Article Text
PDF
Abstract
Objectives Smoke-free policies not only reduce harm to non-smokers, they may also reduce harm to smokers by decreasing the number of cigarettes smoked and increasing the likelihood of a successful quit attempt. However, little is known about the impact of exposure to smoking on patios on smoking behaviour.
Design and participants Smokers from the Ontario Tobacco Survey, a longitudinal population representative cohort of smokers (2005–2011). There were 3460 current smokers who had completed one to six follow-ups and were asked at each follow-up whether or not they had been exposed to smoking on patios in the month.
Main outcome measures Generalised estimating equations and survival analysis were used to examine the association between exposure to patio smoking and smoking behaviour changes (making a quit attempt and time to relapse after a quit attempt), controlling for potential confounders.
Results Smokers who were exposed to smoking on patios (adjusted incident rate ratio (aIRR) = 0.89; 95% CI 0.81 to 0.97) or had been to a patio (aIRR = 0.86; 95% CI 0.74 to 0.99) were less likely to have made a quit attempt than smokers who had not visited a patio. Smokers who were exposed to smoking on patios were more likely to relapse (adjusted HR=2.40; 95% CI 1.07 to 5.40)) after making a quit attempt than those who visited a patio but were not exposed to smoking.
Conclusions Exposure to smoking on patios of a bar or restaurant is associated with a lower likelihood of success in a quit attempt. Instituting smoke-free patio regulations may help smokers avoid relapse after quitting.
- Cessation
- Secondhand smoke
- Environment
Statistics from Altmetric.com
Introduction
The health risks of second-hand smoke are well established with no risk-free level of exposure.1 Associated health effects include stroke, lung cancer, coronary heart disease, low birth weight, nasal irritation, middle ear disease, respiratory symptoms, impaired lung function, lower respiratory illness, and sudden infant death syndrome.1 Several jurisdictions have found dramatic decreases in hospital admissions for worsening of chronic obstructive pulmonary disease (COPD) and acute coronary syndrome following the implementation of smoke-free policies.2–4 This evidence supports eliminating exposure to second-hand smoke as a public health priority.
In many jurisdictions, including the province of Ontario, most indoor workplaces, bars and restaurants are covered by smoke-free laws, bylaws, and ordinances. While some outdoor environments such as transit areas, open air stadiums and school grounds are regulated at the local or national level; a number of jurisdictions in Australia, New Zealand, the USA and Canada have sought to reduce second-hand smoke exposure further by implementing more comprehensive outdoor smoke-free laws and regulations including prohibiting smoking on restaurant and bar patios.5 The Smoke-Free Ontario Act implemented in May 2006 banned smoking in all enclosed public spaces and workplaces, however the legislation permitted smoking on patios that are not covered by a roof.
Outdoor smoke-free patio ordinances are generally justified in three ways. First, waitstaff and patrons in outdoor areas can be exposed to substantial concentrations of tobacco smoke,6 particularly within 1–2 m of a lit cigarette. Outdoor smoke may also drift into protected indoor spaces.7 Second, outdoor smoke-free laws are believed to change social norms by influencing the social climate around tobacco use; thus, reducing exposure could lead to denormalisation and decreased use among current and susceptible smokers. Numerous jurisdictions have shown changes in attitudes and norms following smoke-free regulations.8–10 Third, smoke-free laws may have a direct effect on reducing harm to smokers by encouraging quitting behaviour through eliminating visual cues and physical exposure to airborne nicotine which can trigger relapse and undermine efforts to quit and maintain abstinence. There is evidence that indoor smoke-free restrictions in workplaces, public places and homes lead to decrease in the number of cigarettes smoked per day and increases in the likelihood of attempting to quit.11–15 Smokers in Ireland also reported that their comprehensive smoke-free legislation helped them sustain their abstinence.15
To the best of our knowledge, there are no quantitative studies of the impact of patio smoking bans on the smoking cessation behaviours of affected patrons. Outdoor bans represent one of the more innovative elements in progressive health policies towards reduction of harm from tobacco. We hypothesise that exposure to smoking on patios decreases the likelihood of a smoker making a quit attempt and increases the chance of relapse following a quit attempt. This study examines whether exposure to smoking on patios is associated with the likelihood of making a quit attempt and the risk of relapse after a quit attempt. We also investigate the correlates of exposure to smoking on patios among smokers and non-smokers.
Methods
Participants
Data from the Ontario Tobacco Survey, a population-based longitudinal survey of adult smokers with cross-sectional samples of non-smokers in Ontario, Canada were used to examine these questions. Smokers were followed for up to 3 years at approximately 6-month intervals. Recruitment took place between July 2005 and June 2008 with the final longitudinal panel ending in 2011. Non-smokers were sampled concurrently with smokers, but were not followed longitudinally. The response rate was 61%, with 80% retention at 12-month follow-up. The distribution of sample characteristics was consistent with census data; further details of the sampling and recruitment processes of the survey are documented elsewhere.16 Research ethics approval for this study was obtained from the University of Toronto and the University of Waterloo.
There were three different analytic sets used in this paper (see figure 1). To examine correlates of exposure to smoking on patios, eligible respondents were smokers and non-smokers with data on exposure to second-hand smoke and restaurant/bar attendance (n=4997). Questions on exposure were not asked in the first two waves of the study. All observations prior to the implementation of the Smoke-Free Ontario Act, which banned smoking inside bars and restaurants in all jurisdictions in May 2006, were excluded from all analyses.
Analytic population sets.
To assess quit attempts, the analytic set included those who were current smokers at baseline and had completed at least one of the 6-month, 12-month or 18-month follow-up interviews (n=3460 for the quit attempt objective). Respondents could contribute multiple observations: 1960 respondents contributed three sets of observations, 846 contributed twice, and 654 once for a total of 8226 observations.
To assess rates of relapse after a quit attempt, a subsample of 373 people from the quit attempt analytic set were examined. To be eligible for this analysis, the individuals must have been current smokers at the previous survey, made a quit attempt in the past 6 months and have been currently abstinent for at least 30 days. These restrictions ensured that the quit attempt preceded the assessment of past 30 day exposure to smoking on a patio, and that the relapse event occurred subsequent to the exposure. For example, if a participant reported a quit attempt that began within the past 2 weeks, they would be excluded because the quit attempt may have begun subsequent to the patio visit.
Measures
Exposure to patio smoking was assessed through a combination of six questions asking if the respondents had been to a bar or a restaurant in the past 6 months and if yes, how often they had sat on patios in the past 30 days. Respondents who sat on patios ‘some’ or ‘all of the time’ were then asked whether or not there were other people smoking around them. This measure could include both direct physical exposure to second-hand smoke and visual exposure to smoking (cues). Exposure to smoking on patios was classified in four ways, based on past 30-day patio attendance and exposure as follows: ‘No bar or restaurant attendance’; ‘bar or restaurant attendance, but no patio attendance’; ‘patio attendance, not exposed to smoking’ and ‘patio attendance, exposed to smoking’.
Outcome measures
A quit attempt was defined using the response to “In the past 6 months, did you try to quit smoking completely?” as a self-reported measure for making a serious quit attempt; smokers who were abstinent for more than 30 days at the time of follow-up were considered to have made a quit attempt.
Time to relapse associated with patio smoking was assessed through length of abstinence subsequent to the quit attempt, derived from a series of questions on smoking behaviour. At the first follow-up, the duration of the quit attempt was the longest period reported without smoking a cigarette. Additional time was added to this figure, for the period of time that the participant reported continuous abstinence in subsequent follow-ups. If still abstinent, the period between survey dates of the individual was added to the length of abstinence at the first follow-up. If the participant reported smoking at a subsequent follow-up, a midpoint censoring rule applied, with one-half of time since last interview added to the duration of abstinence recorded. This process of adding duration of abstinence was repeated until the individual reported smoking, was lost to follow-up, or the study ended. This procedure produced a relatively continuous relapse curve.
Covariates
Potential confounders were assessed, including age, sex, marital status (married vs other), having children under 18 in the home (yes/no), level of education completed (less than high school, completed high school, some college or university, completed college or university), region of Ontario (North, Southwest, East, Greater Toronto Area), self-reported health (excellent to poor), daily or occasional smoking, season of interview (April–October vs November–March), survey date, perceived addiction (very, somewhat, not at all addicted), prior lifetime use of pharmaceutical quit aids (ie, nicotine patch, gum and spray, bupropion, and varenicline: yes/no), and prior lifetime use of behavioural quit aids (ie, self-help, counselling, group therapy: yes/no). For the present study, high and moderate nicotine dependence was based on the Heaviness of Smoking Index (HSI) and defined as a score of greater than 2,17 with less-than-daily smokers assigned a value of zero, although was included in regression models as a continuous variable.
Analysis
Association of individual characteristics and exposure to smoking on patios in Ontario was assessed among smokers and non-smokers. The association of exposure to smoking on patios with making a quit attempt was assessed among eligible smoking participants using Generalised Estimating Equation (GEE) regression models. The association between exposure to smoking on patios and time to relapse after a quit attempt was assessed using a Cox regression model among smokers who had made a quit attempt. Tests of the proportional hazards assumptions (plots of the survival function and Schoenfeld residuals) were performed and found to be satisfactory. Interactions between patio smoking and gender, heaviness of smoking and occasional smoking were examined and were statistically insignificant. All analyses accounted for the complex survey design and were conducted using Stata V.12 (StataCorp).
Results
Twenty per cent of adults in Ontario reported being exposed to smoking on patios between 2006 and 2008. Younger, male, and married respondents were over-represented among those exposed (table 1). Exposure to smoking on patios was more likely in the warmer months, with 68% of participants reporting exposure during April–October, and 32% reporting exposure in the colder months of November–March. Participants from the greater Toronto Area (with no complete bans on patios) were most likely to be exposed, while those in Eastern Ontario (including cities with comprehensive bans on smoking on patios, such as Ottawa and Kingston) were least likely to be exposed. In general, smokers were significantly more likely than non-smokers to report being exposed to patio smoking (74% of those exposed to smoking on patios were current smokers).
Associations between demographic characteristics and past 30-day exposure to smoking on patios among smokers and non-smokers in the Ontario Tobacco Survey at baseline
Those exposed to smoking on patios in the past 30 days were less likely to report making a quit attempt compared with those who did not visit a bar or restaurant (incident rate ratio, IRR: 0.89; 95% CI 0.81 to 0.97; table 2). Smokers who visited a patio but did not report exposure to smoking in the past 30 days were also less likely to report making a quit attempt than those who did not visit a bar or restaurant (IRR = 0.86; 95% CI 0.74 to 0.99). However, there was no difference in quit attempts between those who went to a patio and were exposed to smoking and those who went to a patio and were not exposed (IRR = 0.97; 95% CI 0.82 to 1.15; p=0.733).
Association between exposure to smoking on patios and making a quit attempt in next 6 months (n=460, observations=8226), using generalised estimating equations method
In the analysis examining the relationship of exposure to smoking on patios to relapse, only recent quitters who were abstinent for at least 30 days at the time of the survey were included. Those who were exposed to smoking on patios were more likely to relapse at the following interview than those who did not go to a bar or restaurant (HR = 1.65; 95% CI 1.07 to 2.55; table 3). The relationship was even stronger when compared with those who went to a patio but were not exposed to smoking (HR=2.40; 95% CI 1.07 to 5.40; p=0.033). Unlike the results for quit attempts, there was no significant difference in relapse between those who visited a patio but not exposed to smoking and those who did not visit a patio.
Association between time to relapse after a quit attempt and exposure to smoking on patios among recent quitters who were not smoking at the time of interview (n=373), using survival analysis
Discussion
Getting current smokers to quit and stay quit has an immediate public health benefit with impacts on heart disease and COPD-related deaths.18 While smoke-free patios may also have important health impacts, due to a reduction in physical exposure to smoke, and positive effects on changes in social norms among smokers and non-smokers, this study is the first to conclude that they may help smokers avoid relapse following a quit attempt. However, the results of this study suggest that smoke-free patios do not increase the likelihood of quit attempts.
Current smokers were much more likely than non-smokers to report visiting a patio where smoking was taking place. Smokers may seek out patio environments when visiting hospitality venues, as uncovered patios are one of the few remaining public places where people can smoke. Likewise, non-smokers may seek out smoke-free indoor spaces to avoid possible exposure to smoke on patios.19 Some regional variation in reported smoke exposure can be explained, in part, by regional or municipal by-laws. For example, in Eastern Ontario, the community of Kingston has a local by-law that requires all patios to be smoke-free; however, there were insufficient numbers of participants during the times the by-laws were in place to analyse separately.
Previous studies have found that exposure to second-hand smoke at home, at bars and restaurants, and in workplaces is associated with greater numbers of cigarettes smoked, reduced likelihood of quit attempts, and increased chance of relapse.20 ,21 A three-country study in England, the Netherlands and Ireland, comparing quit attempts and success before and after the introduction of smoke-free laws, reported that a comprehensive bar and restaurant ban (without specific patio coverage) was effective at increasing quit attempts and quit success, while partial laws had no effect.22 ,23 However, most comprehensive bans that do not include patios increase outdoor smoking behaviour, as smokers take their cigarettes outside.19 This study suggests that outdoor exposure does impact quit success; however, there was no effect of smoke exposure over and above the effect of visiting a patio with respect to quit attempts. Most quit attempts are planned,24 and this suggests the relationship between smoke-free indoor places and quit attempts may operate through a different mechanism.
One potential mechanism for the association of relapse behaviour with exposure to smoking on patios is through introducing social cues to smoke. Previous research has found that exposure to social or visual cues for smoking is associated with a decrease in motivation to quit and increased risk of relapse.25–29 This study used a measure of exposure that did not distinguish between exposure to sensory (visual, odour) cues related to smoking and physical exposure to nicotine in second-hand smoke. Further research is needed to distinguish between the relative effects of the visual and odour cues and the direct effects of exposure to smoke.
The interpretation of the association must be taken with caution. People were able to choose whether or not they were exposed to smoke on patios. It is evident that people who go to patios are quite distinct from those who do not go out or do not visit patios. Although we controlled for a number of demographic and smoking-related characteristics, there may still be residual confounding. The reduced likelihood of quit attempts among all individuals who had visited a patio, irrespective of smoke exposure, suggests that there may be characteristics of people who visit patios per se that make them more resistant to quitting. Smokers may seek out patios that permit smoking, particularly for social interactions, while the nature of the visit to the patio (eating, drinking, socialising) was not assessed in this study, we did find that smoke exposure alone, and not simply the characteristics of people visiting patios, was associated with relapse, suggesting that the active mechanism is specific to either physical or social exposure to smoking. Another potential limitation is that the measurement of exposure to smoke and visiting a bar and patio was self-reported, and participants may not have reported exposure to smoke from their own cigarettes. However, this is a greater concern with the analysis of quit attempts, as patio exposure was measured prior to, and independently of, cessation behaviour studied and so the true risk of response-bias due to non-objective measurement of patio attendance should be smaller for the relapse analysis. Exposure to smoke may also vary, as a patio may be an area people are required to pass through, akin to a doorway. Smokers who are struggling with a quit attempt may be more likely to notice and report this type of exposure as they pass through.
The association between exposure to smoking on patios and relapse could also be affected by alcohol use, which was not measured directly in this survey. However, there was no association between relapse and attendance at a bar or restaurant where alcohol use was also likely to have occurred, implying that the effect was specific to social or physical exposure to smoking. In Ontario, the interior of bars and restaurants were officially smoke-free, and exposure to smoking indoors was minimal. Further research could help understand how changes in regulations impact changes in smoking behaviour.
Conclusion
Visiting a patio, irrespective of smoke exposure, was associated with a decreased likelihood of attempting to quit smoking; however, exposure to smoking on patios was associated with relapse among smokers who had made a quit attempt. Exposure to second-hand smoke is known to have direct health effects.30 These results suggest that regulations that reduce smoking on patios may also benefit health by helping recent quitters stay smoke-free.
What this paper adds
Getting current smokers to quit and stay abstinent remains one of the most significant outcomes of tobacco control policy.
This paper provides important evidence of a potential contribution of outdoor smoking bans in meeting this objective.
This population representative longitudinal study of smokers demonstrates that exposure to smoking on patios during an ongoing quit attempt increases the risk of relapse.
References
Footnotes
Contributors MC drafted the manuscript and conducted the analysis. RF, LD, RDK, BZ, JEC and SJB provided substantial intellectual contributions to the writing of the manuscript and the interpretation of results. LD, SJB, RF and JEC contributed to the design of the underlying cohort study. All authors were involved in the design and conceptualisation of the study and provided final approval.
Funding The Ontario Tobacco Research Unit is supported by the Ontario Ministry of Health and Long Term Care. This work is supported by Canadian Cancer Society grant #702160 (MC).
Competing interests None.
Ethics approval University of Toronto/University of Waterloo.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Researchers can apply to use the data from the Ontario Tobacco Survey to conduct secondary analyses for research and evaluation; see the OTS Information and Guidelines for data applicants: http://otru.org/wp-content/uploads/2012/08/OTS_Information_Guidelines_Aug2012.pdf.