- Email: [email protected]
Acceptability and impact of a partial smoking ban followed by a total smoking ban in a psychiatric hospital
Available online at www.sciencedirect.com
Preventive Medicine 46 (2008) 572 – 578 www.elsevier.com/locate/ypmed
Acceptability and impact of a partial smoking ban followed by a total smoking ban in a psychiatric hospital Manuela Etter a , Aqal Nawaz Khan a , Jean-François Etter b,⁎ a b
Department of Psychiatry, Geneva University Hospitals, Geneva, Switzerland Institute of Social and Preventive Medicine, University of Geneva, Switzerland Available online 26 January 2008
Abstract Objective. To assess the impact of a partial smoking ban followed by a total smoking ban in a psychiatric hospital in Switzerland. Methods. In 2003, smoking was allowed everywhere in psychiatric units. In 2004, smoking was prohibited everywhere except in smoking rooms. In 2006, smoking rooms were removed and smoking was totally prohibited indoors. Patients and staff were surveyed in 2003 (n = 106), 2004 (n = 108), 2005 (n = 119) and 2006 (n = 134). Results. Exposure to environmental tobacco smoke (ETS) decreased after the partial ban and further decreased after the total ban. Among patients, after the total ban, more smokers attempted to quit smoking (18%) relative to before the total ban (2%, odds ratio = 10.1, p = 0.01). More smokers said that hospital staff gave them nicotine replacement products after the total ban (52%), compared with before (13%, odds ratio = 7.6, p b 0.001). Many participants (55%) commented that the total ban was too strict, and most (64%) preferred the partial ban. Conclusions. The partial ban decreased exposure to ETS and the total ban further improved the situation and increased the proportion of smokers who attempted to quit smoking and received nicotine medications. The total ban was loosely enforced and was overall acceptable, but most participants preferred a partial ban. © 2008 Elsevier Inc. All rights reserved. Keywords: Smoking; Nicotine dependence; Smoking cessation; Psychiatry, Preventive measures
Introduction Smoking prevalence is much higher in patients with psychiatric disorders relative to the general population (De Leon and Diaz, 2005). The high frequency of smoking in psychiatric units is a cause of exposure to environmental tobacco smoke (ETS), a recognized carcinogen (Boffetta et al., 1998). It also results in annoyance due to the smell of tobacco and may incite smoking in patients. Banning smoking in hospitals protects non smokers from ETS and protects non smokers and former smokers from cues to smoke, thus reducing the risk of smoking initiation or relapse during hospital stays (Glasgow et al., 1991; Hughes, 1993; Rigotti et al., 2000). Also, it reduces fires and cleaning costs (McKee et al., 2003). For all these reasons, U.S. hospitals ⁎ Corresponding author. Institute of Social and Preventive Medicine, University of Geneva, CMU, case postale, CH-1211 Geneva 4, Switzerland. Fax: +41 22 379 59 12. E-mail address: [email protected] (J.-F. Etter). 0091-7435/$ - see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2008.01.004
are required to be smoke free and nearly all have complied (Longo et al., 1998). In Europe, the situation is quite different, because most hospitals have only implemented partial smoking bans (ENSH, 2005; Willemsen et al., 2004). In the U.K., hospitals have to be smoke-free under the Health Act 2006, but psychiatric hospitals may be exempted from this regulation for several years (Jochelson, 2006; Campion et al., 2006). Smoking bans have been implemented less often in psychiatry departments than in somatic care departments because of a fear that psychiatric patients would be more reluctant to comply with smoking bans (McNally et al., 2006; Stubbs et al., 2004). Research showed, however, that smoking bans were in general well accepted in psychiatric inpatient units (El-Guebaly et al., 2002; Harris et al., 2007), and that staff may have anticipated more problems than actually occurred (Lawn and Pols, 2005; Smith and Grant, 1989; Matthews et al., 2005). In a previous article, we examined the impact and acceptability of a partial smoking ban in a psychiatric hospital in Switzerland (Etter and Etter, 2007). We showed that this partial smoking ban
M. Etter et al. / Preventive Medicine 46 (2008) 572–578
573
rooms in each unit, one for patients and one for hospital staff. A steering committee, which consisted of physicians, nurses, psychologists and social workers, implemented the partial smoking ban. Despite the fact that a majority of both patients and staff rejected the idea of a total smoking ban (Etter and Etter, 2007), the hospital administration decided that smoking would be banned everywhere inside hospital buildings beginning January 2006. The smoking rooms were then removed. Smoking continued to be allowed outdoors, and patients (except those in locked rooms) and staff were allowed to leave the unit to smoke outdoors. Before implementation of the smoking bans, health professionals were trained to treat tobacco dependence in their patients, booklets were distributed to smokers, and no-smoking signs were posted throughout the buildings (Etter and Perneger, 2001; Humair and Cornuz, 2003). Smokers who wanted to quit were offered advice and nicotine replacement therapy (NRT). NRT was included in the fixed-price in-patient contract and was therefore available at no additional cost to patients, but not to hospital staff. Bupropion (Zyban) was not included in the in-patient contract, and varenicline was not yet available.
was well accepted and decreased exposure to ETS, despite the fact that it was not strictly enforced. Two years later, a total smoking ban was implemented in the same hospital. The acceptability and impact of total smoking bans in psychiatry hospitals is incompletely documented, in particular in Europe. A recent review found only seven empirical studies on the effects of total smoking bans in psychiatry in-patient units, and all of them were from the USA (El-Guebaly et al., 2002). Another recent review found only 10 studies of total smoking bans in psychiatric settings (Lawn and Pols, 2005). Our aim was to compare the acceptability and efficacy of a partial smoking ban and total ban in an in-patient psychiatric hospital. Methods Setting
Study design
The smoking ban was implemented in the whole Psychiatry Department of the Geneva University Hospitals (10 units, 166 beds), but the study was conducted only in two in-patient, adult units of this Department. The first was an admission and short-stay unit (16 beds, mean duration of stays = 17 days, median = 7 days) and the second was a medium-stay unit (16 beds, mean duration of stays = 37 days, median = 15 days). Patients had mainly psychotic disorders, depression and personality disorders.
We conducted four surveys with the same questions. The first survey took place in October 2003 (before the partial ban), the second in April 2004 (2 months after implementation of the partial ban), the third from October to December 2005 (20 months after the partial ban, but before the total ban) and the fourth from March to May 2006 (3–5 months after the total smoking ban). The questionnaires were brief and self-administered. A physician, a nurse or a psychologist distributed the questionnaires to patients and staff after explaining the study and obtaining written informed consent. Patients answered the survey as soon as their condition allowed for, which in most cases was about one week after admission. The physician, nurse or psychologist completed the questionnaires with patients who were unable to answer by themselves. All questionnaires were anonymous. The third and fourth surveys were approved by the ethics committee of the Geneva University Hospitals, but the first two surveys were not submitted to an ethics committee. Otherwise, the study followed the principles in the Declaration of Helsinki.
Intervention In 2003, patients were allowed to smoke everywhere except in bedrooms and dining rooms, but this policy was not enforced and in fact, patients smoked everywhere. In February 2004, a partial smoking ban was implemented; smoking was banned everywhere except in closed smoking rooms equipped with ventilation and glass doors for surveillance. There were two smoking
Table 1 Characteristics of participants, before and after introduction of a partial and a total smoking ban (Switzerland, 2003–2006)
Patients N participants Intended sample Participation rate (%) Age (mean), years % men Ever smoked 100+ cigarettes Smoking status (%) – Daily smokers – Occasional (non-daily) smokers – Former smokers – Never smokers Staff N participants Intended sample Participation rate Age (mean), years % men Ever smoked 100+ cigarettes (%) Smoking status (%) – Daily smokers – Occasional (non-daily) smokers – Former smokers – Never smokers
No ban 2003
Partial ban 2004
Partial ban 2005
Total ban 2006
p value 2003–2004
p value 2003–2006
p value 2005–2006
49 57 86.0 39.9 59.2 91.8
54 67 80.6 39.5 64.8 74.1
66 124 53.2 42.2 48.5 78.8
77 114 67.5 41.0 60.0 81.6
– – 0.42 0.89 0.57 0.018 0.038
– – 0.010 0.68 0.93 0.11 0.43
– – 0.024 0.28 0.17 0.67 0.25
73.5 6.1 12.2 8.2
64.8 0 9.3 25.9
68.8 6.3 6.3 18.8
65.8 2.6 15.8 15.8
57 57 100 38.8 35.1 64.9
54 55 98.2 38.5 37.0 63.0
53 63 84.1 38.7 35.3 56.6
57 62 91.9 40.7 37.5 57.9
– – 0.31 0.86 0.83 0.83 0.97
– – 0.028 0.41 0.79 0.44 1.0
– – 0.18 0.47 0.81 0.89 0.77
26.3 7.0 22.8 43.9
27.8 5.6 20.4 46.3
24.5 11.3 17.0 47.2
26.3 7.0 22.8 43.9
574
M. Etter et al. / Preventive Medicine 46 (2008) 572–578
Participants
Awareness and satisfaction with the non-smoking policy
The target sample included all patients and staff present at the time of data collection. Given the short duration of most hospital stays and the resulting patient turnover, the samples of patients in the successive surveys consisted mostly of different people, whereas the samples of hospital staff consisted largely of the same people who answered successive surveys.
In 2005, 89% of patients and 94% of staff correctly answered that smoking was allowed only in smoking rooms. In 2006, 90% of patients and 93% of staff correctly answered that smoking was prohibited everywhere in the clinic. In 2003, 2004 and 2005, most participants answered that the rules about smoking in the clinic were adequate, but in 2006 (after the total ban) most answered that these rules were too strict (2005: 14.3%, 2006: 55.0%, p b 0.001) (Table 2). In 2003 (before the partial ban), half the participants (51.6%) answered that the rules about smoking in the clinic were not respected, whereas only 17.6% gave this answer in 2004 (p b 0.001), but this figure increased after introduction of the total ban (2005: 16.1%, 2006: 32.6%, p = 0.021) (Table 2). A minority of participants favored a total smoking ban (19.7% in 2005 and 27.8% in 2006, p = 0.09) and most preferred a partial ban with closed smoking rooms (77.8% in 2005 and 64.3% in 2006, p = 0.09). Nevertheless, most participants in the 2006 survey (52.8%) agreed with the statement, “I am pleased with the smoking prohibition at the clinic”. In all four surveys, with no statistically significant change after introduction of the partial or total ban, most staff members (68–83%) and one third to one half of patients (38–56%) agreed with the statement, ”At the hospital, tobacco smoke is a source of conflict with (other) patients”. More patients agreed with the statement, “tobacco smoke is a source of conflict with hospital staff” in 2006, after the total ban (36.4%) than in 2003, when smoking was allowed everywhere (24.7%, p = 0.005). A substantial minority of patients and staff agreed with the statement:
Measurements The questionnaires covered age, sex and smoking status (Table 1), opinions about the no-smoking policy (Table 2), perceived exposure to ETS (Table 3), smoking behavior and smoking cessation interventions received from hospital staff (Table 4). Hospital staff indicated whether they advised and helped patients to quit smoking, or gave them nicotine replacement medications.
Statistical analyses We used chi-square tests and odds ratios to compare proportions, and independent-sample t tests to compare means.
Results Participation There were 106 participants in 2003, 108 in 2004, 119 in 2005 and 134 in 2006. The intended samples were smaller in 2003–2004 than in 2005–2006 because data were collected during only one month in 2003–2004 and during two months in 2005–2006 (Table 1). Among patients and staff, participation rates were higher in 2003 and 2004 than in 2005 and 2006. The 2003 and 2004 data were reported previously (Etter and Etter, 2007).
Table 2 Opinions about smoking at the hospital, before and after introduction of a partial and a total smoking bans (Switzerland, 2003–2006)
Patients Rules about smoking… are too strict … are adequate “At the hospital…” (% agree) Rules on smoking are not respected Tobacco smoke is a source of conflict with other patients Tobacco smoke is a source of conflict with hospital staff Cohabitation between smokers and non-smokers is very difficult Staff Rules about smoking are… too strict … are adequate “At the hospital…” (% agree) Rules on smoking are not respected Tobacco smoke is a source of conflict with patients Cohabitation between smokers and non-smokers is very difficult All (patients and staff) Rules about smoking are… too strict … are adequate “At the hospital…” (% agree) Rules on smoking are not respected Tobacco smoke is a source of conflict with (other) patients Tobacco smoke is a source of conflict with (other) hospital staff Cohabitation between smokers and non-smokers is very difficult
No ban 2003
Partial ban 2004
Partial ban 2005
Total ban 2006
p value 2003–2004
p value 2003–2006
p value 2005–2006
12.2 73.5
14.8 79.6
15.2 77.3
49.4 46.8
0.32
b0.001
b0.001
40.8 49.0 24.7 32.6
16.7 40.7 24.1 29.6
10.6 56.1 42.6 53.4
22.1 37.7 36.4 44.9
0.023 0.16 0.68 0.70
0.06 0.56 0.005 0.033
7.0 71.9
14.8 81.5
13.2 75.5
59.6 36.8
0.014
b0.001
b0.001
59.9 82.5 54.4
18.6 74.1 31.5
22.7 67.9 35.8
45.6 80.4 43.9
b0.001 0.44 0.19
0.12 0.45 0.027
0.066 0.31 0.53
9.4 72.6
14.8 80.6
14.3 76.5
55.0 43.5
0.006
b0.001
b0.001
51.6 67.0 31.2 44.3
17.6 57.4 27.7 30.5
16.1 62.9 38.1 45.9
32.6 58.7 38.5 44.4
b0.001 0.30 0.46 0.30
0.026 0.24 0.009 0.004
0.021 0.63 0.88 0.82
0.14 0.16 0.62 0.86
M. Etter et al. / Preventive Medicine 46 (2008) 572–578
575
Table 3 Exposure to and annoyance from environmental tobacco smoke (ETS) among non-smokers (patients and staff), before and after introduction of the partial and total smoking bans (Switzerland, 2003–2006) No ban 2003
Partial ban 2004
Patients Exposure to ETS in unit (% “never”) – in bedrooms 69.2 89.5 – in dining rooms 30.8 63.2 – in corridors 23.1 47.4 Annoyance due to ETS in unit (% “absolutely not”) – in bedrooms 61.5 78.9 – in dining rooms 38.5 73.7 – in corridors 38.5 63.2 Staff Exposure to ETS in unit (%“never”) – in bedrooms 16.7 38.5 – in dining rooms 26.2 69.2 – in corridors 9.5 23.1 Annoyance due to ETS in unit (% “absolutely not”) – in bedrooms 23.8 38.5 – in dining rooms 31.0 41.0 – in corridors 23.8 35.9 All (patients and staff) Exposure to ETS in unit (% “never”) – in bedrooms 25.0 54.5 – in dining rooms 35.5 65.5 – in corridors 10.4 30.9 Annoyance due to ETS in unit (% “absolutely not”) – in bedrooms 29.2 52.7 – in dining rooms 35.0 52.7 – in corridors 27.1 45.5
Partial ban 2005
Total ban 2006
p value 2003–2004
p value 2003–2006
p value 2005–2006
60.0 65.0 30.0
88.5 73.1 65.4
0.18 0.17 0.19
0.058 0.09 0.029
0.028 0.48 0.06
75.0 50.0 45.0
76.9 80.8 69.2
0.45 0.16 0.07
0.108 0.007 0.162
0.37 0.061 0.051
35.0 55.0 20.0
31.0 71.4 38.1
0.14 0.054 0.018
0.041 0.004 0.006
0.21 0.39 0.30
47.5 65.0 40.0
45.2 81.0 52.4
0.24 0.010 0.10
0095 b0.001 0.023
0.89 0.43 0.53
51.2 73.8 26.1
55.2 83.3 51.7
0.046 0.037 0.001
0.013 b0.001 b0.001
0.12 0.32 0.045
62.8 71.4 44.7
61.4 90.0 62.7
0.083 0.023 0.068
0.013 b0.001 0.002
0.98 0.045 0.067
“At the hospital, cohabitation between smokers and nonsmokers is very difficult”. Before and after the total smoking ban, most participants (75–78%) agreed with the statement, “For patients in locked rooms, the smoking prohibition is very hard to bear”. The total ban increased the proportion of patients who reported that they “sometimes” or “often” got angry with hospital staff because of the smoking policy (4.5% in 2005 and 24.5% in 2006, p = 0.02, odds ratio = 6.8, 95% confidence interval = 1.2 to 47.3). In parallel, there was a non-significant increase in the proportion of staff members who reported that patients “sometimes” or “often” got angry with them because of the nosmoking rules (47.2% in 2005 and 62.2% in 2006, p = 0.29), and most staff (61.5%) commented that some patients refused to abide by these rules. There was, however, no significant change between 2005 and 2006 in answers from staff members to the question: “If we totally prohibited smoking in the clinic, we would face strong protest from patients” (32.7% agreed in 2005 and 42.8% in 2006, p = 0.28). After the total ban (these questions were not asked in 2005), most staff members (80.7%) reported that some patients smoked in the bedrooms and left the clinic to go out and buy cigarettes (82.4%). Perceived exposure to ETS Across all surveys, staff members reported more exposure to, and annoyance from ETS than patients. Among non-smokers
(patients and staff, Table 3), exposure to ETS decreased in bedrooms after the partial ban, but did not further decrease after the total ban. On third (31%) of the non-smokers still commented that they were “often” or “sometimes” exposed to ETS in bedrooms after introduction of the total ban. Exposure to ETS in dining rooms and corridors decreased after the partial ban and further decreased after the total ban (Table 3). After the total ban, no participant reported that they were “often” exposed to ETS in dining rooms and offices, but 12.0% remained “often” exposed to ETS in corridors. In non-smokers, the average duration of exposure to ETS in the clinic, as self-reported in questionnaires, decreased significantly after the total smoking ban (from 69 min per day in 2005 to 12 min in 2006, p = 0.012). Annoyance from ETS Annoyance caused by ETS in bedrooms decreased after introduction of the partial ban, but did not further decrease after the total ban (Table 3). In contrast, the partial ban decreased levels of annoyance caused by ETS in dining rooms and corridors, and the situation further improved after implementation of the total smoking ban. Nevertheless, after the total ban, 15.8% of non-smokers remained “a lot” or “somewhat” annoyed by ETS in bedrooms and 13.6% in corridors. The situation was best in dining rooms, where only 1.8% of participants were “a lot” or “somewhat” annoyed by ETS after the total ban.
576
M. Etter et al. / Preventive Medicine 46 (2008) 572–578
Table 4 Smoking behavior and smoking cessation interventions, among patients who smoked, before and after introduction of the partial and the total smoking bans (Switzerland, 2003–2006)
Among patients, daily and occasional smokers: Cigarettes/day now, mean Cigarettes/day before admission, mean Since admission in the clinic, smokes (%): – more than before admission – no change – less than before admission During your current hospital stay, did a physician or a nurse… (% Yes) – advise you to quit smoking? – help you to quit smoking? – provide medication to quit smoking (like a nicotine patch or gum or Zyban)? At the clinic… (% sometimes + often) – Do other patients provide you cig.? – Do hospital staff provide you cigarettes? – Do hospital staff forbid you to smoke?
No ban 2003
Partial ban 2004
Partial ban 2005
Total ban 2006
p value 2003–2004
24.1 24.3
33.1 29.4
22.0 24.0
23.7 24.6
0.015 0.21 0.065
0.81 0.17 0.21
0.51 0.87 0.96
42.2 13.2 43.7
65.7 20.0 14.3
42.2 26.7 31.1
39.6 29.2 31.3
15.4 2.6 5.1
5.7 5.7 11.4
30.0 15.0 12.5
42.6 19.6 52.2
0.18 0.49 0.32
0.006 0.015 b0.001
0.23 0.58 b0.001
43.6 10.3 5.1
45.7 2.9 14.3
42.2 11.3 11.3
52.2 14.6 22.4
0.98 0.45 0.20
0.14 0.61 0.21
0.26 0.75 0.61
Smoking behavior The total smoking ban was not followed by any change in smoking prevalence or cigarette consumption (Table 1), but it was followed by an increase in the proportion of patients who attempted to quit smoking during their hospital stay (from 2.2% in 2005 to 18.4% in 2006, p = 0.01, odds ratio = 10.1, 95% confidence interval 1.21 to 222.7). Among patients, a substantial minority of smokers (40% in 2006) said they smoked more at the clinic than before their hospital admission, but most said they smoked the same amount or less, with no significant change between surveys (Table 4). Cigarette exchange among patients was frequent, and both patients and staff reported that staff members sometimes provided cigarettes to patients (Table 4). Between 2003 and 2004, there was an increase in the proportion of never smokers and, among smokers, there was an increase in cigarette consumption. These results should however be considered with caution (see Etter and Etter, 2007 for a discussion of this point). Smoking cessation interventions, as reported by patients Among patients who smoked, there was no statistically significant difference between the 2003–2004 surveys and the 2005–2006 surveys in answers to questions on whether hospital staff advised them or helped them to quit smoking (Table 4). However, there was an important improvement in the proportion of patients who received advice to quit smoking from a staff member between 2003 (no ban, 15.4%) and 2006 (total ban, 42.6%, p = 0.006, odds ratio = 4.1, 95% confidence interval = 1.4 to 11.6), or who received help to quit smoking from a staff member (from 2.6% in 2003 to 19.6% in 2006, p = 0.015, odds ratio = 9.2, 95% confidence interval 1.1 to 76.6). After implementation of the total ban, there was an increase in the proportion of smokers who commented that hospital staff gave
p value 2003–2006
p value 2005–2006
them nicotine replacement medications (patch, gum) (from 12.5% in 2005 to 52.2% in 2006, p b .001, odds ratio = 7.6, 95% confidence interval = 2.3 to 27.0). Patients reported receiving smoking cessation support less frequently than was reported by staff members. Smoking cessation interventions, as reported by hospital staff These questions were asked only in 2005 and 2006. According to hospital staff, the total smoking ban was followed by a doubling in the proportion of patients to whom help was provided to quit smoking (from 26.9% in 2005 to 58.2% in 2006, p = 0.007, odds ratio = 3.8, 95% confidence interval = 1.6 to 9.3) and by an increase in the proportion of smokers to whom staff members provided nicotine replacement medications (from 42.3% in 2005 to 74.5% in 2006, p b 0.001, odds ratio = 4.0, 95% confidence interval = 1.6 to 9.9). Discussion The originality of this study lies in its four time points, which enabled us to compare, in the same psychiatric in-patient clinic, a situation where smoking was allowed almost without restriction, with a partial smoking ban (when smoking was allowed only in closed, ventilated smoking rooms), and with a total smoking ban, when smoking was, in principle, completely banned inside hospital buildings. Both the partial and the total smoking bans decreased exposure to and annoyance from ETS, which was their main objective. Compared with the situation in 2003, when smoking was in fact unrestricted, the total smoking ban represented an important improvement, even though it was only loosely enforced. Importantly, the total ban (but not the partial ban) was followed by a substantial increase in the proportion of smokers who attempted to quit smoking during their hospital stay (from 2% to 18%). This is an important result, as it is usually
M. Etter et al. / Preventive Medicine 46 (2008) 572–578
difficult for psychiatric patients to quit smoking (Etter et al., 2004), and because smoking restrictions do not always motivate psychiatric patients to quit smoking (Downey et al., 1998). The total smoking ban was also followed by a substantial increase in the proportion of patients who received nicotine medications. This result seems robust, since it was observed in answers for both patients and staff. In contrast, the partial smoking ban had no detectable impact on smoking cessation support and medications provided by hospital staff (Etter and Etter, 2007). One may wonder whether staff members responded to the more challenging situation created by the total smoking ban by providing more smoking cessation medications to calm unsatisfied patients. After the total ban, only a minority of smokers who received nicotine medications reported having attempted to quit smoking during their hospital stay, which suggest that these medications were often used by continuing smokers to manage withdrawal symptoms, rather than by recent quitters. Pharmacotherapy combined with counseling is an effective intervention for smoking cessation, and hospitalization is a window of opportunity to provide treatment for tobacco dependence (Emmons and Goldstein, 1992). Even though the smoking bans increased the frequency of tobacco dependence counseling and treatment, compared with the situation when there was no ban, this frequency remained below optimal levels. Specific strategies should be implemented to increase the quality and frequency of smoking cessation interventions provided by hospital staff to psychiatric patients (Addington et al., 1998). Interestingly, patients reported receiving much less smoking cessation support than reported by staff members. It is possible that patient self-reporting underestimated the frequency of smoking cessation support they received, particularly if these were brief interventions. This result could also be explained by social desirability bias if staff members over-reported this activity. The partial smoking ban was well accepted (Etter and Etter, 2007), but the total smoking ban was perceived as too strict by about half the participants, and most preferred a partial ban. Both the partial and total smoking bans proved difficult to enforce, as shown by the fact that a sizeable minority of participants remained exposed to, and annoyed by ETS, even after implementation of the total ban. Many patients reported getting angry at hospital staff after the total ban. However, neither the partial nor the total smoking bans seriously deteriorated the relationship between patients and staff. Similarly, a recent review concluded that although in some cases, psychiatric patients had difficulty complying with total smoking bans, these bans had no major longstanding untoward effects in terms of behavior or compliance (El-Guebaly et al., 2002; Velasco et al., 1996; Greeman and McClellan, 1991). The expenditures needed to install smoking rooms equipped with ventilation and glass doors were unnecessary, given that these smoking rooms were used for less than two years and the total smoking ban did not cause any serious problems, even though it created some dissatisfaction. The money invested in building the smoking rooms would probably have been better invested in communicating the total ban and treating tobacco dependence.
577
Study limitations and strengths We relied on self-reports of patients and staff, which are subject to social desirability bias. We used independentsample t-tests to assess differences between surveys. Given that many hospital members took part in several surveys, and because paired-sample t-tests are statistically more powerful, independent-samples tests are too conservative and may underestimate the statistical significance. The final survey was conducted only three months after implementation of the total ban, which may not reflect the long-term acceptability and impact of this ban. However, acceptability of the partial ban did not improve between 2004 and 2005, which suggests that additional surveys may not reveal useful information. The sample size was relatively small, which increases the risk of type II error. In patients, the response rates were lower in 2005–2006 than in 2003–2004, but the existence and direction of a possible non–response bias is difficult to estimate (Etter and Perneger, 1997). Finally, the study did not include a control group, because the smoking ban was not decided and implemented by the researchers, but by the direction of the hospital in all psychiatric units. A control group would have enabled us to distinguish between the impact of the smoking ban and naturally occurring time trends. Among the strengths of this study, we know of no similar previous evaluation of the successive implementation of a partial and a total smoking bans in a psychiatry unit, and this is also one of very few studies of a total smoking ban in a psychiatric hospital conducted outside the USA (El-Guebaly et al., 2002; Lawn and Pols, 2005). Conclusions A partial smoking ban decreased exposure to ETS and a total smoking ban further improved the situation. In addition, the total ban increased the proportion of smokers who tried to quit smoking during their hospital stay and received nicotine medications, which is an important result, given the difficulty of quitting smoking for psychiatric patients. The total smoking ban was not strictly enforced and was overall acceptable to both patients and staff, but most participants preferred a partial ban. Acknowledgments No competing interests. The authors received no external funding for this study. The authors thank survey respondents for their participation. References Addington, J., el-Guebaly, N., Campbell, W., Hodgins, D.C., Addington, D., 1998. Smoking cessation treatment for patients with schizophrenia. Am. J. Psychiatry 155, 974–976. Boffetta, P., Agudo, A., Ahrens, W., et al., 1998. Multicenter case-control study of exposure to environmental tobacco smoke and lung cancer in Europe. J. Natl. Cancer Inst. 90, 1440–1450. Campion, J., McNeill, A., Checinski, K., 2006. Exempting mental health units from smoke-free laws. BMJ 333, 407–408.
578
M. Etter et al. / Preventive Medicine 46 (2008) 572–578
De Leon, J., Diaz, F.J., 2005. A meta-analysis of worldwide studies demonstrates an association between schizophrenia and tobacco smoking behaviors. Schizophr. Res. 76, 135–157. Downey, K.K., Pomerleau, C.S., Huth, A.C., Silk, K.R., 1998. The effect of a restricted smoking policy on motivation to quit smoking in psychiatric patients. J. Addict. Dis. 17, 1–7. El-Guebaly, N., Cathcart, J., Currie, S., Brown, D., Gloster, S., 2002. Public health and therapeutic aspects of smoking bans in mental health and addiction settings. Psychiatr. Serv. 53, 1617–1622. Emmons, K.M., Goldstein, M.G., 1992. Smokers who are hospitalized, a window of opportunity for cessation interventions. Prev. Med. 21, 262–269. Etter, J.F., Perneger, T.V., 1997. Analysis of non-response bias in a mailed health survey. J. Clin. Epidemiol. 50, 1123–1128. Etter, J.F., Perneger, T.V., 2001. Effectiveness of a computer-tailored smoking cessation program, a randomized trial. Arch. Intern. Med. 161, 2596–2601. Etter, M., Etter, J.F., 2007. Acceptability and impact of a partial smoking ban in a psychiatric hospital. Prev. Med. 44, 64–69. Etter, M., Mohr, S., Garin, C., Etter, J.F., 2004. Stages of change in smokers with schizophrenia or schizoaffective disorder and in the general population. Schizophr. Bull. 30, 459–468. European Network of Smoke-Free Hospitals and Maternity Services (ENSH), 2005. European directory of smoke-free hospitals. ENSH, Paris. Glasgow, R.E., Stevens, V.J., Vogt, T.M., Mullooly, J.P., Lichtenstein, E., 1991. Changes in smoking associated with hospitalization, quit rates, predictive variables, and intervention implications. Am. J. Health Promot. 6, 24–29. Greeman, M., McClellan, T.A., 1991. Negative effects of a smoking ban on an inpatient psychiatry service. Hosp. Community Psychiatry 42, 408–412. Harris, G.T., Parle, D., Gagne, J., 2007. Effects of a tobacco ban on long-term psychiatric patients. J. Behav. Health Serv. Res. 34, 43–55. Hughes, J.R., 1993. Possible effects of smoke-free inpatient units on psychiatric diagnosis and treatment. J. Clin. Psychiatry 54, 109–114.
Humair, J.P., Cornuz, J., 2003. A new curriculum using active learning methods and standardized patients to train residents in smoking cessation. J. Gen. Intern. Med. 18, 1023–1027. Jochelson, K., 2006. Smoke-free legislation and mental health units, the challenges ahead. Br. J. Psychiatry 189, 479–480. Lawn, S., Pols, R., 2005. Smoking bans in psychiatric inpatient settings? A review of the research. Aust. N. Z. J. Psychiatry 39, 866–885. 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. Tob. Control 7, 47–55. Matthews, L.S., Diaz, B., Bird, P., et al., 2005. Implementing a smoking ban in an acute psychiatric admissions unit. J. Psychosoc. Nurs. Ment. Health Serv. 43, 33–36. McKee, M., Gilmore, A., Novotny, T.E., 2003. Smoke free hospitals. BMJ 326, 941–942. McNally, L., Oyefeso, A., Annan, J., et al., 2006. A survey of staff attitudes to smoking-related policy and intervention in psychiatric and general health care settings. J. Public Health (Oxf.) 28, 192–196. Rigotti, N.A., Arnsten, J.H., McKool, K.M., et al., 2000. Smoking by patients in a smoke-free hospital, prevalence, predictors, and implications. Prev. Med. 31, 159–166. Smith, W.R., Grant, B.L., 1989. Effects of a smoking ban on a general hospital psychiatric service. Hosp. Community Psychiatry 40, 497–502. Stubbs, J., Haw, C., Garner, L., 2004. Survey of staff attitudes to smoking in a large psychiatric hospital. Psychiatri. Bull. 28, 204–207. Velasco, J., Eells, T.D., Anderson, R., et al., 1996. A two-year follow-up on the effects of a smoking ban in an inpatient psychiatric service. Psychiatr. Serv. 47, 869–871. Willemsen, M.C., Gort, C.A., Van Soelen, P., 2004. Exposure to environmental tobacco smoke (ETS) and determinants of support for complete smoking bans in psychiatric settings. Tob. Control 13, 180–185.
Preventive Medicine 46 (2008) 572 – 578 www.elsevier.com/locate/ypmed
Acceptability and impact of a partial smoking ban followed by a total smoking ban in a psychiatric hospital Manuela Etter a , Aqal Nawaz Khan a , Jean-François Etter b,⁎ a b
Department of Psychiatry, Geneva University Hospitals, Geneva, Switzerland Institute of Social and Preventive Medicine, University of Geneva, Switzerland Available online 26 January 2008
Abstract Objective. To assess the impact of a partial smoking ban followed by a total smoking ban in a psychiatric hospital in Switzerland. Methods. In 2003, smoking was allowed everywhere in psychiatric units. In 2004, smoking was prohibited everywhere except in smoking rooms. In 2006, smoking rooms were removed and smoking was totally prohibited indoors. Patients and staff were surveyed in 2003 (n = 106), 2004 (n = 108), 2005 (n = 119) and 2006 (n = 134). Results. Exposure to environmental tobacco smoke (ETS) decreased after the partial ban and further decreased after the total ban. Among patients, after the total ban, more smokers attempted to quit smoking (18%) relative to before the total ban (2%, odds ratio = 10.1, p = 0.01). More smokers said that hospital staff gave them nicotine replacement products after the total ban (52%), compared with before (13%, odds ratio = 7.6, p b 0.001). Many participants (55%) commented that the total ban was too strict, and most (64%) preferred the partial ban. Conclusions. The partial ban decreased exposure to ETS and the total ban further improved the situation and increased the proportion of smokers who attempted to quit smoking and received nicotine medications. The total ban was loosely enforced and was overall acceptable, but most participants preferred a partial ban. © 2008 Elsevier Inc. All rights reserved. Keywords: Smoking; Nicotine dependence; Smoking cessation; Psychiatry, Preventive measures
Introduction Smoking prevalence is much higher in patients with psychiatric disorders relative to the general population (De Leon and Diaz, 2005). The high frequency of smoking in psychiatric units is a cause of exposure to environmental tobacco smoke (ETS), a recognized carcinogen (Boffetta et al., 1998). It also results in annoyance due to the smell of tobacco and may incite smoking in patients. Banning smoking in hospitals protects non smokers from ETS and protects non smokers and former smokers from cues to smoke, thus reducing the risk of smoking initiation or relapse during hospital stays (Glasgow et al., 1991; Hughes, 1993; Rigotti et al., 2000). Also, it reduces fires and cleaning costs (McKee et al., 2003). For all these reasons, U.S. hospitals ⁎ Corresponding author. Institute of Social and Preventive Medicine, University of Geneva, CMU, case postale, CH-1211 Geneva 4, Switzerland. Fax: +41 22 379 59 12. E-mail address: [email protected] (J.-F. Etter). 0091-7435/$ - see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2008.01.004
are required to be smoke free and nearly all have complied (Longo et al., 1998). In Europe, the situation is quite different, because most hospitals have only implemented partial smoking bans (ENSH, 2005; Willemsen et al., 2004). In the U.K., hospitals have to be smoke-free under the Health Act 2006, but psychiatric hospitals may be exempted from this regulation for several years (Jochelson, 2006; Campion et al., 2006). Smoking bans have been implemented less often in psychiatry departments than in somatic care departments because of a fear that psychiatric patients would be more reluctant to comply with smoking bans (McNally et al., 2006; Stubbs et al., 2004). Research showed, however, that smoking bans were in general well accepted in psychiatric inpatient units (El-Guebaly et al., 2002; Harris et al., 2007), and that staff may have anticipated more problems than actually occurred (Lawn and Pols, 2005; Smith and Grant, 1989; Matthews et al., 2005). In a previous article, we examined the impact and acceptability of a partial smoking ban in a psychiatric hospital in Switzerland (Etter and Etter, 2007). We showed that this partial smoking ban
M. Etter et al. / Preventive Medicine 46 (2008) 572–578
573
rooms in each unit, one for patients and one for hospital staff. A steering committee, which consisted of physicians, nurses, psychologists and social workers, implemented the partial smoking ban. Despite the fact that a majority of both patients and staff rejected the idea of a total smoking ban (Etter and Etter, 2007), the hospital administration decided that smoking would be banned everywhere inside hospital buildings beginning January 2006. The smoking rooms were then removed. Smoking continued to be allowed outdoors, and patients (except those in locked rooms) and staff were allowed to leave the unit to smoke outdoors. Before implementation of the smoking bans, health professionals were trained to treat tobacco dependence in their patients, booklets were distributed to smokers, and no-smoking signs were posted throughout the buildings (Etter and Perneger, 2001; Humair and Cornuz, 2003). Smokers who wanted to quit were offered advice and nicotine replacement therapy (NRT). NRT was included in the fixed-price in-patient contract and was therefore available at no additional cost to patients, but not to hospital staff. Bupropion (Zyban) was not included in the in-patient contract, and varenicline was not yet available.
was well accepted and decreased exposure to ETS, despite the fact that it was not strictly enforced. Two years later, a total smoking ban was implemented in the same hospital. The acceptability and impact of total smoking bans in psychiatry hospitals is incompletely documented, in particular in Europe. A recent review found only seven empirical studies on the effects of total smoking bans in psychiatry in-patient units, and all of them were from the USA (El-Guebaly et al., 2002). Another recent review found only 10 studies of total smoking bans in psychiatric settings (Lawn and Pols, 2005). Our aim was to compare the acceptability and efficacy of a partial smoking ban and total ban in an in-patient psychiatric hospital. Methods Setting
Study design
The smoking ban was implemented in the whole Psychiatry Department of the Geneva University Hospitals (10 units, 166 beds), but the study was conducted only in two in-patient, adult units of this Department. The first was an admission and short-stay unit (16 beds, mean duration of stays = 17 days, median = 7 days) and the second was a medium-stay unit (16 beds, mean duration of stays = 37 days, median = 15 days). Patients had mainly psychotic disorders, depression and personality disorders.
We conducted four surveys with the same questions. The first survey took place in October 2003 (before the partial ban), the second in April 2004 (2 months after implementation of the partial ban), the third from October to December 2005 (20 months after the partial ban, but before the total ban) and the fourth from March to May 2006 (3–5 months after the total smoking ban). The questionnaires were brief and self-administered. A physician, a nurse or a psychologist distributed the questionnaires to patients and staff after explaining the study and obtaining written informed consent. Patients answered the survey as soon as their condition allowed for, which in most cases was about one week after admission. The physician, nurse or psychologist completed the questionnaires with patients who were unable to answer by themselves. All questionnaires were anonymous. The third and fourth surveys were approved by the ethics committee of the Geneva University Hospitals, but the first two surveys were not submitted to an ethics committee. Otherwise, the study followed the principles in the Declaration of Helsinki.
Intervention In 2003, patients were allowed to smoke everywhere except in bedrooms and dining rooms, but this policy was not enforced and in fact, patients smoked everywhere. In February 2004, a partial smoking ban was implemented; smoking was banned everywhere except in closed smoking rooms equipped with ventilation and glass doors for surveillance. There were two smoking
Table 1 Characteristics of participants, before and after introduction of a partial and a total smoking ban (Switzerland, 2003–2006)
Patients N participants Intended sample Participation rate (%) Age (mean), years % men Ever smoked 100+ cigarettes Smoking status (%) – Daily smokers – Occasional (non-daily) smokers – Former smokers – Never smokers Staff N participants Intended sample Participation rate Age (mean), years % men Ever smoked 100+ cigarettes (%) Smoking status (%) – Daily smokers – Occasional (non-daily) smokers – Former smokers – Never smokers
No ban 2003
Partial ban 2004
Partial ban 2005
Total ban 2006
p value 2003–2004
p value 2003–2006
p value 2005–2006
49 57 86.0 39.9 59.2 91.8
54 67 80.6 39.5 64.8 74.1
66 124 53.2 42.2 48.5 78.8
77 114 67.5 41.0 60.0 81.6
– – 0.42 0.89 0.57 0.018 0.038
– – 0.010 0.68 0.93 0.11 0.43
– – 0.024 0.28 0.17 0.67 0.25
73.5 6.1 12.2 8.2
64.8 0 9.3 25.9
68.8 6.3 6.3 18.8
65.8 2.6 15.8 15.8
57 57 100 38.8 35.1 64.9
54 55 98.2 38.5 37.0 63.0
53 63 84.1 38.7 35.3 56.6
57 62 91.9 40.7 37.5 57.9
– – 0.31 0.86 0.83 0.83 0.97
– – 0.028 0.41 0.79 0.44 1.0
– – 0.18 0.47 0.81 0.89 0.77
26.3 7.0 22.8 43.9
27.8 5.6 20.4 46.3
24.5 11.3 17.0 47.2
26.3 7.0 22.8 43.9
574
M. Etter et al. / Preventive Medicine 46 (2008) 572–578
Participants
Awareness and satisfaction with the non-smoking policy
The target sample included all patients and staff present at the time of data collection. Given the short duration of most hospital stays and the resulting patient turnover, the samples of patients in the successive surveys consisted mostly of different people, whereas the samples of hospital staff consisted largely of the same people who answered successive surveys.
In 2005, 89% of patients and 94% of staff correctly answered that smoking was allowed only in smoking rooms. In 2006, 90% of patients and 93% of staff correctly answered that smoking was prohibited everywhere in the clinic. In 2003, 2004 and 2005, most participants answered that the rules about smoking in the clinic were adequate, but in 2006 (after the total ban) most answered that these rules were too strict (2005: 14.3%, 2006: 55.0%, p b 0.001) (Table 2). In 2003 (before the partial ban), half the participants (51.6%) answered that the rules about smoking in the clinic were not respected, whereas only 17.6% gave this answer in 2004 (p b 0.001), but this figure increased after introduction of the total ban (2005: 16.1%, 2006: 32.6%, p = 0.021) (Table 2). A minority of participants favored a total smoking ban (19.7% in 2005 and 27.8% in 2006, p = 0.09) and most preferred a partial ban with closed smoking rooms (77.8% in 2005 and 64.3% in 2006, p = 0.09). Nevertheless, most participants in the 2006 survey (52.8%) agreed with the statement, “I am pleased with the smoking prohibition at the clinic”. In all four surveys, with no statistically significant change after introduction of the partial or total ban, most staff members (68–83%) and one third to one half of patients (38–56%) agreed with the statement, ”At the hospital, tobacco smoke is a source of conflict with (other) patients”. More patients agreed with the statement, “tobacco smoke is a source of conflict with hospital staff” in 2006, after the total ban (36.4%) than in 2003, when smoking was allowed everywhere (24.7%, p = 0.005). A substantial minority of patients and staff agreed with the statement:
Measurements The questionnaires covered age, sex and smoking status (Table 1), opinions about the no-smoking policy (Table 2), perceived exposure to ETS (Table 3), smoking behavior and smoking cessation interventions received from hospital staff (Table 4). Hospital staff indicated whether they advised and helped patients to quit smoking, or gave them nicotine replacement medications.
Statistical analyses We used chi-square tests and odds ratios to compare proportions, and independent-sample t tests to compare means.
Results Participation There were 106 participants in 2003, 108 in 2004, 119 in 2005 and 134 in 2006. The intended samples were smaller in 2003–2004 than in 2005–2006 because data were collected during only one month in 2003–2004 and during two months in 2005–2006 (Table 1). Among patients and staff, participation rates were higher in 2003 and 2004 than in 2005 and 2006. The 2003 and 2004 data were reported previously (Etter and Etter, 2007).
Table 2 Opinions about smoking at the hospital, before and after introduction of a partial and a total smoking bans (Switzerland, 2003–2006)
Patients Rules about smoking… are too strict … are adequate “At the hospital…” (% agree) Rules on smoking are not respected Tobacco smoke is a source of conflict with other patients Tobacco smoke is a source of conflict with hospital staff Cohabitation between smokers and non-smokers is very difficult Staff Rules about smoking are… too strict … are adequate “At the hospital…” (% agree) Rules on smoking are not respected Tobacco smoke is a source of conflict with patients Cohabitation between smokers and non-smokers is very difficult All (patients and staff) Rules about smoking are… too strict … are adequate “At the hospital…” (% agree) Rules on smoking are not respected Tobacco smoke is a source of conflict with (other) patients Tobacco smoke is a source of conflict with (other) hospital staff Cohabitation between smokers and non-smokers is very difficult
No ban 2003
Partial ban 2004
Partial ban 2005
Total ban 2006
p value 2003–2004
p value 2003–2006
p value 2005–2006
12.2 73.5
14.8 79.6
15.2 77.3
49.4 46.8
0.32
b0.001
b0.001
40.8 49.0 24.7 32.6
16.7 40.7 24.1 29.6
10.6 56.1 42.6 53.4
22.1 37.7 36.4 44.9
0.023 0.16 0.68 0.70
0.06 0.56 0.005 0.033
7.0 71.9
14.8 81.5
13.2 75.5
59.6 36.8
0.014
b0.001
b0.001
59.9 82.5 54.4
18.6 74.1 31.5
22.7 67.9 35.8
45.6 80.4 43.9
b0.001 0.44 0.19
0.12 0.45 0.027
0.066 0.31 0.53
9.4 72.6
14.8 80.6
14.3 76.5
55.0 43.5
0.006
b0.001
b0.001
51.6 67.0 31.2 44.3
17.6 57.4 27.7 30.5
16.1 62.9 38.1 45.9
32.6 58.7 38.5 44.4
b0.001 0.30 0.46 0.30
0.026 0.24 0.009 0.004
0.021 0.63 0.88 0.82
0.14 0.16 0.62 0.86
M. Etter et al. / Preventive Medicine 46 (2008) 572–578
575
Table 3 Exposure to and annoyance from environmental tobacco smoke (ETS) among non-smokers (patients and staff), before and after introduction of the partial and total smoking bans (Switzerland, 2003–2006) No ban 2003
Partial ban 2004
Patients Exposure to ETS in unit (% “never”) – in bedrooms 69.2 89.5 – in dining rooms 30.8 63.2 – in corridors 23.1 47.4 Annoyance due to ETS in unit (% “absolutely not”) – in bedrooms 61.5 78.9 – in dining rooms 38.5 73.7 – in corridors 38.5 63.2 Staff Exposure to ETS in unit (%“never”) – in bedrooms 16.7 38.5 – in dining rooms 26.2 69.2 – in corridors 9.5 23.1 Annoyance due to ETS in unit (% “absolutely not”) – in bedrooms 23.8 38.5 – in dining rooms 31.0 41.0 – in corridors 23.8 35.9 All (patients and staff) Exposure to ETS in unit (% “never”) – in bedrooms 25.0 54.5 – in dining rooms 35.5 65.5 – in corridors 10.4 30.9 Annoyance due to ETS in unit (% “absolutely not”) – in bedrooms 29.2 52.7 – in dining rooms 35.0 52.7 – in corridors 27.1 45.5
Partial ban 2005
Total ban 2006
p value 2003–2004
p value 2003–2006
p value 2005–2006
60.0 65.0 30.0
88.5 73.1 65.4
0.18 0.17 0.19
0.058 0.09 0.029
0.028 0.48 0.06
75.0 50.0 45.0
76.9 80.8 69.2
0.45 0.16 0.07
0.108 0.007 0.162
0.37 0.061 0.051
35.0 55.0 20.0
31.0 71.4 38.1
0.14 0.054 0.018
0.041 0.004 0.006
0.21 0.39 0.30
47.5 65.0 40.0
45.2 81.0 52.4
0.24 0.010 0.10
0095 b0.001 0.023
0.89 0.43 0.53
51.2 73.8 26.1
55.2 83.3 51.7
0.046 0.037 0.001
0.013 b0.001 b0.001
0.12 0.32 0.045
62.8 71.4 44.7
61.4 90.0 62.7
0.083 0.023 0.068
0.013 b0.001 0.002
0.98 0.045 0.067
“At the hospital, cohabitation between smokers and nonsmokers is very difficult”. Before and after the total smoking ban, most participants (75–78%) agreed with the statement, “For patients in locked rooms, the smoking prohibition is very hard to bear”. The total ban increased the proportion of patients who reported that they “sometimes” or “often” got angry with hospital staff because of the smoking policy (4.5% in 2005 and 24.5% in 2006, p = 0.02, odds ratio = 6.8, 95% confidence interval = 1.2 to 47.3). In parallel, there was a non-significant increase in the proportion of staff members who reported that patients “sometimes” or “often” got angry with them because of the nosmoking rules (47.2% in 2005 and 62.2% in 2006, p = 0.29), and most staff (61.5%) commented that some patients refused to abide by these rules. There was, however, no significant change between 2005 and 2006 in answers from staff members to the question: “If we totally prohibited smoking in the clinic, we would face strong protest from patients” (32.7% agreed in 2005 and 42.8% in 2006, p = 0.28). After the total ban (these questions were not asked in 2005), most staff members (80.7%) reported that some patients smoked in the bedrooms and left the clinic to go out and buy cigarettes (82.4%). Perceived exposure to ETS Across all surveys, staff members reported more exposure to, and annoyance from ETS than patients. Among non-smokers
(patients and staff, Table 3), exposure to ETS decreased in bedrooms after the partial ban, but did not further decrease after the total ban. On third (31%) of the non-smokers still commented that they were “often” or “sometimes” exposed to ETS in bedrooms after introduction of the total ban. Exposure to ETS in dining rooms and corridors decreased after the partial ban and further decreased after the total ban (Table 3). After the total ban, no participant reported that they were “often” exposed to ETS in dining rooms and offices, but 12.0% remained “often” exposed to ETS in corridors. In non-smokers, the average duration of exposure to ETS in the clinic, as self-reported in questionnaires, decreased significantly after the total smoking ban (from 69 min per day in 2005 to 12 min in 2006, p = 0.012). Annoyance from ETS Annoyance caused by ETS in bedrooms decreased after introduction of the partial ban, but did not further decrease after the total ban (Table 3). In contrast, the partial ban decreased levels of annoyance caused by ETS in dining rooms and corridors, and the situation further improved after implementation of the total smoking ban. Nevertheless, after the total ban, 15.8% of non-smokers remained “a lot” or “somewhat” annoyed by ETS in bedrooms and 13.6% in corridors. The situation was best in dining rooms, where only 1.8% of participants were “a lot” or “somewhat” annoyed by ETS after the total ban.
576
M. Etter et al. / Preventive Medicine 46 (2008) 572–578
Table 4 Smoking behavior and smoking cessation interventions, among patients who smoked, before and after introduction of the partial and the total smoking bans (Switzerland, 2003–2006)
Among patients, daily and occasional smokers: Cigarettes/day now, mean Cigarettes/day before admission, mean Since admission in the clinic, smokes (%): – more than before admission – no change – less than before admission During your current hospital stay, did a physician or a nurse… (% Yes) – advise you to quit smoking? – help you to quit smoking? – provide medication to quit smoking (like a nicotine patch or gum or Zyban)? At the clinic… (% sometimes + often) – Do other patients provide you cig.? – Do hospital staff provide you cigarettes? – Do hospital staff forbid you to smoke?
No ban 2003
Partial ban 2004
Partial ban 2005
Total ban 2006
p value 2003–2004
24.1 24.3
33.1 29.4
22.0 24.0
23.7 24.6
0.015 0.21 0.065
0.81 0.17 0.21
0.51 0.87 0.96
42.2 13.2 43.7
65.7 20.0 14.3
42.2 26.7 31.1
39.6 29.2 31.3
15.4 2.6 5.1
5.7 5.7 11.4
30.0 15.0 12.5
42.6 19.6 52.2
0.18 0.49 0.32
0.006 0.015 b0.001
0.23 0.58 b0.001
43.6 10.3 5.1
45.7 2.9 14.3
42.2 11.3 11.3
52.2 14.6 22.4
0.98 0.45 0.20
0.14 0.61 0.21
0.26 0.75 0.61
Smoking behavior The total smoking ban was not followed by any change in smoking prevalence or cigarette consumption (Table 1), but it was followed by an increase in the proportion of patients who attempted to quit smoking during their hospital stay (from 2.2% in 2005 to 18.4% in 2006, p = 0.01, odds ratio = 10.1, 95% confidence interval 1.21 to 222.7). Among patients, a substantial minority of smokers (40% in 2006) said they smoked more at the clinic than before their hospital admission, but most said they smoked the same amount or less, with no significant change between surveys (Table 4). Cigarette exchange among patients was frequent, and both patients and staff reported that staff members sometimes provided cigarettes to patients (Table 4). Between 2003 and 2004, there was an increase in the proportion of never smokers and, among smokers, there was an increase in cigarette consumption. These results should however be considered with caution (see Etter and Etter, 2007 for a discussion of this point). Smoking cessation interventions, as reported by patients Among patients who smoked, there was no statistically significant difference between the 2003–2004 surveys and the 2005–2006 surveys in answers to questions on whether hospital staff advised them or helped them to quit smoking (Table 4). However, there was an important improvement in the proportion of patients who received advice to quit smoking from a staff member between 2003 (no ban, 15.4%) and 2006 (total ban, 42.6%, p = 0.006, odds ratio = 4.1, 95% confidence interval = 1.4 to 11.6), or who received help to quit smoking from a staff member (from 2.6% in 2003 to 19.6% in 2006, p = 0.015, odds ratio = 9.2, 95% confidence interval 1.1 to 76.6). After implementation of the total ban, there was an increase in the proportion of smokers who commented that hospital staff gave
p value 2003–2006
p value 2005–2006
them nicotine replacement medications (patch, gum) (from 12.5% in 2005 to 52.2% in 2006, p b .001, odds ratio = 7.6, 95% confidence interval = 2.3 to 27.0). Patients reported receiving smoking cessation support less frequently than was reported by staff members. Smoking cessation interventions, as reported by hospital staff These questions were asked only in 2005 and 2006. According to hospital staff, the total smoking ban was followed by a doubling in the proportion of patients to whom help was provided to quit smoking (from 26.9% in 2005 to 58.2% in 2006, p = 0.007, odds ratio = 3.8, 95% confidence interval = 1.6 to 9.3) and by an increase in the proportion of smokers to whom staff members provided nicotine replacement medications (from 42.3% in 2005 to 74.5% in 2006, p b 0.001, odds ratio = 4.0, 95% confidence interval = 1.6 to 9.9). Discussion The originality of this study lies in its four time points, which enabled us to compare, in the same psychiatric in-patient clinic, a situation where smoking was allowed almost without restriction, with a partial smoking ban (when smoking was allowed only in closed, ventilated smoking rooms), and with a total smoking ban, when smoking was, in principle, completely banned inside hospital buildings. Both the partial and the total smoking bans decreased exposure to and annoyance from ETS, which was their main objective. Compared with the situation in 2003, when smoking was in fact unrestricted, the total smoking ban represented an important improvement, even though it was only loosely enforced. Importantly, the total ban (but not the partial ban) was followed by a substantial increase in the proportion of smokers who attempted to quit smoking during their hospital stay (from 2% to 18%). This is an important result, as it is usually
M. Etter et al. / Preventive Medicine 46 (2008) 572–578
difficult for psychiatric patients to quit smoking (Etter et al., 2004), and because smoking restrictions do not always motivate psychiatric patients to quit smoking (Downey et al., 1998). The total smoking ban was also followed by a substantial increase in the proportion of patients who received nicotine medications. This result seems robust, since it was observed in answers for both patients and staff. In contrast, the partial smoking ban had no detectable impact on smoking cessation support and medications provided by hospital staff (Etter and Etter, 2007). One may wonder whether staff members responded to the more challenging situation created by the total smoking ban by providing more smoking cessation medications to calm unsatisfied patients. After the total ban, only a minority of smokers who received nicotine medications reported having attempted to quit smoking during their hospital stay, which suggest that these medications were often used by continuing smokers to manage withdrawal symptoms, rather than by recent quitters. Pharmacotherapy combined with counseling is an effective intervention for smoking cessation, and hospitalization is a window of opportunity to provide treatment for tobacco dependence (Emmons and Goldstein, 1992). Even though the smoking bans increased the frequency of tobacco dependence counseling and treatment, compared with the situation when there was no ban, this frequency remained below optimal levels. Specific strategies should be implemented to increase the quality and frequency of smoking cessation interventions provided by hospital staff to psychiatric patients (Addington et al., 1998). Interestingly, patients reported receiving much less smoking cessation support than reported by staff members. It is possible that patient self-reporting underestimated the frequency of smoking cessation support they received, particularly if these were brief interventions. This result could also be explained by social desirability bias if staff members over-reported this activity. The partial smoking ban was well accepted (Etter and Etter, 2007), but the total smoking ban was perceived as too strict by about half the participants, and most preferred a partial ban. Both the partial and total smoking bans proved difficult to enforce, as shown by the fact that a sizeable minority of participants remained exposed to, and annoyed by ETS, even after implementation of the total ban. Many patients reported getting angry at hospital staff after the total ban. However, neither the partial nor the total smoking bans seriously deteriorated the relationship between patients and staff. Similarly, a recent review concluded that although in some cases, psychiatric patients had difficulty complying with total smoking bans, these bans had no major longstanding untoward effects in terms of behavior or compliance (El-Guebaly et al., 2002; Velasco et al., 1996; Greeman and McClellan, 1991). The expenditures needed to install smoking rooms equipped with ventilation and glass doors were unnecessary, given that these smoking rooms were used for less than two years and the total smoking ban did not cause any serious problems, even though it created some dissatisfaction. The money invested in building the smoking rooms would probably have been better invested in communicating the total ban and treating tobacco dependence.
577
Study limitations and strengths We relied on self-reports of patients and staff, which are subject to social desirability bias. We used independentsample t-tests to assess differences between surveys. Given that many hospital members took part in several surveys, and because paired-sample t-tests are statistically more powerful, independent-samples tests are too conservative and may underestimate the statistical significance. The final survey was conducted only three months after implementation of the total ban, which may not reflect the long-term acceptability and impact of this ban. However, acceptability of the partial ban did not improve between 2004 and 2005, which suggests that additional surveys may not reveal useful information. The sample size was relatively small, which increases the risk of type II error. In patients, the response rates were lower in 2005–2006 than in 2003–2004, but the existence and direction of a possible non–response bias is difficult to estimate (Etter and Perneger, 1997). Finally, the study did not include a control group, because the smoking ban was not decided and implemented by the researchers, but by the direction of the hospital in all psychiatric units. A control group would have enabled us to distinguish between the impact of the smoking ban and naturally occurring time trends. Among the strengths of this study, we know of no similar previous evaluation of the successive implementation of a partial and a total smoking bans in a psychiatry unit, and this is also one of very few studies of a total smoking ban in a psychiatric hospital conducted outside the USA (El-Guebaly et al., 2002; Lawn and Pols, 2005). Conclusions A partial smoking ban decreased exposure to ETS and a total smoking ban further improved the situation. In addition, the total ban increased the proportion of smokers who tried to quit smoking during their hospital stay and received nicotine medications, which is an important result, given the difficulty of quitting smoking for psychiatric patients. The total smoking ban was not strictly enforced and was overall acceptable to both patients and staff, but most participants preferred a partial ban. Acknowledgments No competing interests. The authors received no external funding for this study. The authors thank survey respondents for their participation. References Addington, J., el-Guebaly, N., Campbell, W., Hodgins, D.C., Addington, D., 1998. Smoking cessation treatment for patients with schizophrenia. Am. J. Psychiatry 155, 974–976. Boffetta, P., Agudo, A., Ahrens, W., et al., 1998. Multicenter case-control study of exposure to environmental tobacco smoke and lung cancer in Europe. J. Natl. Cancer Inst. 90, 1440–1450. Campion, J., McNeill, A., Checinski, K., 2006. Exempting mental health units from smoke-free laws. BMJ 333, 407–408.
578
M. Etter et al. / Preventive Medicine 46 (2008) 572–578
De Leon, J., Diaz, F.J., 2005. A meta-analysis of worldwide studies demonstrates an association between schizophrenia and tobacco smoking behaviors. Schizophr. Res. 76, 135–157. Downey, K.K., Pomerleau, C.S., Huth, A.C., Silk, K.R., 1998. The effect of a restricted smoking policy on motivation to quit smoking in psychiatric patients. J. Addict. Dis. 17, 1–7. El-Guebaly, N., Cathcart, J., Currie, S., Brown, D., Gloster, S., 2002. Public health and therapeutic aspects of smoking bans in mental health and addiction settings. Psychiatr. Serv. 53, 1617–1622. Emmons, K.M., Goldstein, M.G., 1992. Smokers who are hospitalized, a window of opportunity for cessation interventions. Prev. Med. 21, 262–269. Etter, J.F., Perneger, T.V., 1997. Analysis of non-response bias in a mailed health survey. J. Clin. Epidemiol. 50, 1123–1128. Etter, J.F., Perneger, T.V., 2001. Effectiveness of a computer-tailored smoking cessation program, a randomized trial. Arch. Intern. Med. 161, 2596–2601. Etter, M., Etter, J.F., 2007. Acceptability and impact of a partial smoking ban in a psychiatric hospital. Prev. Med. 44, 64–69. Etter, M., Mohr, S., Garin, C., Etter, J.F., 2004. Stages of change in smokers with schizophrenia or schizoaffective disorder and in the general population. Schizophr. Bull. 30, 459–468. European Network of Smoke-Free Hospitals and Maternity Services (ENSH), 2005. European directory of smoke-free hospitals. ENSH, Paris. Glasgow, R.E., Stevens, V.J., Vogt, T.M., Mullooly, J.P., Lichtenstein, E., 1991. Changes in smoking associated with hospitalization, quit rates, predictive variables, and intervention implications. Am. J. Health Promot. 6, 24–29. Greeman, M., McClellan, T.A., 1991. Negative effects of a smoking ban on an inpatient psychiatry service. Hosp. Community Psychiatry 42, 408–412. Harris, G.T., Parle, D., Gagne, J., 2007. Effects of a tobacco ban on long-term psychiatric patients. J. Behav. Health Serv. Res. 34, 43–55. Hughes, J.R., 1993. Possible effects of smoke-free inpatient units on psychiatric diagnosis and treatment. J. Clin. Psychiatry 54, 109–114.
Humair, J.P., Cornuz, J., 2003. A new curriculum using active learning methods and standardized patients to train residents in smoking cessation. J. Gen. Intern. Med. 18, 1023–1027. Jochelson, K., 2006. Smoke-free legislation and mental health units, the challenges ahead. Br. J. Psychiatry 189, 479–480. Lawn, S., Pols, R., 2005. Smoking bans in psychiatric inpatient settings? A review of the research. Aust. N. Z. J. Psychiatry 39, 866–885. 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. Tob. Control 7, 47–55. Matthews, L.S., Diaz, B., Bird, P., et al., 2005. Implementing a smoking ban in an acute psychiatric admissions unit. J. Psychosoc. Nurs. Ment. Health Serv. 43, 33–36. McKee, M., Gilmore, A., Novotny, T.E., 2003. Smoke free hospitals. BMJ 326, 941–942. McNally, L., Oyefeso, A., Annan, J., et al., 2006. A survey of staff attitudes to smoking-related policy and intervention in psychiatric and general health care settings. J. Public Health (Oxf.) 28, 192–196. Rigotti, N.A., Arnsten, J.H., McKool, K.M., et al., 2000. Smoking by patients in a smoke-free hospital, prevalence, predictors, and implications. Prev. Med. 31, 159–166. Smith, W.R., Grant, B.L., 1989. Effects of a smoking ban on a general hospital psychiatric service. Hosp. Community Psychiatry 40, 497–502. Stubbs, J., Haw, C., Garner, L., 2004. Survey of staff attitudes to smoking in a large psychiatric hospital. Psychiatri. Bull. 28, 204–207. Velasco, J., Eells, T.D., Anderson, R., et al., 1996. A two-year follow-up on the effects of a smoking ban in an inpatient psychiatric service. Psychiatr. Serv. 47, 869–871. Willemsen, M.C., Gort, C.A., Van Soelen, P., 2004. Exposure to environmental tobacco smoke (ETS) and determinants of support for complete smoking bans in psychiatric settings. Tob. Control 13, 180–185.