Preventive Medicine 76 (2015) 74–78
Contents lists available at ScienceDirect
Preventive Medicine journal homepage: www.elsevier.com/locate/ypmed
Effect of smoke-free patio policy of restaurants and bars on exposure to second-hand smoke Sunday Azagba ⁎ Propel Centre for Population Health Impact, University of Waterloo, Canada School of Public Health and Health Systems, University of Waterloo, Canada
a r t i c l e
i n f o
Available online 22 April 2015 Keywords: Smoke-free policy Outdoor patio Second-hand smoke Environmental tobacco smoke Passive smoking
a b s t r a c t Objective. While there is increasing support for restricting smoking in restaurant and bar patios, there is limited evidence on the effectiveness of this policy. This study examined the effect of smoke-free patio policy of restaurants and bars on adult second-hand smoke (SHS) exposure. Methods. Data were drawn from the 2005–2012 Canadian Tobacco Use Monitoring Survey (n = 89,743), a repeated cross-sectional survey of youth and adult. Regression analysis, a quasi-experimental design was used to examine the effect of provincial smoke-free patio policy on self-reported exposure to SHS. Results. Analyses suggest that exposure to SHS on patios of bars and restaurants declined following the adoption of provincial smoke-free patio policy. Relative to pre-policy SHS exposure, regression results showed a reduction in the probability of SHS exposure of up to 25% in Alberta. Similarly, in Nova Scotia, the probability of SHS exposure declined by up to 21%. Analyses stratified by smoking status found similar significant effect on both smokers and non-smokers. Conclusions. Findings suggest that provincial patio smoking ban on bars and restaurants had the intended effect of protecting non-smokers from SHS exposure. This study is consistent with a large body of evidence showing that a strong smoke-free legislation is an effective public health measure. © 2015 The Author. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Introduction Globally, over 600,000 deaths each year are linked to exposure to second-hand smoke (SHS), also known as environmental tobacco smoke or tobacco smoke pollution (WHO, 2009; Öberg et al., 2011). SHS has been documented as a risk factor for several diseases such as lung cancer, cardiovascular disease, respiratory problems, asthma and sudden infant death syndrome (U.S. Department of Health and Human Services, 2006; U.S. Environmental Protection Agency, 1992). The International Agency for Research on Cancer as well as the U.S. Environmental Protection Agency has designated second-hand smoke as a human carcinogen for which there is no risk-free level (International Agency for Research on Cancer, 2004; U.S. Environmental Protection Agency, 1992). The U.S. Surgeon General report has also noted that a small amount of SHS exposure can adversely affect the heart, blood and blood vessels (U.S. Department of Health and Human Services, 2006). Public smoking restriction has become ubiquitous in many jurisdictions, in part due to the heightened awareness of the harmful health consequences of SHS exposure (U.S. Department of Health and Human ⁎ Propel Centre for Population Health Impact, Faculty of Applied Health Sciences, University of Waterloo, 200 University Avenue West, Waterloo, Ontario N2L 3G1, Canada. E-mail address: [email protected].
Services, 2006; WHO, 2009). The balance of evidence from the preponderance of research suggests that smoke-free legislation is an important public health intervention measure for protecting both non-smokers and smokers from the adverse health effects of SHS exposure (Akhtar et al., 2009; Bondy et al., 2009; Fong et al., 2006; IARC, 2009; Mulcahy et al., 2005; Repace et al., 2013). A large body of evidence has shown significant reductions in cotinine (a metabolite of nicotine) levels among non-smokers following smoke-free policy implementation in many countries including the U.S. (Jensen et al., 2010; Farrelly et al., 2005), Canada (Bondy et al., 2009), Ireland (Mulcahy et al., 2005), Scotland (Akhtar et al., 2009), Italy (Valente et al., 2007), and New Zealand (Fernando et al., 2007). In addition, some population-based survey studies suggest that smoke-free policy is strongly associated with smoking cessation behaviors (Callinan et al., 2010; Hahn et al., 2008; Fowkes et al., 2008; Azagba and Asbridge, 2013; Nagelhout et al., 2011). For example, a review study by Mills and colleagues found consistent evidence, at a populationlevel, that smoking bans promote smoking cessation and decreased cigarette consumption in adult smokers (Mills et al., 2009). While public support is increasing for outdoor smoke-free legislation (Thomson et al., 2009), not much is known about SHS exposure in outdoor dining areas of bars and restaurants (St. Helen et al., 2012; Klepeis et al., 2007). Prior research has demonstrated similar levels of tobacco smoke concentrations in outdoor and indoor settings, especially
http://dx.doi.org/10.1016/j.ypmed.2015.04.012 0091-7435/© 2015 The Author. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
S. Azagba / Preventive Medicine 76 (2015) 74–78
in a close proximity to smokers (Klepeis et al., 2007). In a crossover study, St. Helen et al. (2012) found a significant increase in salivary cotinine and NNAL (a metabolite of tobacco-related nitrosamines) in nonsmokers following SHS exposure in outdoor patios of a restaurant and bar. These findings underscore the potential public health benefit of banning smoking in outdoor patios. In Canada, four provinces (Newfoundland and Labrador, Nova Scotia, Alberta and Ontario) and Yukon Territory have implemented full province-wide smoking bans on outdoor patios of bars and restaurants. Ontario, the most populated province in Canada implemented province-wide ban in January, 2015. Other provinces (Prince Edward Island, New Brunswick, Quebec, Manitoba, Saskatchewan, Ontario and British Columbia) have implemented weak or no bans (e.g., Prince Edward Island outdoor patio ban is not enforced between 10 pm and 3 am, see Table 1 for details). In the United States, a number of states and local municipalities have also implemented smoke-free outdoor dining policy including bar patios (American Nonsmokers' Rights Foundation, 2014). As the number of jurisdictions implementing smoking bans on outdoor patios of bars and restaurants is increasing, to my knowledge, no study has evaluated the effect of this policy on SHS exposure. This may not be surprising given that only few jurisdictions have so far implemented this policy, and also evaluation of this policy will require collecting information on venue-specific SHS exposure. The Canadian Tobacco Use Monitoring Survey (CTUMS) provides an ample opportunity for evaluating the effectiveness of this policy given that it collects information on a venue-specific information on SHS exposure. The main objective of this study is to address this gap by examining the effect of smoke-free patio policy of restaurants and bars on adult SHS exposure using a quasi-experimental framework. Methods
75
Table 2 Average key sample characteristics (%), 2005–2012. Alberta Gender Male Female Age Education level Less than secondary Secondary College University Marital status Married Separated Single Smoking status Smoker Non-smoker Observations
Nova Scotia
Saskatchewan
50.9 49.1 45.2
48.3 51.7 48.7
49.1 50.9 47.8
11.1 37.0 22.5 29.4
18.3 34.2 20.4 27.1
18.1 41.7 17.9 22.3
72.2 10.0 17.8
67.9 13. 5 18.6
70.7 12.3 17.0
19.6 80.4 12,003
20.0 80.0 11,308
21.1 78.9 11,455
Age is in years.
random sampling framework. In the first phase, households are sampled from telephone numbers using a random digit dialing methodology. Individuals are selected in the second stage based on household composition, with an equal representation of respondents in each of ten Canadian provinces. CTUMS commenced in 1999 and was conducted annually. Data from the 2005–2012 annual CTUMS cycles were used given that our outcome variable was not available in previous iterations. Those in Newfoundland and Labrador were excluded in this study since smoking ban on outdoor patios of bars and restaurants was implemented in this province in 2005 and data is not available to capture baseline SHS exposure. The province of British Columbia was also excluded since most of the large cities/municipalities have implemented a similar policy. Analysis was restricted to those aged 21 and above. Ethics review was not required given that this research relied exclusively on anonymous secondary survey data.
Data Variables
Table 1 Smoke-free policy — outdoor patios of bars and restaurants in Canadian provinces. Source: Non-Smokers' Rights Association (Smoking and Health Action Foundation — March 2014) and author's compilation. Province
Policy date & coverage
Strong or full policy Newfoundland July 1, 2005 — enforced with or without roof Nova Scotia December 1, 2006 — enforced with or without roof Alberta January 1, 2008 — enforced with or without roof Ontario January 1, 2015 — enforced with or without roof Weak or partial policy Manitoba October 1, 2004 — prohibited if more than 25% of the floor area is covered by a roof and more than 50% of its perimeter is more than 50% enclosed New Brunswick October 1, 2004 — prohibited if more than 70% enclosed by walls and/or a roof Quebec May 31, 2006 — prohibited on bar and restaurant patios if they have more than 2 sides and a roof British March 31, 2008 — prohibited if it has a roof or other covering and Columbia more than 50% of the nominal wall space is enclosed such that air flow is impeded Prince Edward September 15, 2009 — prohibited on outdoor patios except Island between the hours of 10 pm and 3 am No policy Saskatchewan
No province-wide ban
Exposure to secondhand smoke The outcome variable was derived from self-reported response to questions on venue-specific exposure to second-hand smoke in the past month. In particular, a dichotomous variable indicating SHS exposure in the past month was created from the question: “The next questions are about exposure to second-hand smoke in places other than your own home. Second-hand smoke is what smokers exhale and the smoke from a burning cigarette. In the past month, excluding your own smoking were you exposed to second-hand smoke on an outdoor patio of a restaurant or bar?” Previous study has shown strong association between selfreported exposure to SHS and biomarkers of nicotine (Okoli et al., 2007). Independent variables Smoking ban on outdoor patios of bars and restaurants was derived by using the dates of implementation of province-wide school smoking ban, respondents' province of residence and date of interview (see Table 1). A number of 33
SHS exposure (%)
This study used a repeated cross-sectional population survey from the Canadian Tobacco Use Monitoring Survey to examine the effect of smoking ban on outdoor patios of bars and restaurants on SHS exposure. The survey is nationally representative of Canadians aged 15+ years. CTUMS primarily collects information about the attitudes and behaviors of adults with respect to tobacco use, as well as corresponding socio-demographic variables. The survey excludes those living in institutions, and Canadians living in the northern Territories (Yukon, Nunavut and Northwest Territories). CTUMS is based on a two-phase stratified
30 27 24 21 18 15 2005
2006
Alberta
2007
2008
Saskatchewan
2009
2010
2011
2012
Linear (Saskatchewan)
Fig. 1. Trend of SHS exposure, 2005–2012 — Alberta and Saskatchewan.
76
S. Azagba / Preventive Medicine 76 (2015) 74–78
SHS exposure (%)
33
Table 4 Estimated effect of smoke-free patio policy on SHS exposure — strong vs. weak policy.
30 27
Outdoor policy Male Age Smoker Observations
24 21 18 15 2005
2006
Nova Scotia
2007
2008
2009
Saskatchewan
2010
2011
2012
Linear (Saskatchewan)
Fig. 2. Trend of SHS exposure, 2005–2012 — Nova Scotia and Saskatchewan.
Model 1
Model 2
−0.049 (−0.065 to −0.033) 0.061 (0.045 to 0.076) −0.003 (−0.004 to −0.002) 0.125 (0.110 to 0.141) 78,288
−0.042 (−0.064 to −0.019) 0.061 (0.045 to 0.076) −0.003 (−0.004 to −0.002) 0.125 (0.109 to 0.141) 78,288
Regression results are obtained from linear probability model with 95% confidence intervals in parenthesis. Treated province: Alberta and Nova Scotia. Control province: Prince Edward Island, New Brunswick, Quebec, Ontario and Manitoba. Ontario was included among provinces with weak policy given that the data analyzed only covered up to 2012. Before January 2015, Ontario only prohibited smoking if patio had partial or complete roof. Model 1 controlled for age, gender, education, smoking status, marital status and unemployment status. Model 2 controlled for age, gender, education, smoking status, marital status, unemployment status, provincial unemployment rates, seasonality, and linear and quadratic trends.
socio-demographic variables were included in the analysis: sex age, marital status, education, smoking status, and unemployment status. replication approach was used in the analysis to enable variance estimates to account for the complex survey design (Judkins, 1990).
Statistical strategy The implementation dates of smoking ban on outdoor patios of bars and restaurants in two Canadian provinces (Alberta and Nova Scotia) are used to create a natural experiment. A difference-in-difference estimator (DD) was used to examine the causal effect of a smoking ban on outdoor patios of bars and restaurants on SHS exposure (Angrist and Pischke, 2009; IARC, 2009). The DD method estimates changes in exposure to SHS for individuals in provinces with the policy of interest, treatment group (Alberta and Nova Scotia) over time relative to individuals in provinces without comparable policy, control group (Saskatchewan). Specifically, a DD model of the following form is estimated: Y ipt ¼ α þ β 1 ðTGÞit þ β 2 ðP Þit þ β 3 ðTGit P it Þ þ φX ipt þ τ t þ Provp þ εipt where Yipt represents outcome of interest (SHS exposure) for individual i in province p at survey time t. TG is a dummy variable that equaled one if the individual is in a treated province and zero otherwise, with the corresponding coefficient, β1, indicating the difference between treated and control groups not due to the policy. The variable P is a dummy variable with a value of 1 for post-policy periods and zero for baseline periods. β2 captures the average change between post-policy and pre-policy periods. A linear probability model was estimated for the ease of interpretation. The interaction of the treated and post-policy variables measures the causal effect of the policy of interest on SHS exposure. The main coefficient of interest, β3 represents the average change in the probability of SHS exposure due to smoking ban on outdoor patios of bars and restaurants. The analysis also controlled for individual socio-demographic characteristics, vector Xipt, including gender, age, education, marital status, smoking status and unemployment status. τt captures time effect (linear and quadratic trends) as well as seasonal effects (month of interview). Prov represents time-variant provincial variable (unemployment rate). In a sensitivity analysis, provinces (Prince Edward Island, New Brunswick, Quebec, Ontario and Manitoba) with weak or partial smoking restriction on outdoor patios of bars and restaurants were used as additional control group. Fay-modified balanced repeated
Results The average key characteristics of the main analysis are reported in Table 2. The trend of exposure to SHS for each year is represented in Figs. 1–2. There has been a decline in SHS exposure following the implementation of smoke-free patio legislation in Nova Scotia and Alberta. In the control province with no patio policy (Saskatchewan), a graphic display showed an upward trend in SHS exposure. Results from the difference––difference estimator showing the average treatment effect of smoke-free patio policy of restaurants and bars on adult SHS exposure are reported in Tables 3–6. Analysis adjusting for age, gender, education, marital status, smoking status and unemployment status is reported in column 1 (model 1) while column 2 (model 2) presents results adjusting for additional covariates (provincial unemployment rates, seasonality, linear and quadratic trends). Regression results were consistent across all model specifications indicating a statistically significant effect of smoking ban on outdoor patios of bars and restaurants on SHS exposure. Specifically, in Table 3, the probability of reporting SHS exposure on outdoor patios of bars and restaurants reduced in Alberta by up to 7.7 percentage points when compared to those living in Saskatchewan (no provincial-level smoking ban on outdoor patios of bars and restaurants). Relative to Alberta average pre-policy SHS exposure of 30.7% in 2007, the 7.7 percentage-point decrease can be expressed as a 25.1% decrease in the probability of SHS exposure. Similarly, the implementation of smoking ban on outdoor patios of bars and restaurants in Nova Scotia reduced the probability of SHS exposure by up to 5.5 percentage points when compared to those living in Saskatchewan. Relative to Nova Scotia average pre-policy SHS exposure of 26.1% in 2006 (January to November), result showed a
Table 3 Estimated effect of smoke-free patio policy on SHS exposure — strong vs. no policy. Alberta
Outdoor policy Male Age Smoker Pre-policy estimated mean of SHS exposure Observations
Nova Scotia
Model 1
Model 2
Model 1
Model 2
−0.077 (−0.102 to −0.052) 0.061 (0.045 to 0.076) −0.003 (−0.004 to −0.002) 0.090 (0.070 to 0.110) 30.7% 23,458
−0.073 (−0.104 to −0.041) 0.061 (0.045 to 0.076) −0.003 (−0.004 to −0.002) 0.090 (0.070 to 0.110) 30.7% 23,458
−0.055 (−0.085 to −0.026) 0.054 (0.041 to 0.068) −0.004 (−0.004 to −0.003) 0.135 (0.117 to 0.152) 26.1% 22,763
−0.045 (−0.077 to −0.012) 0.055 (0.041 to 0.068) 0.004 (−0.004 to −0.003) 0.135 (0.118 to 0.153) 26.1% 22,763
Regression results are obtained from linear probability model with 95% confidence intervals in parenthesis. Control province: Saskatchewan. Model 1 controlled for age, gender, education, smoking status, marital status and unemployment status. Model 2 controlled for age, gender, education, smoking status, marital status, unemployment status, provincial unemployment rates, seasonality, and linear and quadratic trends.
S. Azagba / Preventive Medicine 76 (2015) 74–78
77
Table 5 Estimated heterogeneous effect of smoke-free patio policy on SHS exposure by smoking status: Alberta. Smoker
Outdoor policy Male Age Pre-policy estimated mean of SHS exposure Observations
Non-smoker
Model 1
Model 2
Model 1
Model 2
−0.085 (−0.145 to −0.024) 0.086 (0.045 to 0.076) −0.003 (−0.004 to −0.001) 37.5% 5006
−0.088 (−0.160 to −0.017) 0.087 (0.051 to 0.122) −0.003 (−0.004 to −0.001) 37.5% 5006
−0.077 (−0.105 to −0.050) 0.055 (0.037 to 0.073) −0.003 (−0.003 to −0.002) 28.7% 18,452
−0.071 (−0.104 to −0.038) 0.055 (0.038 to 0.073) −0.003 (−0.003 to −0.002) 28.7 18,452
Regression results are obtained from linear probability model with 95% confidence intervals in parenthesis. Control province: Saskatchewan. Model 1 controlled for age, gender, education, smoking status, marital status and unemployment status. Model 2 controlled for age, gender, education, smoking status, marital status, unemployment status, provincial unemployment rates, seasonality, and linear and quadratic trends.
21.1% decrease in the probability of SHS exposure. As expected, regression results also showed that smokers were more likely to be exposed to SHS. In a sensitivity analysis, provinces with weak or partial smoking restriction on outdoor patios of bars and restaurants were used as additional control group. Similar results were obtained indicating that smoking ban on outdoor patios of bars and restaurants had a statistically significant effect on SHS exposure (see Table 4). Analysis examining whether the effect of the policy varies by smoking status is reported in Tables 5–6. For smokers in Alberta compared to smokers in Saskatchewan, the policy significantly reduced SHS exposure by about 8.8 percentage points yielding a relative decrease of 23.5% based on prepolicy exposure (37.5%). Likewise, relative to non-smokers pre-policy SHS exposure (28.7%), the policy had a larger effect (26.8%) on nonsmokers. Analysis comparing Nova Scotia with Saskatchewan relative to pre-policy SHS exposure showed that the policy had a larger impact on smokers (26.2%) than non-smokers (19.5%). Discussion This study examined the effect of province-wide smoke-free patio policy of bars and restaurants on self-reported SHS exposure in Canada. The results suggest that the policy had a statically significant reduction in SHS exposure. Analysis examining the potential heterogeneous effects of this policy by smoking status also found similar effects showing benefits for both smokers and non-smokers. To my knowledge, this is the first study to provide empirical evidence on the effectiveness of a smoke-free patio policy of bars and restaurants on SHS exposure. The current findings represent a unique and timely contribution to the literature, especially as public support for this policy grows and policy makers in numerous jurisdictions are considering adopting this policy (Thomson et al., 2009). The results of this study are consistent with the extant literature showing that smoke-free legislation is beneficial to health, in part due to a reduction in SHS exposure (e.g., e.g., Hahn et al., 2008; Fowkes et al., 2008; Nagelhout et al., 2011; Millett et al., 2013; Naiman et al., 2010).
The finding of a recent longitudinal study further provides justification for ensuring that outdoor patios of bars and restaurants are smokefree (Chaiton et al., 2014). Examining the effect of exposure to smoking on patios on smoking behavior, Chaiton et al. (2014) found that smokers exposed to smoking on patios was strongly associated with unsuccessful quit and were equally less likely to have made a quit attempt. Another longitudinal study assessing the effects of local restaurant smoking regulations on anti-smoking attitudes in the United States reported positive association between having a strong local regulation and making a quit attempt at follow-up among adult-smokers who already made a quit attempt at baseline (Albers et al., 2007). Beyond the primary benefits of reduced SHS exposure and promoting smoking cessation behaviors, evidence also suggests that smoke-free legislation may positively influence societal norms about tobacco use (Albers et al., 2004, 2007; Brown et al., 2009; Fong et al., 2006). In addition, previous studies have shown that smoke-free policy may encourage positive behavioural change in other settings, referred to as norm spreading (Cheng, Glantz and Lightwood, 2011; Cheng et al., 2015). Future research may want to examine the possibility of ‘norm spreading’ due to smoke-free patio policy. This study has some limitations that warrant noting. First, this study draws on respondent self-report and thus is susceptible to varying forms of bias (recall, social desirability, under-reporting). Though selfreported SHS has been validated (Okoli et al., 2007), there is still a possibility of bias in the results shown here. Second, the results shown in this study may be conservative given that only province-wide smokefree policy on outdoor patios of bars and restaurants was examined, it was not possible to match respondents to local towns/municipalities were similar policy was implemented using publicly available version of CTUMS. For example, a city (Saskatoon) in the control province (Saskatchewan) banned smoking in outdoor patios in July, 2004. Finally, while this study adjusted for educational attainment, other standard measures of SES (e.g., income) are not available in CTUMS. Despite these limitations, the use of this large survey provides the opportunity to assess the population-level effect of smoke-free patio policy of restaurants and bars on adult SHS exposure.
Table 6 Estimated heterogeneous effect of smoke-free patio policy on SHS exposure by smoking status: Nova Scotia. Smoker
Outdoor policy Male Age Pre-policy estimated mean of SHS exposure Observations
Non-smoker
Model 1
Model 2
Model 1
Model 2
−0.010 (−0.174 to −0.025) 0.064 (0.029 to 0.099) −0.004 (−0.006 to −0.003) 38.1% 4872
−0.089 (−0.168 to −0.011) 0.063 (0.028 to 0.097) −0.004 (−0.006 to −0.003) 38.1% 4872
−0.044 (−0.076 to −0.012) 0.055 (0.037 to 0.073) −0.003 (−0.004 to −0.003) 22.6% 17,891
−0.034 (−0.068 to −0.001) 0.055 (0.038 to 0.073) −0.003 (−0.004 to −0.003) 22.6% 17,891
Regression results are obtained from linear probability model with 95% confidence intervals in parenthesis. Control province: Saskatchewan. Model 1 controlled for age, gender, education, smoking status, marital status and unemployment status. Model 2 controlled for age, gender, education, smoking status, marital status, unemployment status, provincial unemployment rates, seasonality, and linear and quadratic trends.
78
S. Azagba / Preventive Medicine 76 (2015) 74–78
Conclusion In summary, as the momentum grows for developing endgame strategy aimed at reducing tobacco use or tobacco-related diseases to the barest minimum worldwide, implementing a comprehensive tobacco-control strategy that includes smoke-free legislation would provide both short- and long-term health benefits (e.g., reduction in tobacco-related healthcare expenditures) (Yao et al., 2014). Conflict of interest statement None.
Acknowledgment This research is supported by the Canadian Cancer Society grant#2011-701019. References Akhtar, P.C., Haw, S.J., Currie, D.B., Zachary, R., Currie, C.E., 2009. Smoking restrictions in the home and secondhand smoke exposure among primary schoolchildren before and after introduction of the Scottish smoke-free legislation. Tob. Control. 18 (5), 409–415. Albers, A.B., Siegel, M., Cheng, D.M., Biener, L., Rigotti, N.A., 2004. Relation between local restaurant smoking regulations and attitudes towards the prevalence and social acceptability of smoking: a study of youths and adults who eat out predominantly at restaurants in their town. Tob. Control. 13 (4), 347–355. Albers, A.B., Siegel, M., Cheng, D.M., Biener, L., Rigotti, N.A., 2007. Effect of smoking regulations in local restaurants on smokers' anti-smoking attitudes and quitting behaviours. Tob. Control. 16, 101–106. American Nonsmokers' Rights Foundation, 2014. http://www.no-smoke.org/pdf/ SmokefreeOutdoorDining.pdf. Angrist, J.D., Pischke, J.S., 2009. Mostly Harmless Econometrics: An Empiricist's Companion. Princeton University Press. Azagba, S., Asbridge, M., 2013. The association between smoking bans and nicotine dependence: a longitudinal analysis of current smokers in Canada. Addict. Behav. 38 (12), 2817–2820. Bondy, S.J., Zhang, B., Kreiger, N., et al., 2009. Impact of an indoor smoking ban on bar workers' exposure to secondhand smoke. J. Occup. Environ. Med. 51 (5), 612–619. Brown, A., Moodie, C., Hastings, G., 2009. A longitudinal study of policy effect (smoke-free legislation) on smoking norms: ITC Scotland/United Kingdom. Nicotine Tob. Res. ntp087. Callinan, J.E., Clarke, A., Doherty, K., Kelleher, C., 2010. Legislative smoking bans for reducing secondhand smoke exposure, smoking prevalence and tobacco consumption. Cochrane Database Syst. Rev. 4. Chaiton, M., Diemert, L., Zhang, B., et al., 2014. Exposure to smoking on patios and quitting: a population representative longitudinal cohort study. Tob. Control. http://dx. doi.org/10.1136/tobaccocontrol- 2014-051761 (tobaccocontrol-2014). Cheng, K.W., Glantz, S.A., Lightwood, J.M., 2011. Association between smokefree laws and voluntary smokefree-home rules. Am. J. Prev. Med. 41 (6), 566–572. Cheng, K.-W., Okechukwu, C.A., McMillen, R., Glantz, S.A., 2015. Tob. Control. 24, 168–174. Farrelly, M.C., Nonnemaker, J.M., Chou, R., Hyland, A., Peterson, K.K., Bauer, U.E., 2005. Changes in hospitality workers' exposure to secondhand smoke following the implementation of New York's smoke-free law. Tob. Control. 14 (4), 236–241. Fernando, D., Fowles, J., Woodward, A., et al., 2007. Legislation reduces exposure to second-hand tobacco smoke in New Zealand bars by about 90%. Tob. Control. 16 (4), 235–238. Fong, G.T., Hyland, A., Borland, R., et al., 2006. Reductions in tobacco smoke pollution and increases in support for smoke-free public places following the implementation of comprehensive smoke-free workplace legislation in the Republic of Ireland: findings from the ITC Ireland/UK Survey. Tob. Control. 15, Siii51–Siii58.
Fowkes, F.J., Stewart, M.C., Fowkes, F.G.R., Amos, A., Price, J.F., 2008. Scottish smoke-free legislation and trends in smoking cessation. Addiction 103 (11), 1888–1895. Hahn, E.J., Rayens, M.K., Butler, K.M., Zhang, M., Durbin, E., Steinke, D., 2008. Smoke-free laws and adult smoking prevalence. Prev. Med. 47 (2), 206–209. International Agency for Research on Cancer, 2004. Tobacco Smoke and Involuntary Smoking. WHO, IARC, Lyon (France) (http://monographs.iarc.fr/ENG/Monographs/ vol83/mono83.pdf). International Agency for Research on Cancer, 2009. IARC handbooks of cancer prevention: tobacco control. Evaluating the Effectiveness of Smoke-free Policies vol. 13. IARC, Lyon. Jensen, J.A., Schillo, B.A., Moilanen, M.M., et al., 2010. Tobacco smoke exposure in nonsmoking hospitality workers before and after a state smoking ban. Cancer Epidemiol. Biomark. Prev. 19 (4), 1016–1021. Judkins, D.R., 1990. Fay's method for variance estimation. J. Off. Stat. 6, 223–239. Klepeis, N.E., Ott, W.R., Switzer, P., 2007. Real–time measurement of outdoor tobacco smoke particles. J. Air Waste Manage. Assoc. 57 (5), 522–534. Millett, C., Lee, J.T., Laverty, A.A., Glantz, S.A., Majeed, A., 2013. Hospital admissions for childhood asthma after smoke-free legislation in England. Pediatrics 131 (2), e495–e501. Mills, A.L., Messer, K., Gilpin, E.A., Pierce, J.P., 2009. The effect of smoke-free homes on adult smoking behavior: a review. Nicotine Tob. Res. ntp122. Mulcahy, M., Evans, D.S., Hammond, S.K., Repace, J.L., Byrne, M., 2005. Secondhand smoke exposure and risk following the Irish smoking ban: an assessment of salivary cotinine concentrations in hotel workers and air nicotine levels in bars. Tob. Control. 14 (6), 384–388. Nagelhout, G.E., Willemsen, M.C., de Vries, H., 2011. The population impact of smoke-free workplace and hospitality industry legislation on smoking behaviour. Findings from a national population survey. Addiction 106 (4), 816–823. Naiman, A., Glazier, R.H., Moineddin, R., 2010. Association of anti-smoking legislation with rates of hospital admission for cardiovascular and respiratory conditions. Can. Med. Assoc. J. 182 (8), 761–767. Öberg, M., Jaakkola, M.S., Woodward, A., Peruga, A., Prüss-Ustün, A., 2011. Worldwide burden of disease from exposure to second-hand smoke: a retrospective analysis of data from 192 countries. Lancet 377 (9760), 139–146. Okoli, C.T., Kelly, T., Hahn, E.J., 2007. Secondhand smoke and nicotine exposure: a brief review. Addict. Behav. 32 (10), 1977–1988. Repace, J., Zhang, B., Bondy, S.J., Benowitz, N., Ferrence, R., 2013. Air quality, mortality, and economic benefits of a smoke-free workplace law for non-smoking Ontario bar workers. Indoor Air 23 (2), 93–104. St Helen, G., Thomas Bernert, J., Hall, D.B., et al., 2012. Exposure to secondhand smoke outside of a bar and a restaurant and tobacco exposure biomarkers in nonsmokers. Environ. Health Perspect. 120 (7), 1010. Thomson, G., Wilson, N., Edwards, R., 2009. At the frontier of tobacco control: a brief review of public attitudes toward smoke-free outdoor places. Nicotine Tob. Res. ntp046. U.S. Department of Health and Human Services, 2006. The Health Consequences of Involuntary Exposure to Tobacco Smoke: A Report of the Surgeon General. 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. U.S. Environmental Protection Agency, 1992. Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders. U.S. Environmental Protection Agency, Office of Health and Environmental Assessment, Office of Research and Development, Washington, DC (EPA/600/6-90/006 F). Valente, P., Forastiere, F., Bacosi, A., et al., 2007. Exposure to fine and ultrafine particles from secondhand smoke in public places before and after the smoking ban, Italy 2005. Tob. Control. 16 (5), 312–317. WHO Report on the Global Tobacco Epidemic, 2009: Implementing Smoke-free Environments. World Health Organization, Geneva (http://www.who.int/tobacco/mpower/ 2009/en/index.html, accessed 28 October 2014). Yao, T., Sung, H.Y., Mao, Z., Hu, T.W., Max, W., 2014. The healthcare costs of secondhand smoke exposure in rural China. Tob. Control. http://dx.doi.org/10.1136/ tobaccocontrol- 2014-051761 (tobaccocontrol-2014).
Contents lists available at ScienceDirect
Preventive Medicine journal homepage: www.elsevier.com/locate/ypmed
Effect of smoke-free patio policy of restaurants and bars on exposure to second-hand smoke Sunday Azagba ⁎ Propel Centre for Population Health Impact, University of Waterloo, Canada School of Public Health and Health Systems, University of Waterloo, Canada
a r t i c l e
i n f o
Available online 22 April 2015 Keywords: Smoke-free policy Outdoor patio Second-hand smoke Environmental tobacco smoke Passive smoking
a b s t r a c t Objective. While there is increasing support for restricting smoking in restaurant and bar patios, there is limited evidence on the effectiveness of this policy. This study examined the effect of smoke-free patio policy of restaurants and bars on adult second-hand smoke (SHS) exposure. Methods. Data were drawn from the 2005–2012 Canadian Tobacco Use Monitoring Survey (n = 89,743), a repeated cross-sectional survey of youth and adult. Regression analysis, a quasi-experimental design was used to examine the effect of provincial smoke-free patio policy on self-reported exposure to SHS. Results. Analyses suggest that exposure to SHS on patios of bars and restaurants declined following the adoption of provincial smoke-free patio policy. Relative to pre-policy SHS exposure, regression results showed a reduction in the probability of SHS exposure of up to 25% in Alberta. Similarly, in Nova Scotia, the probability of SHS exposure declined by up to 21%. Analyses stratified by smoking status found similar significant effect on both smokers and non-smokers. Conclusions. Findings suggest that provincial patio smoking ban on bars and restaurants had the intended effect of protecting non-smokers from SHS exposure. This study is consistent with a large body of evidence showing that a strong smoke-free legislation is an effective public health measure. © 2015 The Author. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Introduction Globally, over 600,000 deaths each year are linked to exposure to second-hand smoke (SHS), also known as environmental tobacco smoke or tobacco smoke pollution (WHO, 2009; Öberg et al., 2011). SHS has been documented as a risk factor for several diseases such as lung cancer, cardiovascular disease, respiratory problems, asthma and sudden infant death syndrome (U.S. Department of Health and Human Services, 2006; U.S. Environmental Protection Agency, 1992). The International Agency for Research on Cancer as well as the U.S. Environmental Protection Agency has designated second-hand smoke as a human carcinogen for which there is no risk-free level (International Agency for Research on Cancer, 2004; U.S. Environmental Protection Agency, 1992). The U.S. Surgeon General report has also noted that a small amount of SHS exposure can adversely affect the heart, blood and blood vessels (U.S. Department of Health and Human Services, 2006). Public smoking restriction has become ubiquitous in many jurisdictions, in part due to the heightened awareness of the harmful health consequences of SHS exposure (U.S. Department of Health and Human ⁎ Propel Centre for Population Health Impact, Faculty of Applied Health Sciences, University of Waterloo, 200 University Avenue West, Waterloo, Ontario N2L 3G1, Canada. E-mail address: [email protected].
Services, 2006; WHO, 2009). The balance of evidence from the preponderance of research suggests that smoke-free legislation is an important public health intervention measure for protecting both non-smokers and smokers from the adverse health effects of SHS exposure (Akhtar et al., 2009; Bondy et al., 2009; Fong et al., 2006; IARC, 2009; Mulcahy et al., 2005; Repace et al., 2013). A large body of evidence has shown significant reductions in cotinine (a metabolite of nicotine) levels among non-smokers following smoke-free policy implementation in many countries including the U.S. (Jensen et al., 2010; Farrelly et al., 2005), Canada (Bondy et al., 2009), Ireland (Mulcahy et al., 2005), Scotland (Akhtar et al., 2009), Italy (Valente et al., 2007), and New Zealand (Fernando et al., 2007). In addition, some population-based survey studies suggest that smoke-free policy is strongly associated with smoking cessation behaviors (Callinan et al., 2010; Hahn et al., 2008; Fowkes et al., 2008; Azagba and Asbridge, 2013; Nagelhout et al., 2011). For example, a review study by Mills and colleagues found consistent evidence, at a populationlevel, that smoking bans promote smoking cessation and decreased cigarette consumption in adult smokers (Mills et al., 2009). While public support is increasing for outdoor smoke-free legislation (Thomson et al., 2009), not much is known about SHS exposure in outdoor dining areas of bars and restaurants (St. Helen et al., 2012; Klepeis et al., 2007). Prior research has demonstrated similar levels of tobacco smoke concentrations in outdoor and indoor settings, especially
http://dx.doi.org/10.1016/j.ypmed.2015.04.012 0091-7435/© 2015 The Author. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
S. Azagba / Preventive Medicine 76 (2015) 74–78
in a close proximity to smokers (Klepeis et al., 2007). In a crossover study, St. Helen et al. (2012) found a significant increase in salivary cotinine and NNAL (a metabolite of tobacco-related nitrosamines) in nonsmokers following SHS exposure in outdoor patios of a restaurant and bar. These findings underscore the potential public health benefit of banning smoking in outdoor patios. In Canada, four provinces (Newfoundland and Labrador, Nova Scotia, Alberta and Ontario) and Yukon Territory have implemented full province-wide smoking bans on outdoor patios of bars and restaurants. Ontario, the most populated province in Canada implemented province-wide ban in January, 2015. Other provinces (Prince Edward Island, New Brunswick, Quebec, Manitoba, Saskatchewan, Ontario and British Columbia) have implemented weak or no bans (e.g., Prince Edward Island outdoor patio ban is not enforced between 10 pm and 3 am, see Table 1 for details). In the United States, a number of states and local municipalities have also implemented smoke-free outdoor dining policy including bar patios (American Nonsmokers' Rights Foundation, 2014). As the number of jurisdictions implementing smoking bans on outdoor patios of bars and restaurants is increasing, to my knowledge, no study has evaluated the effect of this policy on SHS exposure. This may not be surprising given that only few jurisdictions have so far implemented this policy, and also evaluation of this policy will require collecting information on venue-specific SHS exposure. The Canadian Tobacco Use Monitoring Survey (CTUMS) provides an ample opportunity for evaluating the effectiveness of this policy given that it collects information on a venue-specific information on SHS exposure. The main objective of this study is to address this gap by examining the effect of smoke-free patio policy of restaurants and bars on adult SHS exposure using a quasi-experimental framework. Methods
75
Table 2 Average key sample characteristics (%), 2005–2012. Alberta Gender Male Female Age Education level Less than secondary Secondary College University Marital status Married Separated Single Smoking status Smoker Non-smoker Observations
Nova Scotia
Saskatchewan
50.9 49.1 45.2
48.3 51.7 48.7
49.1 50.9 47.8
11.1 37.0 22.5 29.4
18.3 34.2 20.4 27.1
18.1 41.7 17.9 22.3
72.2 10.0 17.8
67.9 13. 5 18.6
70.7 12.3 17.0
19.6 80.4 12,003
20.0 80.0 11,308
21.1 78.9 11,455
Age is in years.
random sampling framework. In the first phase, households are sampled from telephone numbers using a random digit dialing methodology. Individuals are selected in the second stage based on household composition, with an equal representation of respondents in each of ten Canadian provinces. CTUMS commenced in 1999 and was conducted annually. Data from the 2005–2012 annual CTUMS cycles were used given that our outcome variable was not available in previous iterations. Those in Newfoundland and Labrador were excluded in this study since smoking ban on outdoor patios of bars and restaurants was implemented in this province in 2005 and data is not available to capture baseline SHS exposure. The province of British Columbia was also excluded since most of the large cities/municipalities have implemented a similar policy. Analysis was restricted to those aged 21 and above. Ethics review was not required given that this research relied exclusively on anonymous secondary survey data.
Data Variables
Table 1 Smoke-free policy — outdoor patios of bars and restaurants in Canadian provinces. Source: Non-Smokers' Rights Association (Smoking and Health Action Foundation — March 2014) and author's compilation. Province
Policy date & coverage
Strong or full policy Newfoundland July 1, 2005 — enforced with or without roof Nova Scotia December 1, 2006 — enforced with or without roof Alberta January 1, 2008 — enforced with or without roof Ontario January 1, 2015 — enforced with or without roof Weak or partial policy Manitoba October 1, 2004 — prohibited if more than 25% of the floor area is covered by a roof and more than 50% of its perimeter is more than 50% enclosed New Brunswick October 1, 2004 — prohibited if more than 70% enclosed by walls and/or a roof Quebec May 31, 2006 — prohibited on bar and restaurant patios if they have more than 2 sides and a roof British March 31, 2008 — prohibited if it has a roof or other covering and Columbia more than 50% of the nominal wall space is enclosed such that air flow is impeded Prince Edward September 15, 2009 — prohibited on outdoor patios except Island between the hours of 10 pm and 3 am No policy Saskatchewan
No province-wide ban
Exposure to secondhand smoke The outcome variable was derived from self-reported response to questions on venue-specific exposure to second-hand smoke in the past month. In particular, a dichotomous variable indicating SHS exposure in the past month was created from the question: “The next questions are about exposure to second-hand smoke in places other than your own home. Second-hand smoke is what smokers exhale and the smoke from a burning cigarette. In the past month, excluding your own smoking were you exposed to second-hand smoke on an outdoor patio of a restaurant or bar?” Previous study has shown strong association between selfreported exposure to SHS and biomarkers of nicotine (Okoli et al., 2007). Independent variables Smoking ban on outdoor patios of bars and restaurants was derived by using the dates of implementation of province-wide school smoking ban, respondents' province of residence and date of interview (see Table 1). A number of 33
SHS exposure (%)
This study used a repeated cross-sectional population survey from the Canadian Tobacco Use Monitoring Survey to examine the effect of smoking ban on outdoor patios of bars and restaurants on SHS exposure. The survey is nationally representative of Canadians aged 15+ years. CTUMS primarily collects information about the attitudes and behaviors of adults with respect to tobacco use, as well as corresponding socio-demographic variables. The survey excludes those living in institutions, and Canadians living in the northern Territories (Yukon, Nunavut and Northwest Territories). CTUMS is based on a two-phase stratified
30 27 24 21 18 15 2005
2006
Alberta
2007
2008
Saskatchewan
2009
2010
2011
2012
Linear (Saskatchewan)
Fig. 1. Trend of SHS exposure, 2005–2012 — Alberta and Saskatchewan.
76
S. Azagba / Preventive Medicine 76 (2015) 74–78
SHS exposure (%)
33
Table 4 Estimated effect of smoke-free patio policy on SHS exposure — strong vs. weak policy.
30 27
Outdoor policy Male Age Smoker Observations
24 21 18 15 2005
2006
Nova Scotia
2007
2008
2009
Saskatchewan
2010
2011
2012
Linear (Saskatchewan)
Fig. 2. Trend of SHS exposure, 2005–2012 — Nova Scotia and Saskatchewan.
Model 1
Model 2
−0.049 (−0.065 to −0.033) 0.061 (0.045 to 0.076) −0.003 (−0.004 to −0.002) 0.125 (0.110 to 0.141) 78,288
−0.042 (−0.064 to −0.019) 0.061 (0.045 to 0.076) −0.003 (−0.004 to −0.002) 0.125 (0.109 to 0.141) 78,288
Regression results are obtained from linear probability model with 95% confidence intervals in parenthesis. Treated province: Alberta and Nova Scotia. Control province: Prince Edward Island, New Brunswick, Quebec, Ontario and Manitoba. Ontario was included among provinces with weak policy given that the data analyzed only covered up to 2012. Before January 2015, Ontario only prohibited smoking if patio had partial or complete roof. Model 1 controlled for age, gender, education, smoking status, marital status and unemployment status. Model 2 controlled for age, gender, education, smoking status, marital status, unemployment status, provincial unemployment rates, seasonality, and linear and quadratic trends.
socio-demographic variables were included in the analysis: sex age, marital status, education, smoking status, and unemployment status. replication approach was used in the analysis to enable variance estimates to account for the complex survey design (Judkins, 1990).
Statistical strategy The implementation dates of smoking ban on outdoor patios of bars and restaurants in two Canadian provinces (Alberta and Nova Scotia) are used to create a natural experiment. A difference-in-difference estimator (DD) was used to examine the causal effect of a smoking ban on outdoor patios of bars and restaurants on SHS exposure (Angrist and Pischke, 2009; IARC, 2009). The DD method estimates changes in exposure to SHS for individuals in provinces with the policy of interest, treatment group (Alberta and Nova Scotia) over time relative to individuals in provinces without comparable policy, control group (Saskatchewan). Specifically, a DD model of the following form is estimated: Y ipt ¼ α þ β 1 ðTGÞit þ β 2 ðP Þit þ β 3 ðTGit P it Þ þ φX ipt þ τ t þ Provp þ εipt where Yipt represents outcome of interest (SHS exposure) for individual i in province p at survey time t. TG is a dummy variable that equaled one if the individual is in a treated province and zero otherwise, with the corresponding coefficient, β1, indicating the difference between treated and control groups not due to the policy. The variable P is a dummy variable with a value of 1 for post-policy periods and zero for baseline periods. β2 captures the average change between post-policy and pre-policy periods. A linear probability model was estimated for the ease of interpretation. The interaction of the treated and post-policy variables measures the causal effect of the policy of interest on SHS exposure. The main coefficient of interest, β3 represents the average change in the probability of SHS exposure due to smoking ban on outdoor patios of bars and restaurants. The analysis also controlled for individual socio-demographic characteristics, vector Xipt, including gender, age, education, marital status, smoking status and unemployment status. τt captures time effect (linear and quadratic trends) as well as seasonal effects (month of interview). Prov represents time-variant provincial variable (unemployment rate). In a sensitivity analysis, provinces (Prince Edward Island, New Brunswick, Quebec, Ontario and Manitoba) with weak or partial smoking restriction on outdoor patios of bars and restaurants were used as additional control group. Fay-modified balanced repeated
Results The average key characteristics of the main analysis are reported in Table 2. The trend of exposure to SHS for each year is represented in Figs. 1–2. There has been a decline in SHS exposure following the implementation of smoke-free patio legislation in Nova Scotia and Alberta. In the control province with no patio policy (Saskatchewan), a graphic display showed an upward trend in SHS exposure. Results from the difference––difference estimator showing the average treatment effect of smoke-free patio policy of restaurants and bars on adult SHS exposure are reported in Tables 3–6. Analysis adjusting for age, gender, education, marital status, smoking status and unemployment status is reported in column 1 (model 1) while column 2 (model 2) presents results adjusting for additional covariates (provincial unemployment rates, seasonality, linear and quadratic trends). Regression results were consistent across all model specifications indicating a statistically significant effect of smoking ban on outdoor patios of bars and restaurants on SHS exposure. Specifically, in Table 3, the probability of reporting SHS exposure on outdoor patios of bars and restaurants reduced in Alberta by up to 7.7 percentage points when compared to those living in Saskatchewan (no provincial-level smoking ban on outdoor patios of bars and restaurants). Relative to Alberta average pre-policy SHS exposure of 30.7% in 2007, the 7.7 percentage-point decrease can be expressed as a 25.1% decrease in the probability of SHS exposure. Similarly, the implementation of smoking ban on outdoor patios of bars and restaurants in Nova Scotia reduced the probability of SHS exposure by up to 5.5 percentage points when compared to those living in Saskatchewan. Relative to Nova Scotia average pre-policy SHS exposure of 26.1% in 2006 (January to November), result showed a
Table 3 Estimated effect of smoke-free patio policy on SHS exposure — strong vs. no policy. Alberta
Outdoor policy Male Age Smoker Pre-policy estimated mean of SHS exposure Observations
Nova Scotia
Model 1
Model 2
Model 1
Model 2
−0.077 (−0.102 to −0.052) 0.061 (0.045 to 0.076) −0.003 (−0.004 to −0.002) 0.090 (0.070 to 0.110) 30.7% 23,458
−0.073 (−0.104 to −0.041) 0.061 (0.045 to 0.076) −0.003 (−0.004 to −0.002) 0.090 (0.070 to 0.110) 30.7% 23,458
−0.055 (−0.085 to −0.026) 0.054 (0.041 to 0.068) −0.004 (−0.004 to −0.003) 0.135 (0.117 to 0.152) 26.1% 22,763
−0.045 (−0.077 to −0.012) 0.055 (0.041 to 0.068) 0.004 (−0.004 to −0.003) 0.135 (0.118 to 0.153) 26.1% 22,763
Regression results are obtained from linear probability model with 95% confidence intervals in parenthesis. Control province: Saskatchewan. Model 1 controlled for age, gender, education, smoking status, marital status and unemployment status. Model 2 controlled for age, gender, education, smoking status, marital status, unemployment status, provincial unemployment rates, seasonality, and linear and quadratic trends.
S. Azagba / Preventive Medicine 76 (2015) 74–78
77
Table 5 Estimated heterogeneous effect of smoke-free patio policy on SHS exposure by smoking status: Alberta. Smoker
Outdoor policy Male Age Pre-policy estimated mean of SHS exposure Observations
Non-smoker
Model 1
Model 2
Model 1
Model 2
−0.085 (−0.145 to −0.024) 0.086 (0.045 to 0.076) −0.003 (−0.004 to −0.001) 37.5% 5006
−0.088 (−0.160 to −0.017) 0.087 (0.051 to 0.122) −0.003 (−0.004 to −0.001) 37.5% 5006
−0.077 (−0.105 to −0.050) 0.055 (0.037 to 0.073) −0.003 (−0.003 to −0.002) 28.7% 18,452
−0.071 (−0.104 to −0.038) 0.055 (0.038 to 0.073) −0.003 (−0.003 to −0.002) 28.7 18,452
Regression results are obtained from linear probability model with 95% confidence intervals in parenthesis. Control province: Saskatchewan. Model 1 controlled for age, gender, education, smoking status, marital status and unemployment status. Model 2 controlled for age, gender, education, smoking status, marital status, unemployment status, provincial unemployment rates, seasonality, and linear and quadratic trends.
21.1% decrease in the probability of SHS exposure. As expected, regression results also showed that smokers were more likely to be exposed to SHS. In a sensitivity analysis, provinces with weak or partial smoking restriction on outdoor patios of bars and restaurants were used as additional control group. Similar results were obtained indicating that smoking ban on outdoor patios of bars and restaurants had a statistically significant effect on SHS exposure (see Table 4). Analysis examining whether the effect of the policy varies by smoking status is reported in Tables 5–6. For smokers in Alberta compared to smokers in Saskatchewan, the policy significantly reduced SHS exposure by about 8.8 percentage points yielding a relative decrease of 23.5% based on prepolicy exposure (37.5%). Likewise, relative to non-smokers pre-policy SHS exposure (28.7%), the policy had a larger effect (26.8%) on nonsmokers. Analysis comparing Nova Scotia with Saskatchewan relative to pre-policy SHS exposure showed that the policy had a larger impact on smokers (26.2%) than non-smokers (19.5%). Discussion This study examined the effect of province-wide smoke-free patio policy of bars and restaurants on self-reported SHS exposure in Canada. The results suggest that the policy had a statically significant reduction in SHS exposure. Analysis examining the potential heterogeneous effects of this policy by smoking status also found similar effects showing benefits for both smokers and non-smokers. To my knowledge, this is the first study to provide empirical evidence on the effectiveness of a smoke-free patio policy of bars and restaurants on SHS exposure. The current findings represent a unique and timely contribution to the literature, especially as public support for this policy grows and policy makers in numerous jurisdictions are considering adopting this policy (Thomson et al., 2009). The results of this study are consistent with the extant literature showing that smoke-free legislation is beneficial to health, in part due to a reduction in SHS exposure (e.g., e.g., Hahn et al., 2008; Fowkes et al., 2008; Nagelhout et al., 2011; Millett et al., 2013; Naiman et al., 2010).
The finding of a recent longitudinal study further provides justification for ensuring that outdoor patios of bars and restaurants are smokefree (Chaiton et al., 2014). Examining the effect of exposure to smoking on patios on smoking behavior, Chaiton et al. (2014) found that smokers exposed to smoking on patios was strongly associated with unsuccessful quit and were equally less likely to have made a quit attempt. Another longitudinal study assessing the effects of local restaurant smoking regulations on anti-smoking attitudes in the United States reported positive association between having a strong local regulation and making a quit attempt at follow-up among adult-smokers who already made a quit attempt at baseline (Albers et al., 2007). Beyond the primary benefits of reduced SHS exposure and promoting smoking cessation behaviors, evidence also suggests that smoke-free legislation may positively influence societal norms about tobacco use (Albers et al., 2004, 2007; Brown et al., 2009; Fong et al., 2006). In addition, previous studies have shown that smoke-free policy may encourage positive behavioural change in other settings, referred to as norm spreading (Cheng, Glantz and Lightwood, 2011; Cheng et al., 2015). Future research may want to examine the possibility of ‘norm spreading’ due to smoke-free patio policy. This study has some limitations that warrant noting. First, this study draws on respondent self-report and thus is susceptible to varying forms of bias (recall, social desirability, under-reporting). Though selfreported SHS has been validated (Okoli et al., 2007), there is still a possibility of bias in the results shown here. Second, the results shown in this study may be conservative given that only province-wide smokefree policy on outdoor patios of bars and restaurants was examined, it was not possible to match respondents to local towns/municipalities were similar policy was implemented using publicly available version of CTUMS. For example, a city (Saskatoon) in the control province (Saskatchewan) banned smoking in outdoor patios in July, 2004. Finally, while this study adjusted for educational attainment, other standard measures of SES (e.g., income) are not available in CTUMS. Despite these limitations, the use of this large survey provides the opportunity to assess the population-level effect of smoke-free patio policy of restaurants and bars on adult SHS exposure.
Table 6 Estimated heterogeneous effect of smoke-free patio policy on SHS exposure by smoking status: Nova Scotia. Smoker
Outdoor policy Male Age Pre-policy estimated mean of SHS exposure Observations
Non-smoker
Model 1
Model 2
Model 1
Model 2
−0.010 (−0.174 to −0.025) 0.064 (0.029 to 0.099) −0.004 (−0.006 to −0.003) 38.1% 4872
−0.089 (−0.168 to −0.011) 0.063 (0.028 to 0.097) −0.004 (−0.006 to −0.003) 38.1% 4872
−0.044 (−0.076 to −0.012) 0.055 (0.037 to 0.073) −0.003 (−0.004 to −0.003) 22.6% 17,891
−0.034 (−0.068 to −0.001) 0.055 (0.038 to 0.073) −0.003 (−0.004 to −0.003) 22.6% 17,891
Regression results are obtained from linear probability model with 95% confidence intervals in parenthesis. Control province: Saskatchewan. Model 1 controlled for age, gender, education, smoking status, marital status and unemployment status. Model 2 controlled for age, gender, education, smoking status, marital status, unemployment status, provincial unemployment rates, seasonality, and linear and quadratic trends.
78
S. Azagba / Preventive Medicine 76 (2015) 74–78
Conclusion In summary, as the momentum grows for developing endgame strategy aimed at reducing tobacco use or tobacco-related diseases to the barest minimum worldwide, implementing a comprehensive tobacco-control strategy that includes smoke-free legislation would provide both short- and long-term health benefits (e.g., reduction in tobacco-related healthcare expenditures) (Yao et al., 2014). Conflict of interest statement None.
Acknowledgment This research is supported by the Canadian Cancer Society grant#2011-701019. References Akhtar, P.C., Haw, S.J., Currie, D.B., Zachary, R., Currie, C.E., 2009. Smoking restrictions in the home and secondhand smoke exposure among primary schoolchildren before and after introduction of the Scottish smoke-free legislation. Tob. Control. 18 (5), 409–415. Albers, A.B., Siegel, M., Cheng, D.M., Biener, L., Rigotti, N.A., 2004. Relation between local restaurant smoking regulations and attitudes towards the prevalence and social acceptability of smoking: a study of youths and adults who eat out predominantly at restaurants in their town. Tob. Control. 13 (4), 347–355. Albers, A.B., Siegel, M., Cheng, D.M., Biener, L., Rigotti, N.A., 2007. Effect of smoking regulations in local restaurants on smokers' anti-smoking attitudes and quitting behaviours. Tob. Control. 16, 101–106. American Nonsmokers' Rights Foundation, 2014. http://www.no-smoke.org/pdf/ SmokefreeOutdoorDining.pdf. Angrist, J.D., Pischke, J.S., 2009. Mostly Harmless Econometrics: An Empiricist's Companion. Princeton University Press. Azagba, S., Asbridge, M., 2013. The association between smoking bans and nicotine dependence: a longitudinal analysis of current smokers in Canada. Addict. Behav. 38 (12), 2817–2820. Bondy, S.J., Zhang, B., Kreiger, N., et al., 2009. Impact of an indoor smoking ban on bar workers' exposure to secondhand smoke. J. Occup. Environ. Med. 51 (5), 612–619. Brown, A., Moodie, C., Hastings, G., 2009. A longitudinal study of policy effect (smoke-free legislation) on smoking norms: ITC Scotland/United Kingdom. Nicotine Tob. Res. ntp087. Callinan, J.E., Clarke, A., Doherty, K., Kelleher, C., 2010. Legislative smoking bans for reducing secondhand smoke exposure, smoking prevalence and tobacco consumption. Cochrane Database Syst. Rev. 4. Chaiton, M., Diemert, L., Zhang, B., et al., 2014. Exposure to smoking on patios and quitting: a population representative longitudinal cohort study. Tob. Control. http://dx. doi.org/10.1136/tobaccocontrol- 2014-051761 (tobaccocontrol-2014). Cheng, K.W., Glantz, S.A., Lightwood, J.M., 2011. Association between smokefree laws and voluntary smokefree-home rules. Am. J. Prev. Med. 41 (6), 566–572. Cheng, K.-W., Okechukwu, C.A., McMillen, R., Glantz, S.A., 2015. Tob. Control. 24, 168–174. Farrelly, M.C., Nonnemaker, J.M., Chou, R., Hyland, A., Peterson, K.K., Bauer, U.E., 2005. Changes in hospitality workers' exposure to secondhand smoke following the implementation of New York's smoke-free law. Tob. Control. 14 (4), 236–241. Fernando, D., Fowles, J., Woodward, A., et al., 2007. Legislation reduces exposure to second-hand tobacco smoke in New Zealand bars by about 90%. Tob. Control. 16 (4), 235–238. Fong, G.T., Hyland, A., Borland, R., et al., 2006. Reductions in tobacco smoke pollution and increases in support for smoke-free public places following the implementation of comprehensive smoke-free workplace legislation in the Republic of Ireland: findings from the ITC Ireland/UK Survey. Tob. Control. 15, Siii51–Siii58.
Fowkes, F.J., Stewart, M.C., Fowkes, F.G.R., Amos, A., Price, J.F., 2008. Scottish smoke-free legislation and trends in smoking cessation. Addiction 103 (11), 1888–1895. Hahn, E.J., Rayens, M.K., Butler, K.M., Zhang, M., Durbin, E., Steinke, D., 2008. Smoke-free laws and adult smoking prevalence. Prev. Med. 47 (2), 206–209. International Agency for Research on Cancer, 2004. Tobacco Smoke and Involuntary Smoking. WHO, IARC, Lyon (France) (http://monographs.iarc.fr/ENG/Monographs/ vol83/mono83.pdf). International Agency for Research on Cancer, 2009. IARC handbooks of cancer prevention: tobacco control. Evaluating the Effectiveness of Smoke-free Policies vol. 13. IARC, Lyon. Jensen, J.A., Schillo, B.A., Moilanen, M.M., et al., 2010. Tobacco smoke exposure in nonsmoking hospitality workers before and after a state smoking ban. Cancer Epidemiol. Biomark. Prev. 19 (4), 1016–1021. Judkins, D.R., 1990. Fay's method for variance estimation. J. Off. Stat. 6, 223–239. Klepeis, N.E., Ott, W.R., Switzer, P., 2007. Real–time measurement of outdoor tobacco smoke particles. J. Air Waste Manage. Assoc. 57 (5), 522–534. Millett, C., Lee, J.T., Laverty, A.A., Glantz, S.A., Majeed, A., 2013. Hospital admissions for childhood asthma after smoke-free legislation in England. Pediatrics 131 (2), e495–e501. Mills, A.L., Messer, K., Gilpin, E.A., Pierce, J.P., 2009. The effect of smoke-free homes on adult smoking behavior: a review. Nicotine Tob. Res. ntp122. Mulcahy, M., Evans, D.S., Hammond, S.K., Repace, J.L., Byrne, M., 2005. Secondhand smoke exposure and risk following the Irish smoking ban: an assessment of salivary cotinine concentrations in hotel workers and air nicotine levels in bars. Tob. Control. 14 (6), 384–388. Nagelhout, G.E., Willemsen, M.C., de Vries, H., 2011. The population impact of smoke-free workplace and hospitality industry legislation on smoking behaviour. Findings from a national population survey. Addiction 106 (4), 816–823. Naiman, A., Glazier, R.H., Moineddin, R., 2010. Association of anti-smoking legislation with rates of hospital admission for cardiovascular and respiratory conditions. Can. Med. Assoc. J. 182 (8), 761–767. Öberg, M., Jaakkola, M.S., Woodward, A., Peruga, A., Prüss-Ustün, A., 2011. Worldwide burden of disease from exposure to second-hand smoke: a retrospective analysis of data from 192 countries. Lancet 377 (9760), 139–146. Okoli, C.T., Kelly, T., Hahn, E.J., 2007. Secondhand smoke and nicotine exposure: a brief review. Addict. Behav. 32 (10), 1977–1988. Repace, J., Zhang, B., Bondy, S.J., Benowitz, N., Ferrence, R., 2013. Air quality, mortality, and economic benefits of a smoke-free workplace law for non-smoking Ontario bar workers. Indoor Air 23 (2), 93–104. St Helen, G., Thomas Bernert, J., Hall, D.B., et al., 2012. Exposure to secondhand smoke outside of a bar and a restaurant and tobacco exposure biomarkers in nonsmokers. Environ. Health Perspect. 120 (7), 1010. Thomson, G., Wilson, N., Edwards, R., 2009. At the frontier of tobacco control: a brief review of public attitudes toward smoke-free outdoor places. Nicotine Tob. Res. ntp046. U.S. Department of Health and Human Services, 2006. The Health Consequences of Involuntary Exposure to Tobacco Smoke: A Report of the Surgeon General. 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. U.S. Environmental Protection Agency, 1992. Respiratory Health Effects of Passive Smoking: Lung Cancer and Other Disorders. U.S. Environmental Protection Agency, Office of Health and Environmental Assessment, Office of Research and Development, Washington, DC (EPA/600/6-90/006 F). Valente, P., Forastiere, F., Bacosi, A., et al., 2007. Exposure to fine and ultrafine particles from secondhand smoke in public places before and after the smoking ban, Italy 2005. Tob. Control. 16 (5), 312–317. WHO Report on the Global Tobacco Epidemic, 2009: Implementing Smoke-free Environments. World Health Organization, Geneva (http://www.who.int/tobacco/mpower/ 2009/en/index.html, accessed 28 October 2014). Yao, T., Sung, H.Y., Mao, Z., Hu, T.W., Max, W., 2014. The healthcare costs of secondhand smoke exposure in rural China. Tob. Control. http://dx.doi.org/10.1136/ tobaccocontrol- 2014-051761 (tobaccocontrol-2014).
