success of the policy on this unit. This has only been made possible by the professionalism and skill of this team. You should all take pride in proving that SFGP can be implemented in psychiatric inpatient settings. [Document: email memo from manager to all HCP staff at Site B, November 5, 2008] The memo was printed and posted on the staff room bulletin board and highlighted the team?s successes.  When resistance surfaced among some HCPs, the managerial response was to deliver more education and promote open dialogue. A HCP, a former smoker, confirmed the importance of education: It?s really all about education and preparation. [We] really paid attention to this. All staff [at Site B] had to do the ?Changing Minds? PowerPoint1 presentation that originated in Nova Scotia, then got modified for Alberta, then Fraser Health, then rebranded for [here]. [Participant conversation August 5, 2009, HCP30/B] In addition to the in-hospital education, most HCPs had undertaken extra education around tobacco use, interventions, and support measures to aid reduction and/or cessation. Education was valued for its perceived ability to displace personal beliefs around tobacco use that were                                                  1 The Changing Minds PowerPoint is a presentation aimed at HCPs working in psychiatric settings that provides education about the importance of addressing tobacco use in mental health populations, encourages framing tobacco use as an addiction to be treated like other addictions, and dispels commonly held beliefs and resistance about implementing SFGPs in psychiatric settings based on experiences reported in the literature at other facilities.     75 contrary to SFGP objectives and lobby for the use of targeted smoking reduction and cessation interventions. These philosophical underpinnings are in contrast to Site A, where only one HCP reported having recently completed a tobacco education program.  While education can change practice, it was also clear that personal beliefs fuelled inner tensions about human rights and the scope of one?s professional practices. For a HCP who smoked, the SFGP was difficult to reconcile: I think there?s a lot of real frustrated staff. . . . I had to change my mind because I really don?t believe in a non-smoking policy per se, even though I agree with it, only because you know, smoking is not illegal. But it?s like so many things, like the patients on the ward here can?t . . . have caffeinated beverages on the ward. That?s the other thing we?re policing, so now not only are we policing the cigarettes, we police coffee, pop, cocoa, tea, anything family brings in to them. [HCP24/B] Managers were expected to address the compliance issues with the rest of the hospital as a means to both affirm and support the efforts of the HCPs working in the psychiatry unit. The battle that ensued between management and other HCPs led a HCP, a former smoker, to suggest: Administrators are tired and they are starting . . . to pick their battles, too. I think that smoking is fairly low on their agenda . . . they?re bogged down into things they consider much more important . . . and [these meetings with other department heads] are fairly meaningless [when it comes to advancing the enforcement of the SFGP]. [HCP30/B]  Evident in this HCP?s account is an understanding that individual hospital managers were divided in their commitment levels to the SFGP, and their direct managers? fatigue suggests that the efforts to implement the policy had begun to wane. In line with this, the HCPs on the psychiatry unit struggled to sustain their implementation efforts. That smoking was not a high priority throughout the hospital likely also reflected broader community acceptance of tobacco. Community standards viewed smoking as a comparatively lower health risk than standards about access to care concerns and primary healthcare challenges. This was revealed in management?s priorities about other unspecified issues at the hospital and within the region.      76 Healthcare professionals? discretion in practice There are rules and there are rules. To understand nuance, one had to appreciate the subtleties of meaning within the policy and determine the personal judgments and interpretations that underlay the implementation of the SFGP. The policy was enacted within an array of cultural ideals and practices that shaped the inpatient psychiatric unit. Central to this were HCPs? discretion and choices about their practices around this and how best to prioritize those practices.  The policy required HCPs to confiscate tobacco and tobacco products at time of patient admission and whenever visible. Tobacco was generally confiscated at admission, but confiscation at other times during a patient?s stay was less uniformly enforced by HCPs. Confiscation efforts provided the most dramatic examples of how diverse HCPs? practices could subterfuge policy. Too much force or HCP aggression in confiscating tobacco was understood as mustering an assortment of patient tactics for hiding tobacco. So, while HCPs understood confiscation as an essential control element of the policy, they sought a balance between search and seizure measures and turning a blind eye to some patients possessing tobacco and smoking off the unit. Tobacco that was not visible to HCPs on the unit was tolerated, but when tobacco was visible it was confiscated. Wider tobacco searches and confiscation measures were also deployed when patient rule-breaking escalated. Discretion in confiscating tobacco also complied with tobacco reduction, rather than cessation, which was central to the SFGP.  Finding a balance between strict compliance to create a tobacco-free environment and deploying strong measures to locate patients? hidden tobacco demanded assessment of a number of factors. Judgment and discretion rested with individual HCPs, and SFGP practice-related decisions were influenced by diverse understandings of the policy and beliefs about smoking and management guidance. A HCP, a non-smoker, said of the need for discretion: This particular unit is not very stringent with room checks. Sometimes when you go in someone?s room you will see it [tobacco] just laying out and . . . it?s up to individuals nurses? discretion whether it?s taken away. The policy reads that it will be confiscated . . .  some nurses are more lenient than others and will just say ?listen, if you?re going to have that, you have to keep it concealed a bit more.? I guess it is hypocritical in some ways, but I think it?s also heavy-handed to enforce the policy in others.[HCP32/B]     77 For this HCP, the policy afforded HCP discretion to contextualize the implementation. This discretion presided as a practice norm, whereby the level of enforcement was known to vary across HCPs.  HCPs determined their own tolerance for diverse practices. Some staff divisions around acceptable levels of policy implementation were evident whereby occupational health and safety concerns were made. A former smoker and HCP explained: I?ll police it to a point, but I won?t jeopardize my relationship. I won?t jeopardize any of the situations or my safety within it, too. I will turn a blind eye to stuff if it?s going to mean that I?m not being abused, verbally or so on. I protect myself as well, if it?s going to protect other people. [HCP 25/B] Another HCP confirmed:  I?m not going to put myself at risk of getting punched, so if they really need a cigarette, I?m not going to stop them. [Participant conversations October 19, 2010] Ensuring personal safety was a cultural norm among the HCPs, and while policy-makers had anticipated potential violence in enforcing the SFGP, respect and politeness (strong Canadian cultural norms) could also inform lapses in implementing the policy. The following passages from employee training and informational documents provided an example of anticipated violence and policy-makers? expectations that cultural norms guide employees:      78   As a Canadian cultural norm, politeness was an organizational value that prevailed in Northern BC. Advocating ?good citizenship? was code for valuing community membership, and this is particularly important to Northern BC. Well-mannered approaches were believed to help avoid violence and conflict when one individual sought to confront another about the policy. Advocating for all also reflects policy-makers? commitment to providing employees with a safe workplace. Health authorities in Canada are generally expected to function from a caring perspective as part of their professional practice codes. HCPs indicated that difficulty arose with acquired brain injury patients because of the cognitive impairments that limited their ability to comprehend and abide by some of the policy ambiguities. These patients were also thought to be the most likely to react violently to strong enforcement measures as a result of their misunderstandings of the policy [Participant conversations: October 19, 2010, and also HCP interviews]. Locating when and how to exercise HCP discretion around the SFGP required insider knowledge unique to the unit.  When approaching someone who is smoking on the property, please use an approach of ?good citizenship? using polite reminders and education about the policy. There is no expectation that staff will put themselves at risk in any way when doing this.? [from Compliance Protocol revised November 2007] ?If you are wondering how to approach a person to talk about [observing them smoking on hospital property], here are some suggestions: 1. Be polite and non-confrontational . . . Do not raise your voice and do not sound angry. . .  2. Be respectful and non-judgmental. . .  Do not get involved in further discussion if the person does not respond politely. You should remain polite and be considerate if you want your message to be heard. [from Talking Tips document revised November 2007, emphasis included]     79 Ambushed?The challenges of broad compliance Participants described unanticipated events and experiences related to unmet expectations around the SFGP-related behaviours of other HCPs that led me to characterize their experiences as ?ambush.? Examples included diversity in implementing the SFGP between Site B HCPs and other employees at the hospital, HCP challenges with physician orders, challenges arising from patient transfers from other hospitals, and reactions of community members to the SFGP.  Site B HCPs noticed inconsistencies in policy uptake. The hospital had a strong culture around accepting tobacco, whereby smoking among staff and patients was a longstanding norm. After nearly 18 months of strict compliance, the morale of HCPs working at Site B was wavering, at least in part, due to their understandings that employees in other areas of the hospital were not enforcing the SFGP. The tensions that emerged from this discordance grew over time, and continuing challenges with SFGP implementation were internalized by some and gave rise to self-doubt about the appropriateness of the policy and implementation efforts.  HCPs complained about how smoking went virtually unchecked around the hospital, but of a willingness of hospital staff to report psychiatric patients? violations of the SFGP while ignoring other smoking groups. Lack of security personnel was a factor that left smoking on the hospital grounds unchecked; however, reporting violation on the part of psychiatric inpatients reflected norms around managing that patient subgroup. These inconsistencies in the application of the SFGP increased conflict between staff working on Site B and those working on other hospital units. A HCP who was a former smoker said of these inconsistencies: With this patient that came down yesterday . . . this patient was getting out regularly and now we have him and now he can?t get out. . . . The Kardex [stated]: ?allows liberal smoking passes +++[many times]as per doctor?s orders as long as he stays in his pyjamas and has his IV pole and the next day, has had numerous smoking breaks +++ [many times]with his IV pole, [he was]compliant to same.? That was on the 17th and we get him on the 18th. So is that setting us up for a huge situation? I mean potentially it creates a Code White [security alert for violence] if it really comes down to it, when you have somebody who is absolutely reactive. . . . I don?t know what some people     80 think. They have them sign the smoking policy upstairs, but it?s okay, you can smoke up there, it?s just down here [that you can?t]. [HCP26/B] That staff outside of the psychiatric unit acted upon psychiatric patients? breaches of the SFGP while ignoring the breaches of others was problematic for Site B HCPs. The actions of other staff members may signal defensive posturing to deflect attention from their own failures to more broadly implement the SFGP. The dynamic suggested that HCPs on the psychiatric unit were doing their jobs in regard to the SFGP and were ambushed by those who were not.  Nurses were challenged in receiving new patients from other areas of the hospital with physician orders directing that patients be given ?smoking passes.? Physicians outside of the psychiatric ward routinely ordered patient smoking passes, but lacked understanding of the SFGP and/or were unaware that off-unit passes were not transferrable to the psychiatric ward. Practice hierarchies made it difficult for nurses to counter or explicitly contest physician orders. Obtaining amended orders also increased HCPs? work, and patients could contest changes, arguing that their doctor had granted them smoking privileges, a condition that underpinned some patients agreeing to being admitted. Indeed, patients challenged HCPs who reneged on the previous order. A HCP, a non-smoker, described these difficulties: The doctors sometimes are who writes passes . . . for smoking . . . [and] they need to realize it?s very difficult for us sometimes to deal. You may be dealing with a very strong personality. . . . I was recently in a situation like that. ?I know you have passes, I know the doctor said you can go on passes with smoking, but please bear with me. . . . I need to follow protocol and you need to get new passes?. . . . Trying to read the smoke-free unit policy, I can see the person getting angry and upset with me. I was like ?oh my God? and all I could do was complete my smoke-free policy along with my paper and just try to get a doctor . . . and tell [my team leader]. It?s difficult, I don?t like that position. [HCP28/B] Such HCP hierarchies were embedded within organizational structures, and many participants suggested that there was pressure to be complicit in accepting those power dynamics.  Northern Health, as the governing organization, oversees all patient transfers within the region. Joint responsibility in policy implementation was clearly the intent of the policy-    81 makers. Two documents made available to employees across Northern Health illustrated this interpretation:      The language used by the policy-makers and managers emphasized institutional values, and joint responsibility and solidarity would yield benefits for all. Emphasis on ?responsibility? suggested individuals were expected to actively implement the policy. A HCP, a non-smoker, when describing how she saw the rest of hospital enforcing the SFGP, laughed, suggesting: All one has to do is really walk around the hospital to see . . . in Emergency especially. . . . I know at one time, it was flat-out conflict. . . . We?d say you need to have [patients] sign the non-smoking policy [the SFGP acknowledgement] before they come down [to our unit] and . . . be told by the Emerg nurse ?well, it?s not our policy, it?s yours.? I think that speaks to it. [HCP32/B] The policy implementation had inadvertently created new boundaries within the hospital, which in turn threatened to ?other? the HCPs working on the psychiatric ward. Smoke-free Grounds Policy Let?s Clear the Air! It?s Everyone?s Responsibility  [Document: large-lettered text on coloured poster, document no. 10-405-4008 Rev11/07mac ,  posted throughout Northern Health hospital sites] As employees of Northern Health, we all share the responsibility to ensure the Smoke-free Grounds Policy is implemented and followed to provide a healthy environment for all.   [Document: 2.2 Compliance Protocol November 2007?a staff support document issued by the health authority for all employees]       82 Initially, patients were asked to sign an acknowledgement of the SFGP before arriving on the unit when being transferred from within or from another hospital, to make clear the expectations around tobacco and smoking on the psychiatric unit. However, this approach was abandoned within 18 months. When asked why, HCPs confirmed that the signed SFGP acknowledgement was not done by HCPs at other hospitals or on other units within the hospital; participants speculated that informing patients of the policy could incite conflict and stall the transfer [Participant conversations, October 19, 2010]. HCPs were aware of and responsive to community norms and expectations when they acted in their professional roles, including how they implemented the SFGP. They looked for concordance in their roles as HCP and community members. For example, the tobacco-related litter that was once contained and managed on hospital grounds had spread to public areas surrounding the hospital grounds [Field observations, July 6 and 7, 2009/B]. Teaching staff at the school adjacent to Site B complained that people from the hospital were drifting onto the schoolyard and creating excess litter and second-hand smoke [PO field notes, August 10, 2009]. This caused the hospital executive to respond with assurances that the hospital supported the community and would address the issue. Communities were supportive of the SFGP, but wanted to prevent smoking-related problems from moving to their schools, parks, and yards. Challenges existed around finding places for smokers to go off hospital grounds, while not infringing upon neighbouring properties. The presence of the ashtrays on hospital grounds facilitated violation of the SFGP but waylaid community concerns.  Scarcity of resources A scarcity of HCPs and community and organizational resources affected implementation of the SFGP and led HCPs to make practice choices. When unit workload was particularly demanding and HCPs? energy flagged, implementation of the SFGP lapsed. A HCP, a former smoker, revealed how declining energy levels permitted these lapses:   Some people say, I just get so tired of being yelled at for this thing and another, and this is one issue I?m not willing to be yelled at for. So we?ve had a tremendous increase in workload over the summer. It just, we?re used to fairly quiet summers and it just wasn?t so, and it decreased energy level for staff to invest in this and also people have worked so much overtime that, you know, people go in survival mode, myself included some days. [HCP30/B]     83 Self-protection efforts included prioritizing practice choices. In discussing how implementation of the SFGP lapsed in response to seasonal workload changes, another HCP, a former smoker, stated: Like anything, you start out maybe strong in something and then it just kind of relaxes a little bit . . . we were short-staffed and we were getting a lot of agency nurses for a while, too, so and it could?ve been that. In the winter time we definitely had that, so people aren?t as on board with it, too, as people who are flying in for two weeks or four weeks or something. [HCP25/B] The use of agency nurses was intended to address increased workloads, but few opportunities were available to orient new and transient workers to the SFGP and its nuances. Moreover, only two or three nurses were on duty at any one time so use of agency nurses created proportionately high use of temporary staff. While temporary staff might be willing to follow direction, fewer HCPs were championing the policy. There was also some evidence that summer months coincided with policy implementation lapses as a by-product of longer days and weather that was more conducive to smoking outside. Site B managers had taken steps to ensure that both registered nurses and registered psychiatric nurses could independently order NRT for patients who smoked. While ordering and dispensing NRTs was within the scope of practice for registered nurses under provincial regulations, on the initiative of unit management, special dispensation was obtained from the BC College of Registered Psychiatric Nurses to permit psychiatric nurses the same practice. Those in formal leadership roles encouraged taking an active role in assessing, ordering, and dispensing NRT, and tobacco assessment and treatment was seen as a nursing task.  One other factor at Site B was that there was no resident psychiatrist. Consultant psychiatrists were flown in to see patients two days a week. Sporadic presence of a psychiatrist contributed to the creation of practice norms that resulted in HCPs assuming increased independence in their practices. Although this also translated into independent practices around tobacco assessment and NRT supports, promoting smoking reduction or cessation was not the norm. A HCP, who did not smoke, highlighted how the structures of shift changes and a fast-paced work environment curtailed initiatives to support quit attempts with patients:  Right now if a [patient] really wanted to quit . . . we can make time [to talk about what we can do] but it wouldn?t be consistent and this unit . . . moves fast. You?re     84 rarely here more than a few weeks. And how the shifts are done, it?s really difficult because I may be able to get to talk to you on the first day, the second day maybe not, the nighttimes you might not want to talk to me, so just the whole system. [HCP28/B]  Because human resource shortages and the shift work structures were perceived as organizationally bound and outside the control of HCPs, they in turn perceived an inability to provide consistent care to garner tobacco reduction.  Recurring shortages also operated to create resourcefulness and fostered a spirit of self-reliance reflective of Northern BC attitudes. In particular, Site B had difficulty attracting registered nurses and registered psychiatric nurses to fill vacant positions. They relied heavily on nursing agencies outside Northern BC to provide temporary relief, but also turned to hiring licensed practical nurses (LPNs) to help address the shortage. In BC LPNs do not receive specific education and training in acute psychiatric care and therefore do not typically work in acute psychiatry. On-site education was provided to assist LPNs to develop necessary practice competencies and aid integration into the HCP team. Based on my personal knowledge about the tensions between registered nurses and LPNs in BC respecting practice fits, this unit demonstrated a novel approach that worked. The process was likely aided by strong community spirit that reinforced a need to be respectful of others, but it also required strong informal leadership amongst members of the nursing team. The situation also reiterated the importance placed on education at this site and that education was easily available to HCPs only if provided locally.  HCPs found creative ways to support smoking reduction/cessation in the face of HCP shortages. Tobacco intervention expertise was also developed among recreational therapists by educating and training them to counsel patients who smoked. They attended the psychiatric unit regularly to provide recreational therapy and tobacco reduction support to patients. They could also follow patients through to the community care continuum post-discharge, and in this regard, nursing staff could defer, at least in part, to other HCPs to assist patients with smoking cessation efforts. Education predominated as the key to changing attitudes and beliefs about tobacco use, in building strengths and uniformity in supporting the SFGP, and as a means to bolstering local resources. Finally, the lack of security personnel at Site B was an important resource scarcity that increased the potential impact of violence related to the SFGP. If a violent incident were     85 to occur on the unit, wait times for the local police to arrive could be significant. Strategies to avoid violence were, therefore, especially important. HCPs thoughtfully considered strategies to ensure their personal safety. One HCP stated that a reason that the SFGP worked was that HCPs are ?skilled at de-escalating situations? [Participant conversation, August10, 2009]. Scarcity of resources created resourcefulness and self-reliance that affected HCP practices, their professional practice decisions, and implementation of the SFGP.  Conclusions Comparison and contrast of the two sites revealed cultural knowledge and contexts impacting SFGP-related experiences of HCPs. In terms of similarities, HCP experiences at both sites were laden with challenges and struggles. Site A was primarily engaged in preserving pre-policy practices (the status quo) whereas Site B was more fully engaged in implementing the policy while addressing new practice challenges. Site A provided an example of resistance to the SFGP to a point where the policy seemed to have little impact. Site B provided an example of concerted efforts to implement the policy, where external resistance was managed and addressed, at least in part, with management guidance and the policy presence, and varying degrees of impact were visible.  Energy at both sites came from multiple sources, including individuals, group processes, education, and leadership. At Site B, implementation more closely mirrored policy-maker expectations, but during data collection this unit experienced increased incidents of SFGP rule-breaking that resulted from demoralization, fatigue, and lack of support from other hospital areas. This team of HCPs possessed hardiness and sense of pride in the fact that, when asked to perform, they did their best [PO field notes, August 16, 2010]. Here, HCPs were challenged by management to adjust pre-policy attitudes and adopt evidence-based practice to support the SFGP implementation. Across the two units, different unit cultures and histories influenced and were influenced by HCPs to shape their responses to management efforts designed to bolster policy implementation. Strategic leadership and management at Site B was contrasted with front-line HCP leader groups at Site A. Site B had a policy champion who encouraged a sense of HCP ownership of the policy, and Site A did not. Site B put in place support for changes to HCP practices as a result of the SFGP; there was less energy directed toward this at Site A. Across     86 both sites, the introduction of the SFGP resulted in HCPs beginning to question their own practices, knowledge, and attitudes toward tobacco use in this patient population. HCP experiences indicated that SFGP implementation and HCP practices within these contexts required nuanced approaches or subtleties of understanding and tacit cultural knowledge about implementation that must be developed over time.  Tobacco assessment and treatment were taken up as nursing tasks at Site B, but not at Site A, and this differentially influenced overall awareness and acceptance of tobacco. At Site B, programs were underway to use recreational therapists to start cessation efforts with inpatients who could be followed through to community care after hospital discharge, while recreational therapists were not part of the HCP complement at Site A. Site A HCPs, in efforts to maintain status quo, held onto traditional views of the importance of tobacco to patients that included patient-centred, rights-based arguments to avoid HCP responsibilities associated with the SFGP. At Site B, traditional views on tobacco were openly challenged in ways that fostered team building and HCP professional practice development. Overall, HCPs revealed the importance of various factors in influencing SFGP implementation. These factors included leadership, staff morale, education, sense of team or cohesiveness, cultural norms, passage of time, personal beliefs, resource availability, organizational structures, and northern contexts. The contexts of Northern BC created unique work environments that included additional considerations of community connections and resource availability. Concepts of caring, safety, autonomy, and fairness would undergird many HCP practice choices. HCPs located patients at the centre of their concerns in making practice choices related to SFGP implementation, and tobacco-related activities continued to heavily influence daily routines. HCPs at the two sites demonstrated both similarities and differences, and different contexts created unique experiences of SFGP implementation despite policy-makers? initial desire to produce uniform implementation.      87  It?s almost like putting you on a new train but you don?t want to lose the old train and the new train of course will get you to where you?re going, but what is that if you leave all your belongings behind?  [PT48/B ? a smoker in his 50s admitted for schizophrenia and bipolar disorder speaking of his experiences with medication changes and being without cigarettes while an inpatient at Site B]   In this chapter, findings are presented that address the research question, What are the smoke-free grounds policy (SFGP)-related experiences of patients in inpatient psychiatric units, and how do cultural norms and institutional structures shape these experiences? Many similarities existed among patients at Sites A and B, and the findings are integrated across sites in what follows. The findings are arranged into three themes: 1) resigned to smoking, 2) responding to power and control, and 3) policy responses. Where significant site-specific differences in patient perspectives occurred, they are described within the thematic findings. The chapter concludes with the summary and closing remarks.  Resigned to smoking Smoking was widely accepted as part of the culture of the psychiatric units. For patients who smoked, most had started smoking in their youth, and it became tied to coping with their mental illness. Patients spoke of how smoking helped manage mood and reduce stress. Smoking was positioned as central to their lives, a position also ratified even by patients who did not smoke. A patient in his 20s, a smoker hospitalized with depression and schizophrenia, in considering others?, illustrated these beliefs: I mean if they take away smoking from people who are already under such big stress . . . it can be way too hard because they?re already dealing with more than they can handle. [PT2/A] Permitting smoking at the hospital was viewed as humanitarian and led many patients to expect healthcare professionals (HCPs) and hospitals to accommodate their smoking. A Chapter 5 -  FINDINGS: PATIENTS? PERSPECTIVES     88 patient who was a non-smoker in his 30s, admitted for what he described as a ?mental illness,? also spoke of the importance of tobacco to co-patients: They don?t seem to have too much to look forward to and they put a lot of their interests and energy into their smoking being the one [thing] that helps them really feel better . . . it?s a tough addiction. [PT22/A] For this patient, smoking was understood as pervasive and broadly accepted, and was positioned as easing patients? anxieties by offering respite, stimulation, and connection to other smokers. Smoking cessation was not made a top priority by patients and, therefore, neither was the SFGP. Illustrating the long-term relationship to tobacco among patients who smoked, a patient in her 50s, admitted for depression and medication adjustments, explained the role of tobacco in her life: Well, it?s my life. It?s over 30 years that I?ve been smoking. . . . I started when I was around 12 and quit for 10 months [just prior to my hospital admission], [when] everything went out of whack . . . and first thing I could do was buy a pack of smokes. So it?s been a major part of my life. [PT1/A] She suggested being a smoker was linked to her identity as well as her mental illness, and she turned to smoking as a familiar means of coping. Other patients, including a patient in his 40s with schizophrenia and some cognitive deficits, directly linked smoking and mental illness: Smoking . . . takes the illness away for me . . . the smoker knows the risk of getting cancer and that you can die, but as long as you do all things in moderation . . . [smoking] shouldn?t affect you much. [PT19/A] For this patient, as with others, the risks of smoking were traded off for the benefits derived. Mental illness management received priority over the potential of smoking-induced illnesses. This patient ruled out quitting and believed a HCP would determine if and when he should quit. He believed he would safely smoke until a doctor suggested otherwise: The doctor is going to tell you when to quit, and that?s when the person should quit . . . when he gets the paper in the mail saying that the smoker has cancer. . . . the doctor catches the cancer in time . . . that?s when I think they should quit. [PT19/A] Some patients lacked full understanding about the risks of smoking and subscribed to traditional behaviourist patient?HCP interactions and hierarchies that positioned HCPs as     89 expert. As such, for some patients, responsibility for their smoking defaulted to their HCP and, without HCP intervention, patients remained resigned to smoking.  An established routine of HCP?patient interactions around fresh air breaks (FABs) revealed broad acceptance of tobacco and resignation to smoking. Most smokers regularly utilized FABs that required some interaction with HCPs to facilitate a short leave from the unit. If conversation was included, each HCP?patient interaction facilitating a FAB presented an opportunity to increase therapeutic connection. My observations confirmed that Site A practices in adopting set times for FABs and requiring patients to write their name on a white board reduced personal interactions to little more than a gesture to get the attention of the HCP and indicate white board sign-out. Site B, in contrast, had not adopted a sign-out procedure or a strict ?one break for every hour? approach to FABs, and patients? FAB requests often included information gathering, assessments, and expressions of concern from HCPs. Site A procedures were normalized as routine and behaviourally focused on patient management, while Site B activities remained more humanistic and focused on individuals. That said, HCPs at both sites continued to act within the SFGP protocols. Most patients perceived of FABs as connected to smoking activity so FABs provided regular reminders of the presence of tobacco at both sites. An increased use of nicotine replacement therapies (NRTs) accompanied the SFGP and created a new set of routines and norms on the units. For most patients, NRTs were poor substitutes for smoking that could occur at a FAB. But for most newly admitted patients FAB privileges were not immediately acquired. A patient in her 40s, who smoked and had been admitted for anxiety, said:  It?s just hard to see the people who have passes going out because I know what they?re doing, which makes it more difficult. The withdrawals are no fun. They?ve given me the [NRT] patch, but you still need something to do with your hands. ?Follow the rules and try to get a [FABs] pass? [is what I would tell a new patient here]. [PT39/B] The desire to get FABs was strong, and many newly admitted patients to the units initially refused NRTs from HCPs, believing that if they accepted NRTs they would be precluded from getting FABs. When they realized that use of NRTs did not affect FABs, they were     90 more likely to utilize NRTs to avoid cravings while waiting for FAB privileges [PO field notes, August 15, 2009].  The preferred form of NRT was the inhaler. The NRT inhalers were short, white, plastic apparatus that looked similar to a cigarillo tip at one end, blunted at the other, with an air intake hole. With tip to mouth, patients inhaled to draw air through the nicotine cartridge and receive a dose of nicotine. Replaceable cartridges inside were designed to deliver as much nicotine as would a standard cigarette. The degree to which patients handled and used NRT inhalers like cigarettes was remarkable: A new patient to the unit crosses in front of the nursing station and surreptitiously drops an item into the hand of another passing patient, then smiles. He has ?handed off? his inhaler to another and the new owner initially hides the inhaler in a paper he is carrying. He then takes a puff from the inhaler in a rather secretive fashion, returning the inhaler to his cupped hand to conceal it as if contraband. [PO field notes, August16, 2009] Polite and regular patient interactions with HCPs in obtaining replacement cartridges were reminiscent of the old system, whereby HCPs dispensed cigarettes to patients hourly on request. My observation notes reflected some of the routines associated with inhaler use: A patient arrives at the nursing station window. ?Can I have another capsule please?? . . . He takes the cartridge, assembles the unit, walks away, returning a few seconds later with the inhaler in his mouth. He removes it, interrupting his inhaler activity to ask the nurse a question. His posture and practices suggested he is directly substituting smoking rituals in how he uses the inhaler. [PO field notes, August 18, 2009] A non-smoker in her 50s, hospitalized for a medication adjustment, talked about the NRT activity on the unit: They are giving [patients] little white cigar thingies . . . a young patient is using one who never smoked before . . . so they?re not checking . . . she may have said ?oh yeah, I smoke,? and she gets it just to be cool or something. [PT38/B] Descriptions of the NRT inhaler as a ?cigar? reinforced its likeness to smoking, and here, NRT use had social implications. In a setting where smokers often outnumber non-smokers, finding social acceptance may come in the form of possessing and using NRTs.      91 Talk about NRTs and use of inhalers in manners resembling cigarette use offered a visual cue that might trigger cravings to smoke for some patients [PO field notes, my reflection, August 13, 2009]. To the degree that gum and inhaler NRTs required regular replenishment to meet patient cravings, they became quasi-tobacco products. HCPs were the keepers and dispensers of the requested source of nicotine when patients had cravings or sought the hand-mouth activities associated with smoking behaviours. Symbolically, the use of NRTs helped maintain the centrality of tobacco. One patient in her 40s was observed to maintain a 12-day quit attempt using NRT gum [PO field notes, August 12, 2009/B]. Re-admitted 15 months later for anxiety, she had returned to smoking 4 months after quitting because she was ?hanging around a smoker.? She revealed how the influence of peers and the presence of tobacco eventually countered her quit attempt. She also believed smoking helped relax her and was observed to draw heavily on the NRT inhaler she brought to the interview [PT44/B]. Patients viewed smoking as stress relieving, and most viewed stress as influenced by their surroundings and as being beyond their control. Endorsing hospitalization as stress relieving for those with mental illness, a patient in his 20s admitted for depression, who smoked, said: It?s pretty fun up here, you know. It?s a stress-reliever and everybody?s nice and you don?t worry about too much. You get your three meals a day, you get your snacks, get your juice. It?s pretty good, man . . . relax and enjoy. [PT16/A] For this patient, the unit offered elements of social connection, safety, and security that were missing in his life off the unit.  Cigarettes formed an integral part of the psychiatric patient culture that included sharing, regularly asking each other for cigarettes, picking up cigarette butts, and assisting those who needed help to get outside to smoke. The less visible, but continued, presence of cigarettes contributed to an environment that facilitated smoking. A patient in her 50s, a smoker admitted for depression, believed hourly FABs aligned with the physiology of tobacco addiction:  I think they have been wonderful; every [hour there is a] 15 minute . . . break . . . [and] if you are an addict, it is about right, so the hospital is very much taking that into consideration. [PT1/A]     92 That she construing the timing and provision of FABs as ?wonderful? revealed beliefs that the hospital had a duty to be compassionate and caring by accommodating smoking. This patient underscored the embedded nature of resignation to smoking. In commenting about his general impression of the SFGP, a patient in his 40s with schizophrenia, who smoked, suggested the need for designated smoking areas while continuing to support the policy more generally: Outside is the best way to keep it [but] . . . we should have areas that are covered like for rain and stuff. [PT19/A] This patient also suggested an outdoor ventilated dome with heaters during the winter. Protection from the elements was part of the duty to accommodate, especially in cold northern climates. No matter the degree to which patients who smoked supported any good intent of the SFGP, almost all believed that designated smoking areas on hospital grounds addressed issues important to patient safety, comfort, and treatment, to demonstrate humanistic caring. Paternalistic actions of HCPs were sometimes viewed by patients as compassionate. This invited HCPs into the handling and management of tobacco on behalf of vulnerable patients. A patient in her 50s admitted for unspecified reasons, who had smoked for over 50 years but quit six months prior to her admission because of lung disease, told of how she felt HCPs had a continuing role in cigarette management: I notice that [patients] will keep [cigarettes] in their room. And I said, ?you?re taking a big chance because if you have them in your room, someone might come along and steal them?. . . . I?m trying to encourage them to get their name put on them by the nurses . . . and take one or two at a time. That way they know they?ve got so many left and nobody?s got into them. [PT21/A] She encountered resistance from some patients on the suggestion that they turn their property over to staff, but estimated that 40% of patients did turn their tobacco over to HCPs. This large number was not substantiated by participant observations or HCP interviews, but some patient tobacco was known to be held by HCPs. Remarkably, despite recognizing that smoking caused her lung disease, she continued to endorse smoking as necessary for other patients. Her tone and behaviour during the interview suggested that she craved smoking as a coping mechanism. This patient had had many hospital admissions and recalled how she used     93 the smoking room on the maternity unit after delivering her baby. She, like other patients, resisted a shift away from the past when tobacco use at hospitals was widely accepted.  Patients also sensed HCPs? resignation to them smoking. Admitted for depression and schizophrenia, a patient in his 20s who was asked to recall if HCPs talked to him about his smoking said: One of the nurses was asking me . . . when was the last time I smoked marijuana so I told him about 3? years ago and he said that?s good and then I forget what else he said but I asked him, well should I quit smoking, too? And he . . . told me it doesn?t affect me the same way as marijuana does. . . . And one of the other nurses was asking me if I?d ever quit before and how I went about doing that. [PT2/A] Asked what these conversations were like for him, he said ?it was good? and he liked them. Evident here, however, was how a patient?s contemplations of quitting as an expressed interest in achieving improved health was not always supported or followed-up by HCPs. A patient, a smoker in her 40s, admitted for anxiety, was asked if anyone had talked to her about smoking during her hospitalization and she said, ?Just the nurses say I can get NRT whenever I want? [PT44/B]. Other patients confirmed that patient conversations with HCPs were often limited to offers of NRT, while some suggested that they were never asked about quitting. This operated to affirm further patients? smoking and their beliefs that smoking was not a priority health issue for many HCPs.  Responding to power and control Patients? experiences with the SFGP were closely tied to perceptions of power and control, and patients responded to these dynamics. Patients found innovative ways to circumvent the SFGP and adapt to various structures of control. Institutional messaging about the SFGP was perceived by patients to govern tobacco use amid an array of other controls. Patients? experiences illustrated restrictions on freedoms to act, possess contraband, and speak about cigarettes. At Site A, many patients highlighted a small picnic table behind the Emergency Department entrance as the unofficially designated place for ?staff smoking.? Patients observed staff smoking there, and a large metal can filled with cigarette butts left lingering evidence. Patients understood the space as excluding them. A patient in his 20s, a     94 non-smoker with depression and addiction issues on his way to be voluntarily admitted into a detoxification inpatient unit, spoke about sitting at the picnic table, an action that contravened the staff?patient segregation:  A few staff members came out and asked me to move . . . from their picnic bench and . . . they were there to smoke. . . . I was not doing anything other than sitting there reading the newspaper and I felt really judged. . . . One of them mentioned a couple times that ?you?re not allowed to be sitting here? which I know was not true. . . . They were quite rude, let me tell ya. [PT18/A]  The patient singled out one of the staff, and while ?not wanting to judge,? he surmised the employee was unlikely to be a nurse because of her use of foul language and the unprofessionalism she exhibited. Though angered, he abided and ?walked away? to avoid any trouble. He described how the actions and words of the employee stigmatized and marginalized him. While he was taking important steps to improve his health, his lack of power rendered him compliant and perhaps weak. Here, too, hospital staff?s violation of the SFGP confused some patients because they had been told on admission that smoking could only occur at the hospital boundary sidewalks closest to the public streets. As a further type of discrepancy for patients to try to reconcile, inequities in the application of the SFGP were frequently mentioned by patients. These inequities helped further define the perceived authority of HCPs and the hospital. A patient in his 50s, hospitalized with addiction issues, also commented on the imbalance of someone sitting in a vehicle and smoking directly outside an entrance door:   I don?t understand how that rule works. I?m out there by the ashtray and this guy?s right in front of Emerg, smoking a cigarette in his truck . . . both windows were open . . . I didn?t say nothing, I just thought of it for a minute because I?ve walked all the way to an ashtray when I?m sick to do it. The guy pulls up in his $50,000 truck with his stereo going, smoking a cigarette. Kind of a slap in the face, you know what I mean? [PT20/A] Revealing aspects of patient life and cultures, the participant highlighted his subordinate status and the inequity he endured. The institution in this respect invoked a set of rules on patients? tobacco use that others routinely broke.  At both sites, confiscation of tobacco from patients was part of the SFGP. A recently discharged patient in his 50s, who had schizophrenia and bipolar disorder, spoke of a sense     95 of loss that resulted when his cigarettes were taken away on the unit and of his loss of autonomy in having to ask others for cigarettes. He also indicated fear of being seen as bothersome to other patients by having to ask for cigarettes. Reflecting on past experiences, he spoke of being labelled as mentally ill and of the impact of bothering other people:  Smoking is getting to the point where it?s, people get picky about things and if you keep bothering them and you get a name for yourself that you?re bothering people . . . then you have an incident. Well, the wheels start to turn and you can get killed because somebody doesn?t like you. [PT48/B]  Throughout the interview, this participant spoke of a distrust of people and the mental health system, the perceived control of the system over some patients, and the double stigma impact of smoking and mental illness. He used an analogy of a robin amongst a flock of pigeons and how all the pigeons would turn on the robin as bothersome, identifying as the robin if he continued to ask others for cigarettes or overstepped the kindness of others. He internalized stigma by virtue of not being able to smoke, but needing to. He described the increasingly negative social attitudes toward smokers, and this heightened concern about his personal safety. A patient in his 50s, who smoked and had been admitted for addiction-related issues, also spoke of experiencing the dual stigma of smoking and mental illness:  It is society?s perception of, you know, what they attached it to. Who they hung that jacket on. It would be us, you know? Normal people at home that smoke, [they?re] okay, let?s pin . . . that bad habit on somebody else. Let?s not pin it on the hardworking guy and you know the guy with two jobs, the wife that has to take care of the kids all day and doing really well, the Jones?s next door that are doing well. You can?t put that on them, so they try to just deflect that kind of blame in different cultures and areas; [to] lower cultures of people. It?s just more prevalent in a lower class of people. Drug addicts, whatever, prostitutes, mental patients, whatever you want to call them. [PT20/A] This patient highlighted that marginalized subgroups smoke and, therefore, smokers were categorized as likely embodying a lower socio-economic class and, perhaps, mental illness. Smokers were compartmentalized as having deficits that validated exclusion and status loss or that predisposed them to be smokers. Stigma of smoking was associated with those other marginal, subordinate states, or social determinants of peoples? lives.      96 External stigma emerges from power imbalances favouring dominant cultures that distinguish difference and labels, discriminates, and negatively stereotypes minority groups. The factors underlying the power imbalance necessary for stigma formed part of the social, economic, and political structures to which patients responded. For most patients, stigma was part of what shaped their SFGP-related experiences. The SFGP literally made smoking and smokers more visible by moving them out of any type of shelters and onto the sidewalks by the streets. A patient in her 50s, hospitalized for unspecified reasons, who had quit smoking six months before her admission, said: If I was smoking still, I would probably be smoking [on the hospital grounds] because I wouldn?t want to be [near the street] where everyone can see me smoking, because I would be embarrassed by everybody. [PT21/A] The SFGP conveyed mixed messages to patients about institutional values of caring because of its effects in further stigmatizing patients who smoked. Some patients who experienced stigma as a result of increased visibility responded by isolating their smoking activities away from others. A patient in her 40s, hospitalized for unspecified reasons, in talking about the lack of compliance with the SFGP at Site B, was quick to indicate that people avoided ?hassling? her over smoking because they identified her as a psychiatric patient, stating:  People are pretty careful about approaching anybody from [the psychiatric unit] . . . you just look at somebody funny and they?d be backing away from you. [PT45/B] This patient found ways to exert her own power in subtle ways. A patient in his 50s, admitted for addiction-related issues, said of the social impact of smoking and having mental illness:  [Mental health patients] click together, too, because they know they?re sort of balked at in a certain way, socially balked at. As soon as they find out you?re coming off the [psychiatric unit], people have a different opinion of you. I can feel it in the air right behind me here. I can feel the stares, the smirks or whatever, right? I just bulldoze through it and keep doing what I?m doing. It?s there, though. [PT20/A] For these patients, mental illness was isolating, but the common activity of smoking generated social bonding and acceptances they do not get from others. Patients understood how the discrimination of being a psychiatric patient might at times work for them, rather than against them, in facilitating smoking on hospital grounds. Site B patients seemed more     97 willing to indulge in flagrant violations of smoking on the hospital grounds than those from Site A, perhaps because they knew there was no security personnel to enforce the policy, or that the smaller community and hospital at this site made them known to others as psychiatric patients and thus avoided them.  The external stigma patients experienced was influenced by regional geographies and location. Living in a small Northern BC town and being admitted to an acute psychiatric ward, even if outside their home community, threatened anonymity and made these patients vulnerable to additional sources of stigma related to mental illness. Many sought to reduce the stigma of mental illness by hiding their related hospitalizations from community members. Absences from small remote communities were difficult to explain. One patient, a non-smoker in her 50?s hospitalized for medication adjustments, illustrated this point:  I?m ashamed to be here because I?m afraid other people from . . . my small town might come in to visit somebody or also be coming in, and then the label that I get is so awful and I?m struggling with that. [PT38/B] Coming from a small settlement in the north, she risked being ostracized if known to have a mental illness. Additional comments she made expressed shame that she was getting better and occupying a bed on the unit that someone else might more urgently need. She reflected community values of allocating scarce resources to those most in need, and self-evaluated her own actions as inconsistent.  Several patients spoke of and characterized some public policy and big business actions as conspiracies that targeted and stigmatized those who were most marginalized. A patient in her 40s, who smoked and was admitted for unspecified reasons, surmised: [The government] can?t tax people to death on cigarettes and have legal sales of drugs like that and then say it?s harmful to you and you shouldn?t be smoking. [Tobacco] taxes are high because [governments claim to] put it into the healthcare system. Of the taxes taken from cigarette sales, how much is put towards health care? [It?s] not very high . . . it?s all propaganda. It?s all rhetoric. [PT45/B] Drawing attention to the inconsistencies of government actions, she sourced numerous examples of how governments discriminated against marginalized subgroups in Canadian society. At a macro level, government tobacco legislation and tobacco taxes also eroded some participants? limited economic power. The same patient explained:     98  I?ve [complained] about the government?s policy on anti-smoking, especially [for its effect on] people who don?t have money, are on social assistance, or are in hospital. It?s a joke, really, because what happens is those kind of people that are told no, we don?t give you money for [cigarettes], well, they just get the tobacco anyway. I mean they harvest it; they literally pick up the butts. I?ve done it lots of times. [PT45/B]  Tobacco taxation was viewed to further marginalize those addicted to smoking. Additionally, high tobacco prices forced some participants to illegally purchase cigarettes from First Nations (Aboriginal) sources. In Canada, historical treaties exempt Aboriginal peoples from paying taxes on tobacco, and despite a prohibition on re-selling these tax-exempt cigarettes, the price difference is enough to stimulate a black market. Others smoked cheaper hand-rolled cigarettes and/or picked up discarded cigarette butts to counter their economic disadvantage.  Patients perceived tobacco companies as intentionally promoting and supporting their addiction to tobacco and often in deceptive and manipulative ways. Patients knew of some of the chemical constituents of cigarettes and viewed their use as purposefully manufactured. Many patients commented on a graphic poster at Site B of a cigarette with numerous harmful chemical constituents of a cigarette listed. Beside each chemical name was a picture illustrating its other uses, including car batteries, nail polish remover, rat killer, and insect killing sprays [Observation field notes, August 10, 2009/B]. A patient in his 30s and a non-smoker admitted for what he labelled ?mental illness? said of cigarettes: Maybe they should make smokes without any of those chemicals because I think a lot of it is the big companies make it so it?s impossible to quit . . . it?s like a crime. [PT22/A] Reflective of assertions that tobacco companies were culpable for enhancing the addictive effects of cigarettes through use of chemical additives, this participant was sympathetic to the expressed helplessness of others related to smoking addiction. Shifting blame for addiction to tobacco companies was one way to deflect the public stigma of smoking. Stigma here was rooted in Canadian society?s emphasis on autonomy and personal responsibility for addiction that pointed to defects in those who smoked, the same position utilized by tobacco companies to disclaim responsibility for people knowing the health risks of smoking.     99 Smoking was seen as a legal activity that patients expected they could continue. In addressing what she thought of the SFGP, a patient in her 50s, admitted for a medication change, who did not smoke, said: I understand that . . . there?s a ban in public places but the option is allowed for the employee to go to a place to [smoke] yet they don?t allow that freedom to a [psychiatric] patient. . . . So they?re taking that freedom to say, ?I choose to still smoke because it is not illegal.? . . . I think that?s wrong. [PT38/B] In the psychiatric patient context, smoking restrictions imposed upon psychiatric inpatients were viewed as distinct from those imposed upon employees or other patients because they already endured restricted freedoms to come and go from the unit. The psychiatric units also imposed restrictions on commonly available stimulants. Site B, for example, had switched from caffeinated to decaffeinated coffee and restricted the availability of soft drinks containing caffeine. Patients viewed these restrictions as additional controls. A patient in her 40s, a smoker admitted with anxiety, had been drinking more than 20 cups of coffee a day at home, and since her admission had suffered persistent headaches as a result of caffeine withdrawal. While HCPs provided her with headache medications, she also sought relief through drinking the caffeinated soft drinks a friend had sneaked in for her. Contravening the rules, however, fostered her guilt and fear of being caught, which in turn did little to waylay the anxiety for which she had been admitted to the ward: This is all confidential, right? No nurses will know. That my friend snuck me in some Pepsi. . . . So the headache?s not so bad . . . I?m not going to torture myself right now. I?m in here to get help, not to torture myself. Even though if they found it I?d probably get my ass kicked. [PT39/B] She conveyed the contradictions and counterproductive practices that can be deployed on psychiatric units under the guise of caring for patients. Restrictions were bundled, and tobacco comprised but one of the restricted, policed substances, while the cumulative effect rendered patient cultures subordinate and without the autonomy to choose. The SFGP also prohibited discussion about tobacco, to prevent triggering cravings in patients who were unable to smoke. A patient in her 20s admitted with depression subverted this rule by covertly discussing her own tobacco cravings with another patient. Asked about why she did not talk to a HCP on the unit about her cravings, she said:     100 They?re nurses. You generally know what?s going to be said about [tobacco and smoking], like rules and regulations. [PT47/B] This patient perceived interaction with HCPs around tobacco cravings to be rule-bound with little therapeutic effect or that the addiction of tobacco could not be talked through with them. This meant that for some patients the SFGP was understood to close down dialogue about tobacco addiction despite policy intent to empathize and support patients with the addiction.  During participant observations, a HCP on Site B was seen to escalate a patient request for a FAB to a conflict, by taking issue with the patient?s late return from an off-unit break the previous day. Observed to be particularly rule-bound in patient interactions, this HCP was quick to remind the patient that passes were at the ?discretion of the nurse? and ?a privilege, not a right.? The patient, confused by the HCP?s unreasonableness and demeanour, and frustrated by her own helplessness in the situation, expressed increased paranoia about HCP motivations for such practices [PO field notes, August 2009]. Patients perceived an ?us and them? in their interactions with HCPs that was supported by the structures that organized HCPs? work, the various rules imposed upon patients, and the physical structures of the unit. While most patients viewed HCPs to be motivated by positive therapeutic intent in their actions, some HCP actions challenged that viewpoint. A patient in her 50s, a non-smoker who was admitted for undisclosed reasons, told of the control HCPs have over patients who were late returning from FABs: Nurses will control them and tell them that they?re late and that they?ll sometimes lose their privileges to go out again, you know, and stuff like that. [PT21/A] This patient provided insights to the mechanisms that sustained unit norms and indicated that privileges would be lost for what she called ?misbehaving.? Examples of misbehaving she described included ?screaming, yelling, and throwing things.? FABs were understood by all patients as privileges, so FABs became closely tied to a reward system to persuade patient behaviours.     101 Policy responses The SFGP simultaneously impinged upon patients? personal freedoms and sought to protect their health. Patients struggled to reconcile competing views about the intent of the SFGP and balance the self-interests and rights of others. Patients at times indicated understanding of and compliance with the policy, but at other times resistance.  Patients? compliance with the SFGP was influenced by broader community norms and the social norms of the unit. A few patients pointed to the complexities and challenges inherent in developing and implementing tobacco control policies. A patient in her 20s, a smoker admitted for depression, stated: It?s fairly reasonable. I mean, some people don?t really get why [the policy is in place]. . . but I mean you look at it and it?s not only them that matter, it?s the other people too. Like all the people in the hospital, it?s kind of directed to everybody being healthy, staying healthy and this is a place where people have to come in order to get better and if people smoke anywhere close to it or inside, then smoke can [get too near people] and it?ll generally hurt everybody else, too. [PT47/B] She contextualized the SFGP as one of many hospital policies, all of which were intended to advance the health and well-being of patients and staff. A patient in his 40s, who smoked and had been admitted for medication changes for schizophrenia, provided an example of the concern patients showed toward exposing others to second-hand smoke:  I do agree, though, that people should smoke outside instead of inside the building. People should be concerned about each other, a smoker or a non-smoker. [PT19/A] Compassionate, this patient validated the SFGP for removing smoking from inside the hospital as a legitimate means to protect others from second-hand smoke. A non-smoker patient in his 30s, admitted for unspecified mental illness, spoke about ?different variables in different cities,? suggesting that places with higher numbers of ?poor? or ?native? residents who were known to smoke more could counter smoking cessation education and interventions [PT22/A]. He also explained the complexities of addiction and the influence of tobacco company marketing and messaging that downplayed the dangers of tobacco use. He also said how tricky it was to try to find a policy that was humanitarian for inpatients with     102 mental illness who smoked, while also trying to advance the health of all patients and employees. Linking their concerns with broader social norms, patients highlighted how implementation of the SFGP created new litter issues. There were limited options to control cigarette butt refuse because the policy did not allow for ashtrays on hospital grounds. In speaking of the appearances of the grounds near the hospital, a patient in her 50s who smoked, hospitalized for medication adjustment, underscored the environmental effects of cigarette butts:  It?s disgusting! . . . There are cigarette butts absolutely from here to Timbuktu. . . . I don?t ever leave them on the ground; I put them out and put them in the garbage. I don?t leave them behind. I mean there?s animals, there?s water. . . . The sprinklers come on and wash all the cigarette butts down the [storm] drain. Right into the river or wherever . . . even to a smoker, that is not acceptable. [PT1/A] This patient avoided and advocated against littering as a reflection of strong cultural norms respecting the natural environment. In Northern BC, historically, natural resources were harvested with minimal regard to environmental impact. Shifting cultural norms now emphasized stewardship, conservation, and protection of natural ecologies. The problem of cigarette butt trash was also observed in field observations, and large scatterings of cigarette butts were located in areas where people could comfortably sit and smoke, and near building entrances. Litter revealed both the activity and end by-products of smoking.  In terms of passive resistance, patients also reported how they acted to protect other patients from getting into trouble by remaining silent about rule infractions witnessed. This maintained a cultural norm of protecting each other and maintaining group membership. A patient in her 20s, admitted for depression, and a smoker, said: There was a couple of instances that I never told the nurses about where I go into like the shower room and I?d smell cigarettes so I wasn?t sure if people were going in there to smoke or not, but I smelled it. [PT47/B] Blending in and social acceptance were two powerful forces that influenced patient behaviours. Examples of this cultural norm included patients relating accounts of SFGP infractions to me during interviews while not revealing patient identities. A patient, a smoker in his 20s admitted for depression said: ?I?m not giving names or tattling or anything, but I     103 know they were sneaking a lot to go smoke? [PT46/B]. When asked what he observed of HCPs interacting with other patients around smoking, a patient in his 30s, a smoker admitted for medication issues, said:  Just one on one, I don?t pay attention to what happens on the unit. What?s their business is their business. We know that. Absolutely, the people on the unit is [HCPs?] business. We don?t intermingle. That?s [the HCPs?] job to deal with it. We don?t get involved. [PT15/A] This patient revealed insider patient knowledge about the need to mind one?s own business and the patient?HCP divide on the unit. A patient, a non-smoker hospitalized for medication adjustments, illustrated how some patient knowledge remained concealed from HCPs. In knowing where patients hid tobacco, she said: They hide [cigarettes] in the bushes. There?s a little door under the [patient?s] bed and they?ve opened that and put them in there . . . and there?s sort of a false ceiling . . . they use Velcro and put it on one side of their [cigarette] pack and stick it up behind the false ceiling. [PT38/B] This patient revealed how group cohesion protected disclosure of some information to HCPs, which operated to further define and preserve an ?us? and ?them? binary.  Galvanizing patient bonds with each other was important to circumventing the SFGP and in finding social acceptance. Patients routinely expressed empathy and care for one another, whereby sharing tobacco was premised as embodying concern for others. In commenting about a need to share tobacco with other patients, a patient in her 50s, hospitalized for medication adjustments, empathized with patients without cigarettes who wanted to smoke:  I end up giving them one because I know they are stressed out, because well . . . you gotta share, I guess, sometimes. [PT1/A] She was motivated by insider knowledge about the therapeutic value of smoking to reduce stress and provide reward and respite from an array of mental illness issues, practices informed by patient group norms.  Patients? desire to help each other also emerged in relation to where smoking took place. A patient in her 50s, a non-smoker hospitalized for medication changes, helped cover     104 up patient smoking on the enclosed patio at Site B. Her actions were driven by concern about the added stress that strict SFGP enforcement invoked on other patients: They?ve gone out and they?ve smoked in the corner of the patio . . . they?ll say ?can you stand at the door to see if anybody?s coming? because then those cigarettes will be confiscated . . . so they are puffing away as fast as they possibly can . . . and I feel like I have to take part in that conspiracy which I don?t really want to . . . [but] I don?t like to see people with added stress. [PT38/B] This patient located smoking as therapeutic, despite not being a smoker herself. Also revealed were her insider understandings about what is needed for patients to break with the official rules of the SFGP. In this respect, knowing the official rules afforded her, and many other patients, knowledge about how to resist governance attempting to erode long-standing smoking practices. Patients resisted elements of SFGP implementation and assertions by some HCPs who sought to strictly enforce the policy. Some patients? agency in resisting control was illustrated by a patient in her 50s, who smoked and had been hospitalized for unspecified reasons, who spoke of her compliance and of other patients? resistance: [Other patients are] asking me where to smoke and I?m telling them, ?go off the property,? but they?re saying well I?m smoking on the bench or I?m smoking on the side of the building and I?ll say, ?well, no, because the [air intake] vents go there and the smoke is coming in.? . . . Some people even smoke in the bathrooms and you have to watch for that. [PT21/A] This patient was complying and argued against resistance. Other patients, including one in her 40s, with anxiety, who smoked, participated in an underground resistance suggesting ?you . . . can?t smoke on the grounds, but you could sneak out back (laughs)? [PT/44B]. The same applied to a patient in her 30s with depression, who indicated the ease by which the rules can be contravened without consequence: ?They prefer you go off hospital grounds [to smoke] but I?ve never gone off hospital grounds, and as an inpatient I?ve never been asked to step off the property? [PT/35B].  Security guards at Site A and hospital employees at Site B were viewed as the potential enforcers of the SFGP; however, it was understood by patients that there were few if any consequences for smoking on the hospital grounds. In effect, while mindful of security     105 personnel and hospital employees, most patients suggested that the policy was not strongly enforced and if they were ?caught? smoking, the consequences were negligible.  A patient in his 20s who smoked and was undergoing medication changes for depression highlighted the visibility of smoking restrictions: It?s all over the place, you see signs, people see signs oh I can?t smoke here, I can?t smoke there, I can?t smoke here. Well, if over here [referring to a place away from the doors but on hospital grounds] they allowed smoking, I think it would be better, put some seats out there, you know. Sit down relax. [PT16/A] He believed smokers had the right to some ?relaxation,? through smoking. For him, smoking was part of his coping and an individual right, which the SFGP infringed upon. Similarly, a patient in her 20s, admitted for depression, contested the inequality she experienced:  I don?t think it?s fair. We have enough stuff to deal with and enough med changes going on and enough withdrawal with different things in our body . . . cigarettes should be the last thing and it should be allowed. . . . I can?t just say hey ?can I go across the street and have a smoke?? Which is my right, if I choose to smoke . . . [it] shouldn?t be stopped. . . . But . . . they?re controlling it. [PT35/B] In this particular context, she revealed smoking as therapeutic in coping with the medication changes for her depression. The risks around smoking are relative in this scenario, and restrictions around smoking are understood as potentially harmful and undermining her efforts toward self-management of her depression.   At both sites, participants viewed the smoking policies as more relaxed in the smaller outlying communities from which many came. This was shown to affect the expectations of patients admitted to the psychiatric unit. A patient in her 50s, a non-smoker hospitalized for medication adjustments, stated: I live . . . [way] up north . . . and some of the bars there don?t feel that anybody?s going to walk in and complain. There?s very few people that live up there so they risk it and allow it. . . . There?s nobody there to police it so they just [allow smoking]. [PT38/B] Patients? expectations around smoking permissions reflected the standards of their individual communities of residence. Each of the research sites served large geographic areas, so a diverse set of community standards present in individuals meant that adjusting to SFGP     106 governance could be challenging. It also reflected how local community norms could amplify or reduce stigma associated with smoking. Conclusions Influenced by power, control, and stigma amid resignation and acceptance of smoking, the SFGP invoked an array of patient perceptions and actions. For patients who smoked, the central role of smoking in their lives related to group inclusion, long-standing relationships with smoking, their mental illness, and coping mechanisms. Smoking was most often referenced by patients as helping to cope with stress and hospitalization. Patients who were non-smokers endorsed these views and supported the continued presence of tobacco at psychiatric units. Because of patients? relationships and histories with smoking, they harboured beliefs that hospitals had a duty to accommodate those who smoked. Adequate FABs and smoking areas where patients could feel safe and comfortable were expected. Smoking areas were advocated to remedy litter issues stemming from implementation of the SFGP. For most patient participants who smoked, quitting was not a current concern, but some believed that HCPs held expert knowledge about when to quit, and some reported experiences where this help was not shared with them.  From patient perspectives, the SFGP was incorporated into a larger structure of rules and governance at the units. FABs were valued by patients, could accommodate smoking activity, and were key motivators in helping regulate patient compliance with the rules. NRTs were seen as poor substitutes for smoking, and the provision of NRTs viewed by patients as mere routine rather than as therapeutic. NRT handling often mimicked smoking activity and reinforced acceptance of smoking. Mostly compliant, patients were adaptive and creative if they sought to circumvent SFGP rules. Being identified as a psychiatric patient was stigmatizing, but sometimes afforded power to self-isolate and be left undisturbed to smoke on hospital grounds. Patients struggled at times to reconcile competing interests between self and others as they weighed rights and desires to smoke against broader community norms and protection of others from second-hand smoke. Often patients vacillated between understanding the SFGP and resisting the restrictions it entailed on smoking. Understandings of power, control, and the dual stigma of mental illness and smoking helped position patients? SFGP experiences. Patients also viewed governments and     107 tobacco companies with suspicion, and attributed malevolent intent to their actions relating to tobacco pricing and addiction. Individuals? actions, beliefs, and attitudes impacted SFGP enactment and were shaped by the multifaceted contexts in which they were located. These individuals? contexts included personal histories, group norms, institutional structures, socio-political structures, locale, and community norms.      108 The purpose of this ethnographic study was to improve understandings of how the implementation of smoke-free grounds policies (SFGPs) is affected by institutional cultures. The findings bring to the forefront the importance of understanding policy implementation within inpatient psychiatric settings and the cultural factors that influence this process. The experiences and perspectives of healthcare professionals (HCPs) and patients regarding the SFGP at two inpatient psychiatric units point to how local and specific contexts and cultures influenced policy implementation. Findings revealed prevailing attitudes and beliefs about tobacco acceptance and tobacco control that mirrored some of those previously reported in the literature (Lawn, 2004; Voci et al., 2010), while adding to existing results drawn from other research focused on SFGP implementation (Ratschen, Britton, & McNeill, 2008; Schultz, Bartmanovich, et al., 2010; Schultz, Finegan, Nykiforuk, & Kvern, 2011; Schultz, James, et al., 2010; Shetty, Alex, & Bloye, 2010). The current study offers new evidence about the importance of local context to SFGP implementation and reports for the first time how rurality may influence SFGP implementation in psychiatric settings in Northern British Columbia, Canada. The findings illustrated challenges faced by HCPs that were not necessarily anticipated by policy-makers, by revealing an array of strategies and justifications by which participants subscribed to as well as circumvented and subverted policy implementation. In this chapter, a conceptual spheres of influence model (McLeroy, Bibeau, Steckler, & Glanz, 1988) has been adapted to highlight the importance of cultural contexts in SFGP implementation. Key findings are discussed in relation to existing knowledge amid distilling interconnections between participants? beliefs and attitudes, the influence of group norms, leadership, and the influence of community in thoughtfully considering future SFGP implementation. The chapter concludes by detailing the limitations of the current study while making recommendations for what might prevail to improve policy implementation.  Chapter 6 -  DISCUSSION AND RECOMMENDATIONS     109 Importance of context Consideration of the findings in light of the research questions posed pointed to the importance of contexts to policy implementation. Aligned with the ethnographic methods used, conceptually, the SFGP was implemented through the actions and interactions of individuals within an array of social contexts. The study findings confirm the existing research and policy viewpoint (McKenzie & Wharf, 2010; Voci et al., 2010; Wu, Ramesh, Howlett, & Firtzen, 2010), asserting that understanding how context influences the actions, beliefs, and behaviours of individuals is key to effective policy implementation. In considering and encouraging others to broadly consider the matters that affect policy implementation, a socio-ecological model provided a means of helping advance the findings and represent the complexity of factors at work in policy implementation. The spheres of influence model?Figure 6.1?is provided to conceptually locate individuals as nested within contexts and offers a means of visualizing and layering analyses onto the study findings to illuminate a myriad of factors affecting SFGP implementation. The spheres of influence model is based on socio-ecological theory and an ecological model for health promotion developed by McLeroy et al. (1988). McLeroy and colleagues located patterned behaviour as an outcome of interest determined by five factors: 1) intra-personal factors, 2) inter-personal processes, 3) institutional factors, 4) community factors, and 5) public policy. Their model drew from the work of Bronfenbrenner (1977) who conceptualized individual development and behaviours as affected by the relationships individuals have with immediate settings and larger social contexts that change over time (the ecological environment).      110   Bronfenbrenner labelled the levels as micro-, meso-, exo-, and macro-systems. As in the work of Bronfenbrenner, fundamental to the current study findings is the understanding that individuals act within contexts and that individual behaviours can be viewed as both influencing and influenced by the inter-relationships of the aforementioned spheres of influence. While the spheres of influence model is illustrated with boundaries between the spheres, the sphere should be understood as deeply entwined and perhaps overarching in nature. There are elements of culture that cross through all of the spheres, influencing and influenced by many or all of the other spheres. For example, cultural norms including self-reliance and hardiness permeated all the spheres in specific contexts, synonymous with wider cultures of BC northerners. Intra-personal context: Individual beliefs and attitudes The long-standing prominence of tobacco on inpatient psychiatric units was sustained by enduring beliefs and attitudes that predated the current policy implementation and formed ENVIRONMENT (public policy, laws, socio-political) COMMUNITY (community norms, locale) INSTITUTIONAL FACTORS (policy, workforce, space) INTER-PERSONAL (GROUP NORMS) (unit culture, affiliations, social groups) INTRA-PERSONAL (INDIVIDUAL) (knowledge, self-concept, skills, actions/behaviours, attitudes, beliefs)  Figure 6.1  Spheres of Influence Adapted from: McLeroy, Bibeau, Steckle & Glanz, 1988     111 key components of the intra-personal sphere of influence. Others have observed that an individual?s knowledge, attitudes, beliefs, behaviours, and self-concept help define an intra-personal sphere (McLeroy et al., 1988), and this conceptualization was applied to help identify the factors relevant to the research questions asked in the current study. Intra-personal factors influenced the behaviours and interactions of HCPs and patients, and oftentimes beliefs and attitudes were co-constructed by members of these two participant groups. For example, patient beliefs that smoking helped manage some symptoms of mental illness were derived from personal experience and co-constructed through interactions with other patients and HCPs. HCPs held similar beliefs that manifested in a number of ways, including HCPs advocating for patients around their right to smoke. Prominent was how patients remained resigned to smoking largely because of established relations with tobacco and beliefs about the therapeutic value of smoking for them. The centrality of smoking was rooted in personal beliefs sustained by both HCPs and patients. For example, these patients? beliefs were driven by embodied experiences of feeling less anxious, self-distracted from symptoms, and a sense of purpose through activity and interaction associated with smoking and getting off the unit. These HCPs? beliefs were driven by better compliance from patients, observations of more amenable patient state, and less visible anxiety and agitation, all justified as preserving therapeutic relationships. Beliefs about the central role of tobacco in psychiatric settings lobbied against top-down SFGP implementation to suggest shortcomings in relying solely on hierarchical directives to implement SFGPs.  HCPs found it challenging to reconcile the day-to-day practice demands of fresh air breaks (FABs), nicotine replacement therapies (NRTs), and personal beliefs with fully subscribing to implementing the SFGP. Though HCPs at both sites regarded provision of FABs and NRTs as humane responses to patients? addictions to nicotine, the many associated routines and behaviours operated to support a culture of tobacco use. This left unanswered questions about how best to provide empathetic responses without reinforcing or being complicit in patient?s tobacco use, a tension that contributed to the varied HCPs? responses to implementing the SFGP. This dilemma was further complicated by tobacco confiscation activities and personal beliefs related to patient rights and the purpose and intent of the SFGP. Reduced physical handling of tobacco and tobacco-related paraphernalia was a     112 welcome change, but FABs required as much HCP management as the smoke breaks that preceded the policy change. For patients, FABs continued to facilitate smoking, and many chose not to utilize NRTs once smoking opportunities were made available.  HCPs and patients held mixed beliefs about whether the SFGP was about quitting smoking, improving health, or reducing harm. As a result, the SFGP was interpreted and acted upon differently. A key feature of policy implementation is ensuring that the goals align with the issues and problems to be addressed. The current study findings suggest that the goal of the policy was neither explicitly stated nor unitarily understood. In addition, because a primarily top-down approach to policy development and directives to implement the SFGP was used in the two study settings, the policy-makers? perspectives about the need for changes were not shared by many of those who were directly responsible for implementing the policy. With some success, education and dialogue increased at Site B shifting individual beliefs and attitudes to better align with SFGP implementation goals. However, HCPs and patients often adopted behaviours based on experiential or personal knowledge while discounting evidence-based knowledge, and this aided in maintaining the status quo. Additionally, the SFGP did help raise questions about the beliefs HCPs held and, for those who sought to maintain the status quo, it often stimulated self-reflection and perhaps interrogation about some of their current professional practices. Such reflexive practice might be a first step to changing HCPs? beliefs and attitudes toward tobacco use among people with severe and persistent mental illness (SPMI) (Miller & Rollnick, 2002).  The mixed beliefs about the intent of the SFGP supported interactions between HCPs and patients that were not necessarily congruent with the policy-makers vision for implementation. These interactions brought some stability, and new norms and routines to the units, but often times perpetuated a culture of smoking. A prime example of this was how FABs and NRT use were enacted, interpreted, and influenced by HCP and patient practices. FABs also invoked and perhaps preserved power differentials, in that they were code for smoking, and while NRT use addressed nicotine withdrawal, it could be used to mimick tobacco use in ways that were reminiscent of pre-policy interactions between patients and HCPs. Many of the beliefs and attitudes of HCPs revealed through the current study support other recent Canadian studies (Johnson, Moffat, & Malchy, 2010; Schultz, Bartmanovich, et     113 al., 2010; Schultz, James, et al., 2010). In supporting the status quo, HCPs demonstrated genuine concern for the health and welfare of patients, but endorsed the use of tobacco for therapeutic purposes, avoided tobacco conversations unless initiated by patients, and positioned tobacco cessation as residing outside their realm of expertise and thereby outside their role. Moreover, complaints of ?policing? or authoritarian roles that were advanced under the SFGP support findings from a study involving smoke-free policies in community mental health services in British Columbia (Johnson et al., 2010). In addition, beliefs that acute inpatient hospital stays were not the time to prioritize quitting, and that patients needed to smoke as self-management and coping mechanisms of underlying mental illness, support findings in other western Canadian adult psychiatric units (Schultz, James, et al., 2010). However, in the current study, in contrast to the Johnson et al. (2010) and Schultz, James, et al. (2010) studies, findings revealed how some HCPs were comfortable using NRTs to treat patients who could not smoke and that this was prompted by new concerns around clinical practice responsibility. Education and resourcefulness have enhanced HCPs? comfort and competency in use of NRTs. Diffusion is defined as a passive spread and acceptance of an innovation (here the SFGP) as distinguished from the active and planned efforts of implementation (Greenhalgh et al., 2004). Conceptually, diffusion helps link policy implementation to the informal processes within an organization (Denis, H?bert, Langley, Lozeau, & Trottier, 2002). Strong beliefs about tobacco use in mental health settings differed among individuals, some of which ran counter to the intent of the SFGP. According to Denis and colleagues (2002), this makes it difficult to map the policy (i.e., innovation) onto all of the values and beliefs of the individuals involved in implementation, thus confounding intended implementation plans (Kennedy, Lawless, & Slater, 2009). This finding suggests identification of the values and beliefs of those most directly involved with implementing SFGPs is pivotal to aligning informal and formal processes and improving implementation results, and identifying where values and beliefs might be shifted or policy amended. Additionally, ideological differences between policy-makers and others involved in policy implementation can divide acceptance of the policy (Denis et al., 2002; Kennedy et al., 2009; Langley & Denis, 2011). In the current study, implementation of the SFGP was rooted in advocacy for cessation based on     114 empirical knowledge about the cause?effect of smoking on specific diseases and based on top-down approaches. However, study findings around advocating for and maintaining rapport with patients, and other humanistic actions, suggested that social contexts in providing care in psychiatric settings routinely trumped strictly medical approaches.  Patients were influenced by knowledge from various sources. Political and social attitudes found in patient environments impacted patient beliefs and attitudes about tobacco use, and most found that tobacco use in face of stricter tobacco use laws and regulations contributed to stigma. In some, this stigma reinforced negative beliefs about self and acceptance of tobacco. Findings in the current study advise against stricter tobacco controls in acute inpatient psychiatric units as the conduit to patient smoking cessation. While supportive of aims to reduce health risks to others implementing the SFGP, patients generally endorsed the use of tobacco in those with mental illness and often advanced individual rights arguments to support the right to choose to smoke. Individual rights arguments and increased stigma associated with the visibility of smoking as contraband resulting from the SFGP contributed to patients? advocating for provision of designated smoking areas, as was the case in another Canadian study (Schultz, Bartmanovich, et al., 2010). Beliefs and attitudes of individuals importantly influenced their behaviours and impacted SFGP implementation in the current study.  Inter-personal context: The influence of group norms Group norms, factors in the inter-personal sphere of influence, were significant in understanding participants? experiences with implementation of the SFGP and tendencies toward resisting or subverting the policy implementation. Some HCP subversion of the SFGP was evident at both sites. Prior and Barnes (2011), in social policy analysis, have conceptualized three types of subversion that were evident to varying degrees in the current study: (a) a re-interpretation of policy to specific contexts to subvert the intent of the policy with ?official? compliance, but unintended outcome, (b) use of alternate strategies and practices developed through individual or group processes where such practices were initially intended to be outside of the policy intent, and (c) a passive type of resistance that is tantamount to outright refusal to engage in actions consistent with the policy. For example, in the context of the current study and the SFGP, HCPs drew limits on enforcement of the     115 policy when it was assessed as detrimental to the management of or rapport with the patients, and where violent reactions might emerge. Subversion was seen in actions and conversations that coded ?FABs? to mean ?smoke breaks,? and HCPs also used these privileges instrumentally to affect behaviours among patients who smoked. Subversion was also observed in HCP holding-dispensing actions to regulate cigarette consumption for some patients. Finally, resistance was evident in actions and cultures of ignoring. Common to all these subversive actions was insider knowledge. Prior and Barnes (2011) have termed these types of resistive actions as ?counter-agency? or ?agencies of resistance,? suggesting they can occur where policies permit individuals implementing the policy opportunity to choose an alternate course of action that has not been fully prescribed by the policy. This ?freedom to choose? permeated findings in the current study where HCPs and patients interacted to subvert the SFGP. This added complexity to implementing the policy, from both HCP and patient perspectives, and sometimes obfuscated communications between these groups and management. SFGP implementation was in effect mediated by and negotiated within and between patients, HCPs, and the organization through group processes. Strong consultative leadership was a resource that assisted HCPs in making professional practice decisions to balance patient care with patient management, while addressing an inability to prescribe actions for all circumstances arising under the SFGP.  How HCPs worked to preserve the status quo illustrated the importance of group cohesiveness. Findings indicated that displacing HCPs? views about tobacco use among patients with SPMI was contingent on shifting group norms. Education and staff development can aid in changing group norms necessary to replace views about the therapeutic value of smoking for mental illness symptoms with evidence-based information to inform practices (Williams et al., 2011). Education and role modelling by leaders were used at Site B, but revealed partial success in disrupting cultural norms that blocked implementation of the SFGP. That said, there is some evidence that, over time, simply implementing a SFGP assists in shifting HCP norms espousing acute inpatient stays as the wrong time to intervene to reduce patient tobacco use (Ashton, Lawn, & Hosking, 2010; Voci et al., 2010). Variations in the implementation of the SFGP between the sites in this study suggested that the fixed nature of pre-policy beliefs about tobacco use, affected by     116 group norms, also differed and that multifaceted approaches and attention to local contexts are needed when implementing SFGPs.  Inter-personal context: Leadership and consensus building Leaders influenced the inter-personal interactions around SFGP implementation and were an important conduit of information and influence within the organizational structure. In viewing the SFGP as an organizational change, the importance of effective leadership is consistently highlighted in the business and management literature and more recently with respect to SFGP initiatives in Lawn and Campion?s (2010) study. Participants? frequent references to group membership suggested those leading policy implementation must pay careful attention to social identity and group processes. Power imbalances reflected inter-personal factors involving groups and, in the current study, implementation power favoured the HCP group above patient groups. Consistent with the research of Lawn & Campion (2010), policy leaders in the current study who successfully fostered team cohesiveness and consistency in application of the policy, and who facilitated staff education that supported HCPs negotiating many of the inter-personal challenges associated with the SFGP implementation, were most likely to support policy implementation successes. Strong formal leadership at Site B worked toward consensus-driven, team approaches that focused on consistent approaches to tobacco control that aligned with the provisions of the SFGP. Strong consultative leadership, including input and participation by those most directly responsible for policy implementation, offered the strongest indication that policy-maker intent could be implemented. Managers at Site A were less effective than those at Site B at leading SFGP implementation, and they failed to adequately check and engage the HCPs? group processes, a situation that gave rise to both implicit and overt HCP resistance to the policy. Some change management literature suggests that finding early adopters and champions to locate intentional, informal leaders (Greenhalgh, Robert, MacFarlane, Bate, & Kyriakidou, 2004) is an effective means to supporting new policy implementation. However, even these early adopters and champions require direction and management to ensure their goals align with those of the policy-makers. Managers at Site A might have capitalized on HCP group processes had they located and supported HCPs who were early adopters and champions of the SFGP. A combination of strong leadership and management approaches may aid those     117 involved in SFGP implementation to find the best approaches. Current study findings suggest that top-down and bottom-up communication are key to garnering strategies toward improving the fidelity and ultimately the implementation of the SFGPs, as previously reported by Wu et al. (2010).  Leadership qualities that may improve implementation efforts of SFPGs in psychiatric settings can be abstracted in reviewing the current study findings. Specifically, the most effective leaders were committed to tobacco reduction efforts, championed the SFGP, were well-educated on the subject, and were formal in their leadership style yet attuned to the front-line challenges. They also focused on consensus building, consistently demonstrated active engagement with HCPs around the issues of the policy, provided education to HCPs, and role modelled and supported problem-solving efforts while addressing resistance in informed and productive ways to build momentum for SFGP implementation. These leadership qualities could also effectively influence other wards and ward leaders to build organizational momentum. Strong relationships between leaders and HCPs required reciprocated trust, a finding supported in the literature (Brooks, Pilgrim, & Rogers, 2011). HCPs recognized effective leaders as supportive and connected to the everyday practice challenges they faced in front-line implementation of SFGPs. The challenges that arose when broad SFGP compliance was not achieved across the organization have not been previously reported in the literature. In addressing these challenges, HCPs sought counsel with senior management and policy-makers to evaluate the organization?s true commitment, which in turn influenced the amount of energy they expended on fully implementing the policy. Findings in the current study, for example, revealed challenges around patient transfers to the psychiatric units that suggested HCPs lacked broad organizational commitment to implementing the SFGP. Discordance between official policy documents and post-implementation management actions was perceived by HCPs who observed low implementation rates of the policy outside of their own units. Patients also recognized and reported inconsistencies in the application of the SFGP in other areas of the respective hospitals. Site B HCPs reported unexpected tensions between staff on the psychiatric unit who were attempting to implement the policy and those staff in the rest of the hospital who tended to ignore the SFGP, a finding     118 supported by previous studies (Schultz, James, et al., 2010; Shetty et al., 2010). It is possible that the health region and its hospitals did not more fully embrace the SFGP because the organization lacked individuals who could champion the policy and span the boundaries of various hospital units and hospitals. There is support for this explanation in the business literature (Greenhalgh et al., 2004) and in related observations in the current study. For example, HCPs reported dissatisfaction with a perceived lack of senior management leadership to drive the policy organization-wide. Senior management was perceived absent in guiding between-unit and between-hospital conflicts that arose under the SFGP. This is not to say that leadership was absent, but rather its efficiencies and influence were perceived by HCPs as ineffectual at the unit level. Without front-line workers perceiving the authentic investment, commitment, and empathy of senior management for the mechanistic challenges to implementing the SFGP, there may be minimal effects toward mobilizing the energies of those charged with implementing the policy. In the current study as in other studies (Lawrence et al., 2011; Ratschen, Britton, & McNeill, 2009; Schultz, James, et al., 2010), HCPs significantly influenced SFGP implementation and therefore required the most attention and support. SFGP implementation efforts are likely enhanced through committed actions of senior management that are visible and supportive of HCPs in addressing the everyday professional practice challenges associated with the policy. Addressing the challenges of broad acceptance of the SFGP within the organization has strong potential to improve implementation of SFGPs in inpatient psychiatric settings. This is also an area where managers with influence beyond the acute psychiatric settings can act to support front-line HCP efforts. HCPs and patients were affected by the actions and attitudes of employees from other hospital units and provided examples of the interplay between those within the unit, other employee groups, and institutional factors that held particular relevance to psychiatric units. Diverse levels of commitment to policy implementation by policy-makers, managers, HCPs, and patients were evident. Patients and HCPs referenced inconsistencies in policy implementation as unfair. Prominent at Site B was the inter-personal and institutional segregation that HCPs and patients experienced. Patients characterized this segregation as related to stigma and sometimes leveraged this to contravene the SFGP. As seen in discussions about the challenges of garnering broad     119 compliance and genuine commitment, HCPs who strictly enforced the SFGP risked ostracism and overt challenges from hospital employees working outside of the psychiatric unit. Prominent at Site A was the inter-personal segregation of nursing HCP groups from other HCP groups and front-line management that hampered hospital-wide policy implementation. Reducing the separation of inpatient psychiatric units from the rest of the organization and promoting cultural norms of inclusiveness across organizational units may support broader acceptance of SFGPs, may better support HCPs working in acute psychiatric settings, and may relieve some of the professional practice challenges faced by HCPs implementing SFGPs on inpatient psychiatric units. Diverse and dislocated approaches to the SFGP across the organization were reported by HCPs and patients. Building uniformity of approach has been viewed by some as a form of social movement that includes eliciting support through coalition building, collaborative activity, and group cohesiveness, collectively described as democratic processes to change (Dixon-Woods, Amalberti, Goodman, Bergman, & Glasziou, 2011). Some of these democratic processes were found at the unit levels in the current study, but the inpatient psychiatric unit staff reported limited success in coalition building with other units around the SFGP. Coalition building is understood to be temporary unions between two or more groups for joint action (Merriam-Webster, 2003). Dixon-Woods et al. (2011) suggested that examining bureaucratic structures and professional groupings may hold clues to understanding why coalition building fails within organizations.  The importance of team cohesiveness to the organization was reflected in some of the wording of the SFGP. Team cohesiveness references the degree to which members were attracted and motivated to maintain membership through positive relationships with group members (Wendt, Euwema, & van Emmerik, 2009). In the current study, team cohesiveness and workload were affected by factors including unit histories rooted in changes to institutional mandates and policies. Site A provided an example of how pre-existing workplace challenges unrelated to the SFGP undermined confidence in formal power structures that resulted in resistance to policy change. Front-line managers also reported how divisions between employees were created by the way labour was organized into departments or functional areas of practice within the hospitals, and this contributed to the diverse     120 approaches to the SFGP seen across the hospital. Where implementation of the SFGP varied within the organization, management structures appeared to be divided and oftentimes divisive. The spheres of influence model is rooted in ecological models and ecological strategies for policy development, and implementation advocates consensus building (McLeroy et al., 1988). Involving patients and HCPs in simultaneous processes of top-down, bottom-up policy development and implementation offers opportunities for consensus building that may improve policy fidelity. Consensus-building approaches promote policy ownership, provide health education, and potentially shape individual understandings about the cause and responsibility for health and illness while empowering participants to do something about it (McLeroy et al., 1988). These are some of the factors that HCPs reported as influencing their diverse approaches and overall commitment to the SFGP. Patients reported marginalization and power imbalances. Increasing patient and HCP involvement in SFGP processes through consensus building may help offset various imbalances, by recognizing the legitimacy of pluralist value sets in shaping policy implementation and fidelity (Langley & Denis, 2011). At Site B, HCPs had increased input into SFGP implementation and a stronger sense of policy ownership than did HCPs at Site A. Increasing involvement in the development and implementation of SFGPs by those most directly affected by it may have off-setting gains in implementation results.  Patients indicated that additional potential gains could accrue from bottom-up policy implementation. Patients were accepting of the need to protect others from the harmful effects of second-hand smoke, so an inpatient psychiatric unit without smoking controls would run counter to these beliefs. Enhancing patient involvement with SFGP development and implementation might help them locate any ambiguity between their own goals around second-hand smoke reduction, resistance to SFGPs, and smoking, and this may support changes in patients? attitudes and beliefs about smoking that can improve SFGP implementation and patient choices around smoking and improved health (Miller & Rollnick, 2002). Organizational policy development involves inter-personal relations and consensus building, and then this policy becomes part of the institutional structures that help regulate     121 employee actions, thus illustrating one of the interactions between inter-personal and institutional factors. Institutional factors: Organizational change  The organizational change literature helped me to position the current study findings within a policy implementation discussion at an organizational level. A clustering of hospitals into a health authority creates structures that influence the norms about the work of health authorities and their ability to manage and lead. Policy is the organizational mechanism of management and leadership, and the introduction of the SFGP is an important change for the organization.  Change management literature supports the importance of examining contexts in policy implementation (Greenhalgh et al., 2004) as does the literature in clinical innovation (Powell et al., 2012). Defining organizational culture as the context affecting change is a common approach found in the literature; however, conceptualizations of organizational culture often fail to recognize how smaller groups or microcosms within organizations manifest cultural differences, instead focusing on the commonalities that cross the entire organization to essentialize or define a singular organizational culture (Jung et al., 2009; Parmelli et al., 2011). The current study findings can inform policy-makers of the limitations attached to examining SFGP implementation from a singular organizational culture perspective, an approach that is problematic because it obscures locale differences as helping explain diversity in SFGP uptake. Brooks, Pilgrim, and Rogers (2011) suggested that macro-level approaches that focus on whole system or changes to organizational cultures offer limited ability to register and account for context and implementation at lower levels. The current study revealed how effective SFGP implementation relied on the lower levels? buy-in and how a unitary organizational culture did not always transcend these levels to prevail in psychiatric clinical practice cultures. Policy implementation that assumes or attempts to drive a singular homogeneous culture in a large and diverse organization may underestimate the cultural reality, as was shown in the current study. While broad acceptance of the SFGP across the organization was desired, policy-makers and executive hospital managers at institutions where inpatient psychiatric units reside amid non-psychiatric units should anticipate heterogeneous cultures when implementing SFGPs.      122 Models of organizational change examining adoption of innovation through constructs of an organization?s implementation climate and innovation-values fit (Klein & Sorra, 1996; Weiner, Belden, Bergmire, & Johnston, 2011) conceptually place individuals in a similar fashion to the spheres of influence model. A widely accepted definition of implementation climate is the extent to which organization employees perceive the use of an innovation as ?rewarded, supported, and expected within their organization? (Klein & Sorra, 1996, p. 1060). Innovation-values fit suggests that employees will more readily commit to using an innovation based on its perceived fit with their own values (Klein & Sorra, 1996). Weiner et al. (2011) suggested that the concept of innovation-values fit as an implementation factor is popular with mental health and addiction researchers, and the current study supports it as a possible theory to help explain diversity in SFGP implementation. For example, Site A and Site B HCPs expressed different views about the reward attained and support provided in implementing the SFGP implementation, and values-fit to the SFGP was challenged by HCPs and patients at both sites. However, examining SFGP implementation from strictly an implementation climate or innovation-values fit viewpoint assumes something of a macro-level approach that risks over-simplifying policy implementation by neglecting the importance of localized contexts.   The SFGP may be justified as an evidence-based practice innovation. Denis and colleagues (2002) found that evidence-based practice innovations were not always well defined as objects to be adopted. Supporting those findings, HCPs in the current study did not view the SFGP as a clearly defined object. Here, complexity and challenges in SFGP implementation involved the nature of the policy itself. In applying Denis and colleagues? (2002) conceptualizations, there is a hard core and soft periphery to innovation adoption. The hard core ?is the element that is irreducible and that carries the key potential benefit? and the soft periphery ?a gamut of complementary arrangements that are involved in delivering the benefit that may take a variety of different forms? (Langley & Denis, 2011, p. i45). In the current study, the goal of the SFGP was to reduce second-hand smoke exposure of those utilizing hospitals (the hard core), but the SFGP can be implemented in a variety of ways (the soft periphery). The soft periphery is where context is located (Denis et al., 2002), creating various pathways of implementation that led to selective adoption of the SFGP. For example,     123 new HCP practice competencies around managing patient behaviours without the reward of cigarettes and finding ways to effectively negotiate patient challenges to the SFGP were part of the soft periphery of the policy that help account for individual and group differences in implementation. Some HCPs drew on pre-existing practices and cultures to substitute FABs for smoke breaks, while others developed new competencies toward supporting tobacco reduction and smoking cessation. The periphery also includes the political systems within the hospital, and these must be understood for their effects on policy implementation efforts (Langley & Denis, 2011). The soft periphery cannot always be reduced, but it needs to be better understood and incorporated into implementation strategies. As part of an institutional approach, the SFGP was primarily communicated through a top-down approach from policy-makers that relied heavily on formal organizational structures of power and accountability for implementation. A top-down approach is characterized by policy-makers working to transfer a clearly articulated intent downward through the organization to control outcomes at the grassroots or front line (Wu et al., 2010). The local contexts of HCPs and patients were revealed as influencing top-down approaches and part of the complexities involved with tobacco reduction and control in inpatient psychiatric settings, findings that are supported by the existing literature (Lawn, 2011; Lawn & Condon, 2006). The influence of local contexts revealed in the current study suggests that simple top-down approaches to changing the presence and control of tobacco in inpatient psychiatric settings, particularly where there are challenges of broad acceptance and HCP support, are insufficient to make significant progress in tobacco reduction and associated health improvements. Similar to what others (Johnson et al., 2010; Poland et al., 2006) have conceptualized as the social contexts of smoking, bottom-up approaches involving patients and HCPs may prompt all involved with SFGP implementation to [re]consider their relationship, power, and social location in relation to those who smoke to afford alternate worldviews cognizant of the complex contexts in which individuals are located. Re-thinking SFGP implementation suggests increased involvement and support of those most directly involved with implementation to better capture and respond to the complexities and contexts influencing implementation processes.      124 Community: Locale-specific norms  Community may be defined in a number of ways, including those primary groups with which individuals engage directly, relationships among organizations and groups within a local area, or by geographic or political terms (McLeroy et al., 1988). Environment includes public policies, laws, and the socio-political factors affecting or affected by that contained in the inner levels of the spheres of influence. The influence of community and environment, though conceptually distinct in the spheres of influence model, are collapsed and combined here for two reasons: firstly, a complete and separate analysis of the factors within the environmental sphere were limited by the data available; and secondly, with a few exceptions, participants in this study perceived environmental factors as either localized or contained within their perceptions of community. Two sites situated in unique geographical and demographic contexts provided clues about what locale-specific community and factors influenced SFGP implementation in inpatient psychiatric settings. Current study findings indicated that individuals were more responsive to, than directive of community factors. For example, patient responses to policy, perceived limits on HCPs? practice autonomy, and HCPs? responses to scarcity of resources revealed how community and environment could affect SFGP implementation. The socio-political aspects of community influenced the actions of individual participants and the healthcare organization.  In Northern BC, local contexts revealed the importance of community standards, how smaller communities made the socially unacceptable more visible and the impact more profound; how HCPs considered community standards when making practice choices; and how conflict was avoided as part of finding harmonious ways to live within smaller communities, all supporting other Canadian study findings (Kulig et al., 2008; MacLeod et al., 2008). Not previously reported in the literature, however, were how hospitals that were viewed by HCPs as more rural than their own were reported as less likely to comply with the SFGP because small communities were more likely to tolerate and accept smoking. Northern BC comprises many small communities, so this finding might help explain why their smoking rates are above the provincial averages. In the current study, most HCPs were limited to one employer and in the case of inpatient psychiatric units, one work site in the local community. This helped explain a reluctance of HCPs to enforce the SFGP in the     125 hospital more generally, the reluctance of individuals to challenge a breach of policy, and individual need to find acceptance and group membership. Following community norms was part of daily life in smaller, northern communities, and failing to find individual fit and harmony with community may result in an individual being ostracised and othered. Frequently reported concerns about litter issues in the current study reflected how community norms and socio-political environments shaped individuals? attitudes and beliefs. Concerns of litter issues as a result of SFGPs have been reported elsewhere (Ratschen, Britton, et al., 2008; Schultz, James, et al., 2010; Schultz et al., 2011). Defining the nature, boundaries, and implications of community size may provide policy-makers with information about these influences on implementing policy. Consistent with the literature, communities and the socio-political environments were mediating structures of individual actions in policy implementation (McLeroy et al.). Evident in Site B HCP approaches to their work, these smaller communities also demonstrated a resourcefulness and self-reliance that sometimes translated to strengthening group cohesiveness and desire to meet or exceed performance expectations of their employers. The nursing literature supports findings of resourcefulness and self-reliance in Canadian rural settings (Kulig et al., 2008; MacLeod et al., 2008) and may represent opportunity to strengthen policy fidelity if policy and group values align.  In an Australian study involving various types of psychiatric inpatient units, researchers categorized 34% of the urban and 45% of the regional sites as successful in implementing smoke-free policies (Lawn & Campion, 2010). This points to some underlying capacity in regional psychiatric inpatient environments to successfully implement SFGPs. While, intuitively, one might suspect that unit size influences the ability to build cohesion and consistency, Lawn and Campion found that unit size was not a factor. Of the two research sites in the current study, the site in the more rural setting achieved SFGP implementation that more closely matched policy-maker intent. Geography may be an important factor affecting policy implementation, and the findings from the current study are in line with Lawn and Campion?s results. One artefact of smaller community services at Site B was that recreational therapists worked with inpatient populations on site as well as when they transitioned back into the local community. This consistency of care allowed a subgroup of HCPs to develop     126 discussions about tobacco cessation during admission, and then follow-up with patients in the community. This type of follow-through to community may bolster efforts toward SFGP implementation and goals of reduced smoking among those with severe mental illness (Lawrence et al., 2011; Schroeder & Morris, 2010). In the current study, this continuity was only available to those who lived in the community in which the hospital was located. In harbouring both advantages and disadvantages for health care services, the rural environment confirmed patient access inequities found in other studies in Canada and the U.S. (Hunter, 2006; Newhouse & Morlock, 2011; MacLeod et al., 2008). A continuity of care from acute to community settings that includes tobacco reduction supports and strategies may improve acceptance and implementation of smoke-free policies across these settings. The importance of examining the social factors contained within environments where those with severe mental illness interact with the healthcare system is thought by some as important (Williams et al., 2011). These authors posited that ?tobacco-free environment[s] support cessation efforts of individuals and also effect culture change by establishing new accepted norms? (p. 374). As a proposed state-wide initiative, Williams et al. suggest a necessity to change community standards that redefine ethical care by displacing a focus on autonomy and ?right to choose? with obligations to support good health practices through smoking cessation. The same emphasis was heard in accounts of some HCPs in the current study, and elements of change were emergent in the ?new? discussions about tobacco use and the responsibility of HCPs. In summary, SFGP implementation requires complex changes and innovation. Greenhalgh et al. (2004), in conducting a literature review on diffusion of innovation amongst service organizations, offered the summary that ?context and ?confounders? lay at the very heart of the diffusion, dissemination, and implementation of complex innovations? (p. 615). They used the example of a champion as a key determinant of innovation, highlighting that empirical research cannot ?provide a simple recipe for how champions should behave that is independent of the nature of the innovation, the organizational setting, the sociopolitical context, and so on? (p. 615). The current study affirms champions as strong determinants of successful SFGP implementation, but perhaps more importantly revealed are empirical insights to the complexities for catalyzing those efforts toward policy development     127 and implementation. Espousing a single recipe to SFGP implementation is somewhat na?ve in this regard. Although the tendency of policy-makers is to adopt top-down approaches to tobacco control policies in areas that have entrenched cultures of tobacco use and acceptance, bottom-up approaches that reflect local context and simultaneously support development and implementation of these policies may more readily influence necessary shifts in cultures, attitudes, and behaviours. Simultaneous top-down, bottom-up approaches seem most likely to engage HCPs in implementation efforts, and distilling the intricacies of garnering this dialogue and collaboration are important next steps.  Limitations  Although the current study afforded a rich cross-sectional account of the cultures and experiences of the participants, it is limited to the observations, interviews, and documents that were analyzed. The differences in the study-site contexts shaped the social, work, and organizational cultures, and the locale specificities limit what can be claimed as being relevant to other geographies and organizations. Participants were selected with the assistance of HCPs, and it is possible that some potentially alternate experiences were not captured in the data or described in the findings of the current study. Diverse approaches to SFGP implementation were buoyed by departmental boundaries, organizational structures, and varied sets of values and group norms within and between the hospitals within the same organization. However, the current study cannot account for pre-policy implementation efforts nor the processes underpinning current practices in the two study sites. How HCPs responded to the new SFGP may also reflect a generalized response to policy changes in their environments. Recommendations Health policy studies can inform policy-makers about important values, beliefs, and contexts to refine implementation strategies. Further study is needed to improve SFGP efficiencies and successes. Studies involving bottom-up policy development and policy implementation strategies may help policy-makers in structuring policy provisions while incorporating local contexts and complexities. Bottom-up approaches to policy development and implementation also offer opportunities for participatory-type interventions, to shift     128 values and beliefs around tobacco use to improve policy implementation in psychiatric or other hospital settings. While the findings from the current study revealed some evidence of involvement HCPs reported feeling removed from the formal policy development and implementation processes. Moreover they suggested that they had little input to management?s implementation plans. This is not to say that HCPs were not consulted in the lead up to the SFGP implementation, but HCPs by and large perceived their views as having little influence. Increased involvement from, and process improvements  involving those most directly affected by or affecting policy implementation is likely to increase SFGP implementation successes within psychiatric settings. Studies pointing to the importance of champions of innovation also provided clues to important leadership qualities and characteristics that warrant further study in the contexts of psychiatric settings (Brooks et al., 2011; Lawn & Campion, 2010). Grounded theory methods could be used to help define key processes and their relative importance to SFGP implementation in psychiatric settings more generally. Further examination of organizational structures and inter-personal processes within healthcare organizations may inform policy-makers and managers about how best to promote unity across departments around complex policies or how to better develop and align policies within organizations to permit SFGP implementation in pluralistic workplace settings. The current study findings suggest that SFGP implementation research include the influence of contexts on component parts of organizations. Further studies exploring the presence and predominance of medical versus social approaches to patient care, from both patient and HCP perspectives, may demonstrate these ideologies as important factors to include in a framework to assist with policy implementation (Langley & Denis, 2011). Measuring how different ideologies relate to uniform adoption of healthcare institutional policies may be a fundamental factor in understanding policy implementation. There is also indication that geography influences or reveals factors affecting SFGP implementation. Further research into SFGP implementation in rural settings can help identify factors that differ from those in urban settings, and comparison and contrast methods may help determine the relative importance of known factors or reveal those that are unknown.      129  Research utilization and innovation adoption research methods and studies might be adopted and replicated to inform SFGP and other policies implementation in psychiatric settings. For example, researchers might consider replication in psychiatric settings of a recent study examining the Alberta Context Tool in a pediatric/neonatal setting that suggested research utilization was influenced by three dimensions of context: culture, leadership, and evaluation (Cummings, Hutchinson, Scott, Norton, & Estabrooks, 2010). Finding ways to identify relevant elements of context and their relative importance to implementation efforts can enhance implementation of new policies. Quantitative methods can be deployed to help identify the most important variables (factors) and to locate the relationships between these variables and policy implementation outcomes.  The current study leads to various suggestions about HCP practice in supporting and promoting smoking cessation. Findings of this study suggest that HCPs working in inpatient psychiatric settings can develop comfort levels and competencies in effectively using NRTs, and further studies are needed to identify the factors that contributed to this. These studies should then be extended to examining if these same factors have positive influence on NRT use in other hospital settings. Practice-based norms that define HCP roles to largely exclude smoking cessation efforts in those with SPMI who smoke reflected culturally defined roles found in the literature (Williams, Ziedonis, et al., 2009). Locating and shifting these norms requires attention to the contexts in which HCPs practice, and examining processes that affect the ecologies of HCP practice is worthy of further investigation. Ecologies of practice include individual and collective experiences, practices that professionals accumulate and learn in their roles (Stronach, Corbin, McNamara, Stark, & Warne, 2002), and also the socio-political influences and interactions that surround HCP practices, including social interactions, place, organizational structures, and community and broader environments. Examining processes in ecologies of practice offers multiple points of potential intervention and influence that may enhance policy implementation. In the current study, HCP practices and decision making were illustrated as complex and influenced by a wide variety of factors, not simply education about best practices to reveal the need to further understand the social influences at play (Langley & Denis, 2011) and the processes of ecologies of practice.      130 Until cultures of tobacco use change, HCPs in inpatient psychiatric settings implementing SFGPs require long-term commitment of resources and supports that match the increased demands on their professional practices that come with SFGP implementation. Nurses described new challenges in monitoring smoking behaviours, controlling tobacco products, and managing NRTs that required additional energy, negotiation, learning, and enforcement efforts without commensurate increase in resources. These efforts likely need support of additional human resources, consistent support of management at all levels across all organizational divisions, and perhaps reorganization of individual HCP practices. Given the strong representation of nurse participants in the current study?s HCP sub-sample, some specific nursing practice recommendations are suggested. Best practice nursing care environments include policies and policy implementations that recognize the pluralisms of context, groups, and individuals. In this regard nurses, as the HCP group most directly affected by SFGPs, should be credentialed to support individual knowledge acquisition, addiction treatment, and care planning. Important opportunities to advance patient health and to change individual nurse?s practices around issues of smoking need to transition toward evidence based practices rather than inertia driven group norms and generalized beliefs. Changing group norms can be challenging, as can be accurately interpreting and lobbying contemporary ?evidence? toward patient care.  That said, nurses as a collective have the power to implement SFGP?s as a means to ensuring patient equity and fairness. Consistency in actions, based on evidence-informed clinical decision-making is no doubt key in managing tobacco on inpatient psychiatric units. Nurses can champion ?new? group norms and future longitudinal research might examine such transitions within an ecologies framework to guide other policy interventions. This study adds understandings about what factors might be included in intervention modeling. Patients issues including inequity, stigma, discrimination, and reduced autonomy signals the need for nursing practice to focus on individual patient?s strengths. Specifically, nurses might communicate the use of NRTs and FABs as therapeutic actions around addiction rather than awarding privileges. Nursing education can go part way to ensuring nurses assess for and understand what influences and shapes individual?s smoking behaviours, and how local contexts influence and are influenced by  attitudes and beliefs about smoking. Practice environments should also provide opportunities for nurses to     131 reflect upon their own beliefs and practice environments to support patients in their overall self-health and well-being. How best to muster reflective practice is an important next step in nursing research where SFGPs and other polices are implemented. How to best support the acquisition of these competencies and how to encourage nurses working in psychiatry settings to better utilize evidence based practices also suggests further research. Policy-makers are encouraged to consider that policies require fine-tuning to meet locale specific contexts and that increased understanding of the plurality of work and policy environments will inform the details about how (and how not) policies are implemented. No single or unitary organization or unit culture exists to ensure successful policy implementation. Conceptualizing and paying attention to the array of cultural factors that influence policy implementation can be mapped through the spheres of influence model to help devise context-specific strategies. Geographies, rather than unit size also need attention whereby implementation strategies are understood as locale specific as a means to improving policy fidelity. Management guided, bottom-up approaches to policy making and implementation inclusive of  site-specific, local contexts, while supporting a process of policy buy-in can also inform the need for discrete culture and/or policy shifts. In cases like smoke-free policies where top-down policy development may be required, room still exists for locale-specific implementation strategies. Finally, to ensure sustained change among cultures of tobacco use the SFGP implementation must be understood as a process rather than mandating a temporal end-point for its uptake.  This study is the first to report HCP and patient experiences related to the implementation of a SFGP in inpatient psychiatric settings in Northern BC, and the first to report in detail implementation experiences from two geographically separate psychiatric units within the same organization. This permitted an element of contrast between two units that helped isolate the importance of local contexts to SFGP implementation efforts. This study also adds to the systematic findings that report HCP and patient perspectives about the implementation of SFGPs in inpatient psychiatric settings. This study contributes to improved understandings of the challenges faced by those implementing SFGPs in inpatient psychiatric settings and informs management and policy-makers of potential factors important to improved policy fidelity. While further study is suggested, some immediate     132 steps can be taken to improve SFGP implementation that recognize the importance of local contexts and that support the contributions and involvement of those most directly involved with SFGP implementation.       133 REFERENCES American Psychiatric Association. (2006). American Psychiatric Association practice guidelines for the treatment of pscyhiatric disorders: Compendium 2006. Arlington, VA: Author. Anderson, J. M. (1991). Reflexivity in fieldwork: Toward a feminist epistemology. Journal of Nursing Scholarship, 23(2), 115?118. doi:10.1111/j.1547-5069.1991.tb00654.x Anderson, J., Adey, P., & Bevan, P. (2010). Positioning place: Polylogic approaches to research methodology. Qualitative Research, 10(5), 589?604. doi:10.1177/1468794110375796 Appelbaum, P. S. (1995). Do hospitalized psychiatric patients have a right to smoke? Psychiatric Services, 46(7), 653. Ashton, M., Lawn, S., & Hosking, J. R. (2010). Mental health workers' views on addressing tobacco use. Australian & New Zealand Journal of Psychiatry, 44(9), 846?851. Aubin, H. J., Rollema, H., Svensson, T. H., & Winterer, G. (2012). Smoking, quitting, and psychiatric disease: A review. Neuroscience and Biobehavioral Reviews, 36(1), 271?284. Barr, A. M., Procyshyn, R. M., Hui, P., Johnson, J. L., & Honer, W. G. (2008). Self-reported motivation to smoke in schizophrenia is related to antipsychotic drug treatment. Schizophrenia Research, 100(1-3), 252?260. BC Stats. (2007). Summary of smoking rates for BC: September 2006?August 2007. Victoria, BC: Author. Retrieved from BC Stats. (2009a). 2006 Census fast facts. Victoria, BC: Author. Retrieved from BC Stats. (2009b). Health Authority 5?Northern statistical profile. Victoria, BC: Author. Retrieve April 4, 2011 from BC Stats. (2011). Quarterly regional statistics: Fourth quarter 2010. Victoria, BC: Author. Retrieved from Benowitz, N. L. (2008, online). Clinical pharmacology of nicotine: Implications for understanding, preventing, and treating tobacco addiction. Clinical Pharmacology & Therapeutics, 83(4), 531?541. doi:10.1038/clpt.2008.3 Bigos, K. L., Pollock, B. G., Coley, K. C., Miller, D., Marder, S. R., Aravagiri, M., . . . Bies, R. R. (2008). Sex, race, and smoking impact olanzapine exposure. Journal of Clinical Pharmacology, 48(2), 157?165.     134 Bollman, R. D., & Clemenson, H. A. (2008). Structure and change in Canada's rural demography: An update to 2006. Ottawa, ON: Statistics Canada. Retrieved from Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. American Psychologist, 32(7), 513?531. Brooks, H., Pilgrim, D., & Rogers, A. (2011). Innovation in mental health services: What are the key components of success? Implementation Science, 6(1), 120. doi:10.1186/1748-5908-6-120 Campion, J., Lawn, S., Brownlie, A., Hunter, E., Gynther, B., & Pols, R. (2008). Implementing smoke-free policies in mental health inpatient units: Learning from unsuccessful experience. Australasian Psychiatry,(2), 92?97. Campion, J., McNeill, A., & Checinski, K. (2006). Exempting mental health units from smoke-free laws. BMJ: British Medical Journal, 333(7565), 407?408. Charon, J. M. (2007). Symbolic interactionism: An introduction, an interpretation, an integration. (9th ed.). Upper Saddle River, NJ: Pearson Education. Clifford, J. (1986). Introduction: Partial truths. In Clifford, J. & Marcus, G. E. (Eds.), Writing Culture: The poetics and politics of ethnography (pp. 1-26). Los Angeles, CA: University of California Press. Cole, M. L., Trigoboff, E., Demler, T. L., & Opler, L. A. (2010). Impact of smoking cessation on psychiatric inpatients treated with clozapine or olanzapine. Journal of Psychiatric Practice, 16(2), 75?81. doi:10.1097/01.pra.0000369968.80155.3f Corrigan, P. (2004). How stigma interferes with mental health care. American Psychologist, 59(7), 614?625.  Cowan, S., & Langley, L. (2004). Smoke-free hospitals: Supporting a systems approach to change. Christchurch, NZ: Education for Change. Creswell, J. W. (2007). Qualitative inquiry & research design: Choosing among five approaches (2nd ed.). Thousand Oaks, CA: Sage. Crockford, D., Kerfoot, K., & Currie, S. (2009). The impact of opening a smoking room on psychiatric inpatient behavior following implementation of a hospital-wide smoking ban. Journal of the American Psychiatric Nurses Association, 15(6), 393?400. Cummings, G., Hutchinson, A., Scott, S., Norton, P., & Estabrooks, C. (2010). The relationship between characteristics of context and research utilization in a pediatric setting. BMC Health Services Research, 10(1), 168. doi:10.1186/1472-6963-10-168     135 Dalack, G. W., Healy, D. J., & Meador-Woodruff, J. H. (1998). Nicotine dependence in schizophrenia: Clinical phenomena and laboratory findings. American Journal of Psychiatry, 155(11), 1490?1501. Dan???man, A. (2010). Good intentions and failed implementations: Understanding culture-based resistance to organizational change. European Journal of Work and Organizational Psychology, 29(2), 200?220. Davidson, S., Judd, F., Jolley, D., Hocking, B., Thompson, S., & Hyland, B. (2001). Cardiovascular risk factors for people with mental illness. Australian and New Zealand Journal of Psychiatry, 35(2), 196?202. de Nesnera, A. P., Folks, D. G., & Rauter, U. K. (2012). Implementing a smoking ban: Tips for success. Current Psychiatry, 11(6), E1?E2. Denis, J.-L., H?bert, Y., Langley, A., Lozeau, D., & Trottier, L. (2002). Explaining diffusion patterns for complex health care innovations. Health Care Management Review, 27(3), 60?73. Desai, H. D., Seabolt, J., & Jann, M. W. (2001). Smoking in patients receiving psychotropic medications: A pharmacokinetic perspective. CNS Drugs, 15(6), 469?494. Dixon-Woods, M., Amalberti, R., Goodman, S., Bergman, B., & Glasziou, P. (2011). Problems and promises of innovation: Why healthcare needs to rethink its love/hate relationship with the new. BMJ Quality & Safety, 20(Suppl 1), i47?i51. Dome, P., Lazary, J., Kalapos, M. P., & Rihmer, Z. (2010). Smoking, nicotine and neuropsychiatric disorders. Neuroscience and Biobehavioral Reviews, 34(3), 295?342. Dougherty, M. C., Dwyer, J. W., Pendergast, J. F., Boyington, A. R., Tomlinson, B. U., Coward, R. T., . . . Rooks, L. G. (2002). A randomized trial of behavioral management for continence with older rural women. Research in Nursing & Health, 25(1), 3?13. Dwyer, T., Bradshaw, J., & Happell, B. (2009). Comparison of mental health nurses' attitudes towards smoking and smoking behaviour. International Journal of Mental Health Nursing, 18(6), 424?433. el Guebaly, N., Cathcart, J., Currie, S., Brown, D., & Gloster, S. (2002). Smoking cessation approaches for persons with mental illness or addictive disorders. Psychiatric Services, 53(9), 1166?1170. Emerson, R. M., Fretz, R. I., & Shaw, L. L. (1995). Writing ethnographic fieldnotes. Chicago: University of Chicago Press.     136 Etter, M., & Etter, J. F. (2007). Acceptability and impact of a partial smoking ban in a psychiatric hospital. Preventive Medicine, 44(1), 64?69. Evans, D. E., & Drobes, D. J. (2008). Nicotine self-medication of cognitive-attentional processing. Addiction Biology, 14(1), 32?42. doi:10.1111/j.1369-1600.2008.00130.x Explanatory Notes to Health Act 2006 (U.K.), c. 28. Fetterman, D. M. (2010). Ethnography: Step-by-step (3rd ed.). Thousand Oaks, CA: Sage. Fiore, M. C., Ja?n, C. R., Baker, T. B., Bailey, W. C., Benowitz, N. L., Curry, S. J., . . . Wewers, M. E. (2008). Treating tobacco use and dependence: 2008 update. Clinical practice guideline. Rockville, MD: U.S. Department of Health and Human Services. Public Health Service. Retrieved from Fox, S. D., & Shovein, J. (1993). Achieving smoke-free hospitals. Hospital Topics, 71(3), 29. Fraser Health. (2008a). Keeping you informed: Smoke free policy. Surrey, BC: Author. Retrieved from Fraser Health. (2008b). Live well, smoke-free: For patients and clients living with addictions or mental illness. Surrey, BC: Author. Retrieved from  Frohlich, K. L., Ross, N., & Richmond, C. (2006). Health disparities in Canada today: Some evidence and a theoretical framework. Health Policy, 79(2-3), 132?143. doi:10.1016/j.healthpol.2005.12.010 Gardner, P. D., Tapper, A. R., King, J. A., DiFranza, J. R., & Ziedonis, D. M. (2009). The neurobiology of nicotine addiction: Clinical and public policy implications. Journal of Drug Issues, 39(2), 417?441. Grant, B. F., Hasin, D. S., Chou, S. P., Stinson, F. S., & Dawson, D. A. (2004). Nicotine dependence and psychiatric disorders in the United States: Results from the national epidemiologic survey on alcohol and related conditions. Archives of General Psychiatry, 61(11), 1107?1115. doi:10.1001/archpsyc.61.11.1107 Grant, L. G. (2007). A descriptive qualitative study of what informs and influences smoking behaviours in community dwelling persons with severe and persistent mental illness. (Unpublished Master of Science in Nursing Thesis). University of British Columbia, Vancouver, Canada.     137 Gray, N. (2004). The ethics of policies for the prevention of tobacco disease. ACTA Oncologica, 43(1), 8?10. Greenhalgh, T., Robert, G., MacFarlane, F., Bate, P., & Kyriakidou, O. (2004). Diffusion of innovations in service organizations: Systematic review and recommendations. Milbank Quarterly, 82(4), 581?629. doi:10.1111/j.0887-378X.2004.00325.x Hammersley, M. (1992). What's wrong with ethnography? The myth of theoretical description. In What's wrong with ethnography? Methodological explorations (pp. 11?31). London, UK: Routledge. Hammersley, M., & Atkinson, P. (1995). Ethnography: Principles in practice (2nd ed.). London: Routledge. Head, S. (2003). Smoke free hospitals are unethical. British Medical Journal, 327(7406), 104. Health Act 2006 (U.K.), c. 28. Health Canada. (2005). The national strategy: Moving forward. The 2005 progress report on tobacco control. Ottawa, ON: Author.  Hempel, A. G., Kownacki, R., Malin, D. H., Ozone, S. J., Cormack, T. S., Sandoval, B. G., III, . . . Leinbach, A. E. (2002). Effect of a total smoking ban in a maximum security psychiatric hospital. Behavioral Sciences & the Law, 20(5), 507?522. Hirshbein, L. (2010, Fall). "We mentally ill smoke a lot": Identity, smoking, and mental illness in America. Journal of Social History, 44(1), 7?21. Hitsman, B., Moss, T. G., Montoya, I. D., & George, T. P. (2009). Treatment of tobacco dependence in mental health and addictive disorders. The Canadian Journal of Psychiatry / La Revue canadienne de psychiatrie, 54(6), 368?378. Hollen, V., Ortiz, G., Schacht, L., Mojarrad, M. G., Lane, G. M., & Parks, J. J. (2010). Effects of adopting a smoke-free policy in state psychiatric hospitals. Psychiatric Services, 61(9), 899?904. Hughes, J. R., Hatsukami, D. K., Mitchell, J. E., & Dahlgren, L. A. (1986). Prevalence of smoking among psychiatric outpatients. American Journal of Psychiatry, 143(8), 993?997. Hunter, M. E. (2006). Mental health care in rural and isolated areas: Lessons from Northern British Columbia. BC Medical Journal, 48(4), 174?177. Irish Health Promoting Hospitals (HPH) Network. (2006). Best practice guidelines to support compliance with national policy in relation to tobacco management in the mental health setting. Dublin: Author. Retrieved from     138 Jochelson, K. (2006). Smoke-free legislation and mental health units: The challenges ahead. British Journal of Psychiatry, 189, 479?480. Jochelson, K., & Majrowski, B. (2006). Clearing the air: Debating smoke-free policies in psychiatric units. London: The King's Fund. Johnson, J. L., Moffat, B. M., & Malchy, L. A. (2010). In the shadow of a new smoke free policy: A discourse analysis of health care providers' engagement in tobacco control in community mental health. International Journal of Mental Health Systems, 4, 23. doi:10.1186/1752-4458-4-23 Joseph, A. M., Knapp, J. M., Nichol, K. L., & Pirie, P. L. (1995). Determinants of compliance with a national smoke-free hospital standard. Journal of the American Medical Association, 274(6), 491?494. Jung, T., Scott, T., Davies, H. T., Bower, P., Whalley, D., McNally, R., & Mannion, R. (2009). Instruments for exploring organizational culture: A review of the literature. Public Administration Review, 69(6), 1087?1096. Kagan, I., Kigli-Shemesh, R., Tabak, N., Abramowitz, M. Z., & Margolin, J. (2004). Patient rights and law: Tobacco smoking in psychiatric wards and the Israeli Prevention of Smoking Act. Nursing Ethics, 11(5), 472?478. Katz, J. E. (2005). Individual rights advocacy in tobacco control policies: An assessment and recommendation. Tobacco Control, 14(suppl 2), ii31?ii37. Keltner, N. L., & Folks, D. G. (2005). Psychotropic drugs (4th ed.). St. Louis: Elsevier Mosby. Kennedy, R., Lawless, M., & Slater, B. (2009). The ten essentials of large-scale change. British Journal of Healthcare Management, 15(12), 580?585. Klein, K. J., & Sorra, J. S. (1996). The challenge of innovation implementation. The Academy of Management Review, 21(4), 1055?1080. Kroon, L. A. (2007). Drug interactions with smoking. American Journal of Health-System Pharmacy, 64(18), 1917?1921. Kulig, J. C., Andrews, M. E., Stewart, N. L., Pitblado, R., MacLeod, M. L. P., Bentham, D., . . . Smith, B. (2008). How do registered nurses define rurality? Australian Journal of Rural Health, 16(1), 28?32. Kunyk, D., Els, C., Predy, G., & Haase, M. (2007). Development and introduction of a comprehensive tobacco control policy in a Canadian regional health authority.     139 Preventing Chronic Disease[online serial], 4(2), A30. Available from: Langley, A., & Denis, J. L. (2011). Beyond evidence: The micropolitics of improvement. BMJ Quality & Safety, 20(Suppl 1), i43?i46. Lasser, K., Boyd, J. W., Woolhandler, S., Himmelstein, D. U., McCormick, D., & Bor, D. H. (2000). Smoking and mental illness: A population-based prevalence study. JAMA: The Journal of the American Medical Association, 284(20), 2606?2610. Lawn, S. J. (2004). Systemic barriers to quitting smoking among institutionalised mental health service populations: A comparison of two Australian sites. International Journal of Social Psychiatry,(3), 204?215. Lawn, S. (2011). Habit or addiction: The critical tension in deciding who should enforce hospital smoke-free policies. Canadian Medical Association Journal, 183(18), 2085??2086. doi:10.1503/cmaj.111579  Lawn, S., & Campion, J. (2010). Factors associated with success of smoke-free initiatives in Australian psychiatric inpatient units. Psychiatric Services, 61(3), 300?305.  Lawn, S., & Condon, J. (2006). Psychiatric nurses' ethical stance on cigarette smoking by patients: Determinants and dilemmas in their role in supporting cessation. International Journal of Mental Health Nursing, 15(2), 111?118. Lawn, S., & Pols, R. (2005). Smoking bans in psychiatric inpatient settings? A review of the research. Australian and New Zealand Journal of Psychiatry,(10), 866?885. Lawn, S. J., Pols, R. G., & Barber, J. G. (2002). Smoking and quitting: A qualitative study with community-living psychiatric clients. Social Science & Medicine, 54(1), 93?104. Lawrence, D., Lawn, S., Kisely, S., Bates, A., Mitrou, F., & Zubrick, S. (2011). The potential impact of smoke-free facilities on smoking cessation in people with mental illness. Australian & New Zealand Journal of Psychiatry, 45(12), 1053?1060. Lawrence, D., Mitrou, F., & Zubrick, S. (2009). Smoking and mental illness: Results from population surveys in Australia and the United States. BMC Public Health, 9(1), 285. doi:10.1186/1471-2458-9-285 LeCompte, M. D. (2002). The transformation of ethnographic practice: Past and current challenges. Qualitative Research, 2, 283?299. Lee, R. S., Kochman, A., & Sikkema, K. J. (2002). Internalized stigma among people living with HIV-AIDS. AIDS and Behavior, (6)4, 309?319.     140 Leonard, S., Adler, L. E., Benhammou, K., Berger, R., Breese, C. R., Drebing, C., . . . Freedman, R. (2001). Smoking and mental illness. Pharmacology Biochemistry and Behavior, 70(4), 561?570. Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Beverly Hills, CA: Sage Publications. Link, B. G., & Phelan, J. C. (2001). Conceptualizing stigma. Annual Review of Sociology, 27, 363?385. Longo, D. R., Feldman, M. M., Kruse, R. L., Brownson, R. C., Petroski, G. F., & Hewett, J. E. (1998). Implementing smoking bans in American hospitals: Results of a national survey. Tobacco Control, 7(1), 47?55. Lynam, M. J., Browne, A. J., Reimer Kirkham, S., & Anderson, J. M. (2007). Re-thinking the complexities of ?culture?: What might we learn from Bourdieu? Nursing Inquiry, 14, 23?34. MacLean, L. M., Meyer, M., & Estable, A. (2004). Improving accuracy of transcripts in qualitative research. Qualitative Health Research, 14(1), 113?123. MacLeod, M. L., Kulig, J. C., Stewart, N. J., Pitblado, J. R., Banks, K., D'Arcy, C., . . . Zimmer, L. (2008). The nature of nursing practice in rural and remote Canada. Ottawa, ON: Canadian Health Services Research Foundation. Makomaski Illing, E. M., & Kaiserman, M. J. (2004). Mortality attributable to tobacco use in Canada and its regions, 1998. Canadian Journal of Public Health.Revue Canadienne de Sante Publique, 95(1), 38?44. Matthews, A. M., Wilson, V. B., & Mitchell, S. H. (2011). The role of antipsychotics in smoking and smoking cessation. CNS Drugs, 25(4), 299?315. McEvoy, J. P., Freudenreich, O., & Wilson, W. H. (1999). Smoking and therapeutic response to clozapine in patients with schizophrenia. Biological Psychiatry, 46(1), 125?129. McKenzie, B., & Wharf, B. (2010). Connecting policy to practice in the human services. Toronto, ON: Oxford University Press. McLeroy, K. R., Bibeau, D., Steckler, A., & Glanz, K. (1988). An ecological perspective on health promotion programs. Health Education & Behavior, 15(4), 351?377. McNally, L., Oyefeso, A., Annan, J., Perryman, K., Bloor, R., Freeman, S., Ghodse, A. H. (2006). A survey of staff attitudes to smoking-related policy and intervention in psychiatric and general health care settings. Journal of Public Health,(3), 192?196. Mead, G. H., & Morris, C. W. (Eds.). (1934). Mind, self & society from the standpoint of a social behaviorist. Chicago, IL: The University of Chicago press     141 Mental Health Act, R.S.B.C. 1996, c. 288. Merriam-Webster (2003). Merriam-Webster's collegiate dictionary (11th ed.). Springfield, MA: Merriam-Webster, Inc. Miller, W. R., & Rollnick, S. (2002). Motivational interviewing: Preparing people for change (2nd ed.). New York: Guilford Press. Milne, J., & Oberle, K. (2005). Enhancing rigor in qualitative description. Journal of Wound, Ostomy & Continence Nursing, 32(6), 413?420. Morisano, D., Bacher, I., Audrain-McGovern, J., & George, T. P. (2009). Mechanisms underlying the comorbidity of tobacco use in mental health and addicitive disorders. The Canadian Journal of Psychiatry / La Revue canadienne de psychiatrie, 54(6), 356?367. Morse, J. M., & Field, P. A. (1995). Qualitative research methods for health professionals (2nd ed.). Thousand Oaks, CA: Sage. Moss, T. G., Weinberger, A. H., Vessicchio, J. C., Mancuso, V., Cushing, S. J., Pett, M., . . . George, T. P. (2010). A tobacco reconceptualization in psychiatry: Toward the development of tobacco-free psychiatric facilities. The American Journal on Addictions, 19(4), 293?311. doi:10.1111/j.1521-0391.2010.00051.x National Cancer Institute. (2007). Greater than the sum: Systems thinking in tobacco control. Bethesda, MD, U.S. Department of Health and Human Services, National Institutes of Health, National Cancer Institute. NCI Tobacco Control Monograph No. 18. NIH Pub. No. 06-6085, April 2007.  New South Wales Department of Health. (2008). Drug and alcohol withdrawal clinical practice guidelines - NSW. North Sydney: Author. Newhouse, R. P., & Morlock, L. (2011). Rural hospital nursing: Results of a national survey of nurse executives. Journal of Nursing Administration, 41(3), 129?137. doi:10.1097/NNA.0b013e31820c7212 Northern Health. (2009). Quick facts about Northern Health. Retrieved from  Novak, G., Zai, C. C., Mirkhani, M., Shaikh, S., Vincent, J. B., Meltzer, H., . . . Le Foll, B. (2010). Replicated association of the NR4A3 gene with smoking behaviour in schizophrenia and in bipolar disorder. Genes, Brain & Behavior, 9(8), 910?917. O'Mahony, E., & Rahmani, F. (2004). Staff attitudes to smoking in an Irish mental health service. Psychiatric Bulletin, 28(11), 425.     142 Ontario Hospital Association. (2002). Regional health authorities in Canada: Lessons for Ontario: Discussion paper. Research series: Contributing to the debate on the future of health care in Canada. Toronto: Author. Retrieved from$Att)/pspr56wmkq/$FILE/RegionalHealthAuthoritiesinCanada.pdf?OpenElement Ontario Tobacco Research Unit. (2001). Protection from second-hand tobacco smoke in Ontario. Toronto, ON: Author. Retrieved from Ontario Tobacco Research Unit. (2007). Fact Sheet 1.6: Protection from secondhand smoke. Toronto, ON, Author. Retrieved from"buildings"  Onwuegbuzie, A., & Leech, N. (2007). Validity and qualitative research: An oxymoron? Quality and Quantity, 41(2), 233?249. Pal, L. A. (2006). Beyond policy analysis: Public issue management in turbulent times (3rd ed.). Toronto, ON: Thomson Nelson. Parle, D., Parker, S., & Steeves, D. (2005). Making Canadian healthcare facilities 100% smoke-free: A national trend emerges. Healthcare Quarterly, 8(4), 53?57. Parmelli, E., Flodgren, G., Beyer, F., Baillie, N., Schaafsma, M., & Eccles, M. (2011). The effectiveness of strategies to change organisational culture to improve healthcare performance: A systematic review. Implementation Science, 6(1), 33. doi:10.1186/1748-5908-6-33 Pierce, J. P., & Le?n, M. E. (2008). Effectiveness of smoke-free policies. The Lancet Oncology, 9(7), 614?615. Philpot, T. K. (1994). The ethics of smoke-free zones: An exploration of the implications and effectiveness of a non-smoking policy as a health-promotion strategy in the context of an orthopaedic trauma ward. Journal of Clinical Nursing, 3(5), 307?311. Pitblado, J. R. (2005). Best practices in research methods. So, what do we mean by "rural," "remote," and "northern"? Canadian Journal of Nursing Research, 37(1), 163?168. Poland, B., Frohlich, K., Haines, R. J., Mykhalovskiy, E., Rock, M., & Sparks, R. (2006). The social context of smoking: The next frontier in tobacco control? Tobacco Control, 15(1), 59?63. Powell, B. J., McMillen, J. C., Proctor, E. K., Carpenter, C. R., Griffey, R. T., Bunger, A. C., . . . York, J. L. (2012). A compilation of strategies for implementing clinical innovations in health and mental health. Medical Care Research and Review, 69(2), April. doi:10.1177/1077558711430690     143 Praveen, K. T., Kudlur, S. N., Hanabe, R. P., & Egbewunmi, A. T. (2009). Staff attitudes to smoking and the smoking ban. Psychiatric Bulletin, 33(3), 84?88. Prior, D., & Barnes, M. (2011). Subverting social policy on the front line: Agencies of resistance in the delivery of services. Social Policy & Administration, 45(3), 264?279. doi:10.1111/j.1467-9515.2011.00768.x Prochaska, J. J. (2011). Smoking and mental illness?breaking the link. New England Journal of Medicine, 365(3), 196?198. Prochaska, J. J., Fletcher, L., Hall, S. E., & Hall, S. M. (2006). Return to smoking following a smoke-free psychiatric hospitalization. The American Journal on Addictions, 15(1), 15?22. Procyshyn, R. M., Ihsan, N., & Thompson, D. (2001). A comparison of smoking behaviours between patients treated with clozapine and depot neuroleptics. International Clinical Psychopharmacology, 16(5), 291?294. Procyshyn, R. M., Tse, G., Sin, O., & Flynn, S. (2002). Concomitant clozapine reduces smoking in patients treated with risperidone. European Neuropsychopharmacology, 12(1), 77?80. Ratschen, E., Britton, J., & McNeill, A. (2008). Smoke-free hospitals?The English experience: Results from a survey, interviews, and site visits. BMC Health Services Research, 8(41). doi:10.1186/1472-6963-8-41 Ratschen, E., Britton, J., & McNeill, A. (2009). Implementation of smoke-free policies in mental health in-patient settings in England. British Journal of Psychiatry, 194(6), 547?551. doi:10.1192/bjp.bp.108.051052 Ratschen, E., Britton, J., Doody, G. A., Leonardi-Bee, J., & McNeill, A. (2009). Tobacco dependence, treatment and smoke-free policies: A survey of mental health professionals' knowledge and attitudes. General Hospital Psychiatry, 31(6), 576?582. Ratschen, E., Britton, J., Doody, G. A., & McNeill, A. (2009). Smoke-free policy in acute mental health wards: Avoiding the pitfalls. General Hospital Psychiatry, 31(2), 131?136. doi:10.1016/j.genhosppsych.2008.10.006 Ratschen, E., McNeill, A., Doody, G. A., & Britton, J. (2008). Smoking, mental health, and human rights: A UK judgment. The Lancet, 371(9630), 2067?2068. Reilly, P., Murphy, L., & Alderton, D. (2006). Challenging the smoking culture within a mental health service supportively. International Journal of Mental Health Nursing, 15(4), 272?278. Resnick, M. P., & Bosworth, E. E. (1989). A smoke-free psychiatric unit. Hospital & Community Psychiatry, 40(5), 525?527.     144 Rich, D., & Knowlden, S. (2002). Towards a smoke free mental health workplace: Midas Tobacco Project: Mental health goes smoke free. South West Sydney Area Health Service/Fairfield Department of General Practice, Retrieved from  Rudnick, J. (2002). Tobacco use in the workplace. In ILO encyclopedia of occupational health and safety [online version] (pp. 15.1?15.89). Programme on Safety and Health at Work and the Environment (Safework). Geneva: International Labour Organization. Retrieved from R?sch, N., Todd, A.R., Bodenhausen, G.V., & Corrigan, P.W. (2010). Do people with mental illness deserve what they get? Links between meritocratic worldviews and implicit versus explicit stigma. European Archives of Psychiatry and Clinical Neuroscience, 260(8), 617?625 Ryan-Nicholls, K. D., & Haggarty, J. M. (2007). Collaborative mental health care in rural and isolated Canada: Stakeholder feedback. Journal of Psychosocial Nursing & Mental Health Services, 45(12), 37?45. Sandelowski, M. (1995a). Qualitative analysis: What it is and how to begin. Research in Nursing & Health, 18(4), 371?375. Sandelowski, M. (1995b). Sample size in qualitative research. Research in Nursing & Health, 18, 179?183. Sarna, L., Bialous, S. A., Wells, M. J., & Kotlerman, J. (2009). Smoking among psychiatric nurses: Does it hinder tobacco dependence treatment? Journal of the American Psychiatric Nurses Association, 15(1), 59?67. doi:10.1177/1078390308330638 Schroeder, S. A., & Morris, C. D. (2010). Confronting a neglected epidemic: Tobacco cessation for persons with mental illnesses and substance abuse problems. Annual Review of Public Health, 31(1), 297?314. doi:10.1146/annurev.publhealth.012809.103701 Schultz, A. S., Bartmanovich, C., Snowball, N., Kvern, M., Finegan, B., Nykiforuk, C., Ramsden, V., Green, M.(2010). Exploring tobacco use management post-implementation of smoke free hospital grounds study (ETS): The ETS project: Enforcement and compliance: Brief Report. Retrieved from Schultz, A. S., Finegan, B., Nykiforuk, C. I., & Kvern, M. A. (2011). A qualitative investigation of smoke-free policies on hospital property. Canadian Medical Association Journal, 183(18), E1334?E1344. doi:10.1503/cmaj.110235 Schultz, A. S., James, M., Ramsden, V., Green, M., Snowball, N., & Bartmanovich, C. (2010). Exploring tobacco use management on psychiatric units post-implementation     145 of smoke-free hospital grounds study (ETS): Brief Report. Retrieved from Seale, C., & Ankiss, F. (2012). Content and comparative keyword analysis. In C. Seale (Ed.), Researching society and culture (3rd ed., pp. 459?478). Thousand Oaks, CA: Sage Publications. Shetty, A., Alex, R., & Bloye, D. (2010). The experience of a smoke-free policy in a medium secure hospital. The Psychiatrist, 34, 287?289. doi:10.1192/pb.bp.109.027425 Smoke-free (Exemptions and Vehicles) Regulations 2007 (U.K.) SI 2007/765. Snow, T. (2006). Smoking ban 'will be a nightmare' in psychiatric premises. Nursing Standard, 20(46), 9. Solty, H., Crockford, D., White, W. D., & Currie, S. (2009). Cigarette smoking, nicotine dependence, and motivation for smoking cessation in psychiatric inpatients. The Canadian Journal of Psychiatry / La Revue canadienne de psychiatrie, 54(1), 36?45. Solway, E. (2009). Windows of opportunity for culture change around tobacco use in mental health settings. Journal of the American Psychiatric Nurses Association, 15(1), 41?49. doi:10.1177/1078390308329885  Speziale, H. S., & Carpenter, D. R. (2011). Qualitative research in nursing: Advancing the humanistic imperative (5th ed.). Philadelphia: Wolters Kluwer Health/Lippincott Williams & Wilkins. Spradley, J. P. (1979). The ethnographic interview. New York: Holt, Rinehart and Winston. Spradley, J. P. (1980). Participant observation. Toronto, ON: Wadsworth Thomson Learning. Spradley, J. P., & McCurdy, D. W. (1972). The cultural experience: Ethnography in complex society (1988 reissue ed.). Long Grove, Ill: Waveland Press. Statistics Canada. (2006). 2006 Community Profiles. Author. Retrieved from  Statistics Canada. (2007a). Terrace, British Columbia (Code965) (table). 2006 Community Profiles. 2006 Census. Statistics Canada Catalogue no. 92-591-XWE. Ottawa, ON: Author. Retrieved from  Statistics Canada. (2007b). Prince George, British Columbia (Code970) (table). 2006 Community Profiles. 2006 Census. Statistics Canada Catalogue no. 92-591-XWE. Ottawa, ON: Author. Retrieved from      146  Stronach, I., Corbin, B., McNamara, O., Stark, S., & Warne, T. (2002). Towards an uncertain politics of professionalism: Teacher and nurse identities in flux. Journal of Education Policy, 17(1), 109?138. doi:10.1080/02680930110100081 Stubbs, J., Haw, C., & Garner, L. (2004). Survey of staff attitudes to smoking in a large psychiatric hospital. Psychiatric Bulletin, 28(6), 204?207. Thorne, S., Kirkham, S. R., & MacDonald-Emes, J. (1997). Focus on qualitative methods. Interpretive description: A noncategorical qualitative alternative for developing nursing knowledge. Research in Nursing & Health, 20(2), 169?177. Tobacco Control Regulation, BC Reg. 232/2007, as amended to BC Reg. 394/2007. U.S. Department of Health and Human Services. (2004). The health consequences of smoking: A report of the Surgeon General. [online version] Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. Retrieved from U.S. Department of Health and Human Services. (2006). The health consequences of involuntary exposure to tobacco smoke: A report of the Surgeon General. Atlanta, GA: Centers for Disease Control and Prevention, Coordinating Center for Health Promotion, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. Retrieved from U.S. Department of Health and Human Services. (2010). Ending the tobacco epidemic: A tobacco control strategic acton plan for the U.S. Department of Health and Human Services. Washington: Office of the Assitant Secretary for Health. Retrieved from Urichuk, L., Prior, T. I., Dursun, S., & Baker, G. (2008). Metabolism of atypical antipsychotics: Involvement of cytochrome p450 enzymes and relevance for drug-drug interactions. Current Drug Metabolism, 9(5), 410?418. Vancouver Coastal Health. (2008). Smoke-free premises policy. Vancouver, BC: Author.  Voci, S., Bondy, S., Zawertailo, L., Walker, L., George, T. P., & Selby, P. (2010). Impact of a smoke-free policy in a large psychiatric hospital on staff attitudes and patient behavior. General Hospital Psychiatry, 32(6), 623?630. doi:10.1016/j.genhosppsych.2010.08.005 Weiner, B., Belden, C., Bergmire, D., & Johnston, M. (2011). The meaning and measurement of implementation climate. Implementation Science, 6(1), 78. doi:10.1186/1748-5908-6-78     147 Wendt, H., Euwema, M. C., & van Emmerik, I. J. (2009). Leadership and team cohesiveness across cultures. The Leadership Quarterly, 20(3), 358?370. doi:10.1016/j.leaqua.2009.03.005 Williams, J. M. (2008). Eliminating tobacco use in mental health facilities: Patients' rights, public health, and policy issues. JAMA - Journal of the American Medical Association,(5), 571?573. Williams, J. M., Gandhi, K. K., & Benowitz, N. L. (2010). Carbamazepine but not valproate induces CYP2A6 activity in smokers with mental illness. Cancer epidemiology, biomarkers & prevention: A publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology, 19(10), 2582?2589. Williams, S. C., Hafner, J. M., Morton, D. J., Holm, A. L., Milberger, S. M., Koss, R. G., & Loeb, J. M. (2009). The adoption of smoke-free hospital campuses in the United States. Tobacco Control, 18(6), 451?458. Williams, J. M., Ziedonis, D. M., Abanyie, F., Steinberg, M. L., Foulds, J., & Benowitz, N. L. (2005). Increased nicotine and cotinine levels in smokers with schizophrenia and schizoaffective disorder is not a metabolic effect. Schizophrenia Research, 79(2-3), 323?335. Williams, J. M., Ziedonis, D. M., Vreeland, B., Speelman-Edwards, N., Zechner, M. R., Williams, M. T., . . . Eilers, R. (2009). A wellness approach to addressing tobacco in mental health settings: Learning about healthy living. American Journal of Psychiatric Rehabilitation, 12(4), 352?369. doi:10.1080/15487760903248580 Williams, J., Zimmermann, M., Steinberg, M., Gandhi, K., Delnevo, C., Steinberg, M., & Foulds, J. (2011). A comprehensive model for mental health tobacco recovery in New Jersey. Administration and Policy in Mental Health and Mental Health Services Research, 38(5), 368?383. Wing, V. C., Wass, C. E., Soh, D. W., & George, T. P. (2012). A review of neurobiological vulnerability factors and treatment implications for comorbid tobacco dependence in schizophrenia. Annals of the New York Academy of Sciences, 1248, 89?106. Wolcott, H. F. (2008). Ethnography: A way of seeing (2nd ed.). Lanham, MD: Altamira. Wolfenden, L., Campbell, E., Wiggers, J., Walsh, R. A., & Bailey, L. J. (2008). Helping hospital patients quit: What the evidence supports and what guidelines recommend. Preventive Medicine, 46(4), 346?357. Wu, X., Ramesh, M., Howlett, M., & Firtzen, S. (2010). The public policy primer: Managing the policy process. London: Routledge.     148 Wye, P. M., Bowman, J. A., Wiggers, J. H., Baker, A., Knight, J., Carr, V. J., . . . & Clancy, R. (2009). Smoking restrictions and treatment for smoking: Policies and procedures in psychiatric inpatient units in Australia. Psychiatric Services, 60(1), 100?107. Wye, P., Bowman, J., Wiggers, J., Baker, A., Knight, J., Carr, V., . . . & Clancy, R. (2010). Total smoking bans in psychiatric inpatient services: A survey of perceived benefits, barriers and support among staff. BMC Public Health, 10(372). Yanow, D. (2000). Conducting interpretive policy analysis. Qualitative research methods Series 47. Thousand Oaks, CA: Sage Publications. York, L. N. (1997). A phenomenological investigation of cigarette smoking in persons with severe and persistent mental illness (Unpublished Ph.D. Thesis). Saint Louis University, Saint Louis, MO. Retrieved from Ziedonis, D. M., & Williams, J. M. (2003). Management of smoking in people with psychiatric disorders. Current Opinion in Psychiatry, 16(3), 305?315. Ziedonis, D. M., Williams, J. M., & Smelson, D. P. (2003). Serious mental illness and tobacco addiction: A model program to address this common but neglected issue. American Journal of the Medical Sciences, 326(4), 223?230.        149 Appendices     150 Appendix A  - Smoke-free Grounds Policy Document      151         152 Appendix B  - Participant Observation Guide      153      154 Appendix C  - Document Collection Guide      155      156 Appendix D  - Healthcare Professional Interview Guide     157      158 Appendix E  - Patient Interview Guide     159      160 Appendix F  - Key Informant Interview Guide     161      162 Appendix G  - Key Informant Demographics Form     163     164     165      166 Appendix H  - Healthcare Professional Consent Forms H.1 ? Healthcare Professional Informal Discussion Consent Form     167     168     169        170 H.2 ? Healthcare Professional Interview Consent Form     171     172     173      174 Appendix I  - Healthcare Professional Demographics Form     175     176     177      178 Appendix J  - Housekeeping Staff Email Invitation     179      180 Appendix K  - Patient Invitations      181 Appendix L  - Patient Consent      182      183      184      185 Appendix M  - Patient Demographics Form     186  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items