Background Passive smoking is a serious health risk in non-smokers. The strength of tobacco control policies of the EU countries vary. This study aimed to investigate the relationship between secondhand smoke exposure, as assessed by exhaled carbon monoxide (eCO), and the strength of national tobacco control policies, in non-smokers in the EU.
Methods Data were provided from the EU campaign ‘HELP: for a life without tobacco’ during national events settled in the 27 EU countries in 2006–2007. Individual information on age, gender, and eCO was obtained from 58 919 self-reported non-smokers. The strength of national tobacco control policies was scored by the Tobacco Control Scale (TCS). The relationship between eCO and TCS score was investigated using ecological and multilevel approaches.
Results Both analyses reported a significant linear decrease in eCO per unit increase in TCS score, with a smaller estimation resulting from the multilevel analysis (β=−0.03 ppm, 95% CI −0.04 to −0.02 vs −0.05, −0.02 to −0.08).
Conclusion The present study confirms, in a large European non-smoker population, the relevance of strong antismoking policies in reducing exposure to passive smoking. The findings give further reason to encourage European countries to strengthen their tobacco control policies to protect non-smokers from secondhand smoke.
- Air pollution, passive smoking
- prevention PR, public health policy
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Exposure to secondhand smoke (SHS) has been shown to be responsible for a wide range of diseases and premature deaths in non-smokers.1 This has led supranational authorities, including the European Union (EU), to launch initiatives promoting a smoke-free lifestyle, by enforcing the legislative framework and by encouraging health promotion.2 National policies to limit exposure to secondhand smoke nevertheless vary widely among EU member states. Instruments have been developed to evaluate national policies, based on the six most effective tobacco control measures, providing a quantitative ranking of EU countries.3 To raise awareness, the EU funded in 2005 the 4 year campaign ‘HELP: for a life without tobacco’ using media and national events in the 27 EU member states. The present study explores, using data collected during this campaign, the relationship between SHS exposure, as assessed by exhaled carbon monoxide (eCO) and the strength of national-level tobacco control policies, as scored by the Tobacco Control Scale (TCS), in non-smokers in the 27 EU member states.3
The campaign HELP collected data between March 2006 and July 2007 on a convenience sample of 111 835 people interviewed in several cities of the 27 EU countries where national events were organised. Stands were settled in various places such as public places and healthcare facilities. The present cross-sectional study included 58 919 self-reported non-smokers with eCO concentration lower than 10 ppm. This threshold was chosen to exclude smokers who denied their status (5.6% of self-reported non-smoker participants).4
Individual information was obtained on age, gender and smoking habits from a standardised interview and on eCO in expired air after breath-holding with a portable CO analyser, using an established protocol.5 An ecological variable, the TCS, was used to measure the strength of the national tobacco control policies. The TCS, first published in 2005 and updated in 2007, provides an overall national score based on six different tobacco control interventions, namely price and taxes level, clean indoor air laws, comprehensive advertising bans, funding for tobacco control programmes, characteristics of health warnings on tobacco packs and affordability of tobacco dependence treatments.3 6 The 2005 TCS was attributed if no change in tobacco control policies had occurred in the country between 2005 and the date of interview; the 2007 score was attributed, otherwise. This score increases with the strength of tobacco control policies up to a possible maximum of 100 points.
An ecological analysis was initially carried out to estimate the effect of TCS score on eCO, averaged over the country level, using a linear model. Examination of residuals scatterplots found no important deviation from the assumption of normality. A multilevel linear analysis was also performed to assess the association between TCS score and eCO, measured at individual-level, after adjusting for age, gender and survey setting.7 In the multilevel approach, the possible clustering of non-smokers within countries in relation to individual eCO was also analysed by the intraclass correlation coefficient (ICC).8 In both analyses, the strength of the association between TCS score and eCO was appraised by the β coefficient (95% CI). The percentage of country differences in eCO explained by the adjusted model was estimated by the adjusted R2 for the ecological analysis and by the R22 as described by Merlo et al for the multilevel analysis.8
Non-smokers in the sample were interviewed most frequently in public places (49%), followed by healthcare (13%) and educational (13%) facilities, sport (10%), cultural (7%), and music (4%) events, and lastly in workplaces (4%). More than 50% of non-smokers were less than 30 years old and 48.7% were men. Mean eCO ranged from 1.6 ppm (SD 2.1 ppm) in Ireland to 6.6 ppm (SD 1.9 ppm) in Greece. The TCS score varied from 26 in Luxembourg to 74 in Ireland. A significant linear 0.05 ppm decrease in mean eCO per unit increase in TCS score was estimated by the ecological analysis (95% CI −0.02 to −0.08, p=0.00245) (figure 1, table 1). This model explained 23.6% of the variation in mean eCO. The multilevel model showed a noticeable clustering of non-smokers within countries by an intraclass correlation of 0.28 (table 1). Although individual variables (age, gender and survey setting) were significantly associated with individual eCO, they explained only 4% of the differences in eCO between countries (table 1). Adjustment for individual variables and TCS score led to an intraclass correlation decrease and explained 27.7% of the differences in eCO between countries. A significant linear 0.03 ppm decrease in eCO per unit increase in TCS score was estimated by the multilevel analysis (95% CI −0.02 to −0.04, p<0.0001) (table 1).
The present study found that eCO concentration decreased with the strength of tobacco control policies, as scored by the TCS in a large non-smoker European population.
However, this study has several limitations. The HELP campaign aimed to highlight the consequences of active and passive smoking and to promote tobacco-free lifestyles throughout the EU. The CO concentration was used to sensitise individuals to the role of tobacco smoke as a pollutant of the whole body. Its measurement in expired air was chosen as it could be performed non-invasively and for its cost-effectiveness advantages. Exhaled CO was, thus, in this study, the most appropriate marker of individual SHS exposure. However, CO is not specific to tobacco smoke, as it can also reflect exposure to other environmental sources such as traffic, occupational or domestic emissions. Therefore, further investigations are needed to validate the present results. They should include more detailed information on tobacco smoke exposure of non-smokers relating to sources and intensity of involuntary exposure at home or at work. In smaller scale studies, cotinine is commonly used as a specific marker. However, its measure in saliva or urine is costly and complex to perform. Nevertheless, Haw et al showed, by using cotinine concentrations in a national-level study, a large and significant reduction in SHS exposure in non-smoking adults after the implementation of smoke-free legislation.9
The present exploratory findings support these results on a large sample from 27 countries, at the European Union level, by confirming the relevance of strong antismoking policies integrating a set of tobacco control interventions for the reduction of passive smoking. They constitute a further reason to encourage the EU member states with the weakest tobacco control policies to strengthen them, in particular to protect non-smokers from SHS. This could be achieved by launching comprehensive smoke-free legislation with the support of the population, by following the example of Ireland.10
What this study adds
Passive smoking is a serious health risk in non-smokers. In the European Union, the scope and character of national tobacco control policies vary widely among the member states, as indicated by scores on the Tobacco Control Scale (TCS).
The present measures of exhaled carbon monoxide concentration in a large sample of European non-smokers support the relevance of strong antismoking policies, as scored by the TCS, for the reduction of passive smoking.
These findings give further reason to encourage European countries to strengthen their tobacco control policies.
We wish to thank the 27 member states of the study and all HELP events organisers and participants, and national correspondents.
Funding BD manages the Office Français de Prévention du Tabagisme (OFT), a non-governmental organisation, that received partial funding from the European Union for this study.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.