The authors have declared that no competing interests exist.
Conceived and designed the experiments: EF. Performed the experiments: XS MF CM EC ES EF. Analyzed the data: XS JMMS MF MJL. Contributed reagents/materials/analysis tools: RPO JAP. Wrote the paper: XS EF. Critically revised the manuscript and contributed to the final version: XS JMMS MF RPO CM EC MJL ES JAP EF.
In 2006, Spain implemented a national smoke-free legislation that prohibited smoking in enclosed public places and workplaces (except in hospitality venues). In 2011, it was extended to all hospitality venues and selected outdoor areas (hospital campuses, educational centers, and playgrounds). The objective of the study is to evaluate changes in exposure to secondhand smoke among the adult non-smoking population before the first law (2004-05) and after the second law (2011–12).
Repeated cross-sectional survey (2004–2005 and 2011–2012) of a representative sample of the adult (≥16 years) non-smoking population in Barcelona, Spain. We assess self-reported exposure to secondhand smoke (at home, the workplace, during leisure time, and in public/private transportation vehicles) and salivary cotinine concentration.
Overall, the self-reported exposure to secondhand smoke fell from 75.7% (95%CI: 72.6 to 78.8) in 2004-05 to 56.7% (95%CI: 53.4 to 60.0) in 2011–12. Self-reported exposure decreased from 32.5% to 27.6% (−15.1%, p<0.05) in the home, from 42.9% to 37.5% (−12.6%, p = 0.11) at work/education venues, from 61.3% to 38.9% (−36.5%, p<0.001) during leisure time, and from 12.3% to 3.7% (−69.9%, p<0.001) in public transportation vehicles. Overall, the geometric mean of the salivary cotinine concentration in adult non-smokers fell by 87.2%, from 0.93 ng/mL at baseline to 0.12 ng/mL after legislation (p<0.001).
Secondhand smoke exposure among non-smokers, assessed both by self-reported exposure and salivary cotinine concentration, decreased after the implementation of a stepwise, comprehensive smoke-free legislation. There was a high reduction in secondhand smoke exposure during leisure time and no displacement of secondhand smoke exposure at home.
Exposure to secondhand smoke (SHS) has been causally associated with many adverse health effects
Exposure to SHS can occur in different settings, including in the home, at the workplace, in other private and public places (bars, restaurants, cafes, etc.), and inside public and private transport vehicles. Questionnaires, biomarkers, and airborne markers have been used to evaluate SHS among non-smokers. The prevalence of SHS exposure in adult non-smokers varies considerably, depending on the country, the development of the tobacco epidemic
On the 1st of January, 2006, a smoke-free legislation (Law 28/2005) was implemented in Spain to protect the health of non-smokers. The legislation banned smoking in all public and work places, with some exceptions in hospitality venues (no ban in venues of less than 100 m2, and ‘smoking areas’ were allowed in venues over 100 m2)
The objective of this study was to evaluate whether a measurable change in SHS exposure could be detected in the adult non-smoking population with the implementation of the stepped Spanish smoke-free legislation. We compared SHS exposure measurements (self-report data and levels of salivary cotinine) before the first law (2004–05) and after the second law (2011–12) legislation.
This study had a repeated cross-sectional design. We included a representative, random sample of the population of Barcelona (Spain). Surveys were conducted before and after the implementation of smoke-free legislation. The pre-legislation data were obtained between March 2004 and December 2005. We used the same strategy to collect the post-legislation data between June 2011 and March 2012. Detailed information about the pre-legislation survey (sampling, face-to-face questionnaire, saliva collection, and cotinine analysis) has been provided in previous studies
In brief, for each survey, we determined a sample size of 1,560 people with standard procedures (α error of 5%, beta error of 20%, and 20% losses for independent samples). The pre-legislation survey (years 2004–05), included a final sample of 1,245 individuals and the post-legislation survey included a final sample of 1,307 individuals. These sample sizes were sufficient to detect 10% changes in the amount of exposure to SHS at the workplaces or at home (under the least favorable conditions) and a 40% difference in salivary cotinine concentrations between the two surveys. Sample size calculations were performed with 5.2 GRANMO MS Windows (
We obtained data and addresses for Barcelona residents from the updated official city census (years 2001 and 2010) provided by the Municipal Institute of Statistics of Barcelona. Individuals aged 16 years and older were eligible to participate in the study. A letter was mailed to eligible individuals to inform them about the purpose of the study and that they had been selected at random. The letter also informed them that the study required a visit from an interviewer that would administer the questionnaire and collect a saliva sample. The individuals were informed that they were free to decline participation, and that they could find out more about the study with a telephone call or email; the contact information was provided in the letter. Participants that could not be located after several attempts (at different times of the day and different days of the week) and those that declined to participate in the study were replaced at random. The replacements were chosen from eligible individuals of the same sex, within a 5-year age group, and within the same district of residence. Substitutions accounted for 50.7% and 54.6% of the pre- and post-legislation surveys, respectively. Individuals that agreed to participate were interviewed at home by trained interviewers. Participants were asked to sign an informed consent form before proceeding with the face-to-face interview. In case of subjects aged 16 an 17, parental written consent was obtained. The same questionnaire was used in both surveys (on traditional paper in the pre-legislation survey and in computer-assisted form in the post-legislation survey). Additional questions were included in the second survey regarding the smoke-free legislation. The questionnaire included information on socio-demographics, tobacco consumption, self-assessed exposure to SHS in different settings (at home, work/educational venues, during leisure time, and in public and private transportation vehicles), and attitudes toward smoking restrictions. After completing the questionnaire, respondents were asked to provide a sample of saliva for the cotinine analysis, and weight and height were measured. The Research and Ethics Committee of Bellvitge University Hospital approved the study protocols and the informed consent forms, including parental written consent.
Non-smokers were defined as individuals that, at the time of the interview, reported that they did not smoke, and they had a salivary cotinine concentration ≤10 ng/mL
We asked the participants to provide a saliva sample to determine the cotinine levels. Cotinine is the main metabolite of nicotine; it is a stable, specific, sensitive biomarker of tobacco smoke in biological fluids, with a half-life of 15–17 h, and it reflects SHS exposure in the last 5–7 days
We calculated prevalence rates (%) and 95% confidence intervals (CI) for exposure to SHS among non-smokers in the different settings. Results were stratified by sex, age (16–44, 45–64, and ≥65 years), and educational level (less than primary and primary school, secondary school, and university). The data were fitted with multivariate log-binomial models to assess the prevalence ratios (PR) and 95% CI of exposure to SHS among non-smokers before and after the implementation of the legislation. The models were adjusted for sex, age, and educational level. Geometric means (GM) and geometric standard deviations (GSD) were computed to describe the cotinine concentrations among non-smokers, due to its skewed distribution
A total of 2,552 participants were interviewed; 1,245 subjects were in the pre-legislation survey and 1,307 were in the post-legislation survey. The samples were similar in the proportions of men and women, but we found significant differences in age and educational level. 879 (70.6%) participants in the pre-legislation survey and 947 (72.5%) participants in the post-legislation survey were self-reported non-smokers. Of the non-smokers, 110 (62 in the pre-legislation and 48 in the post-legislation surveys) were not included in the analysis, because they did not provide a saliva sample; in addition, 12 (10 in the pre-legislation and 2 in the post-legislation survey) were excluded, because cotinine analysis was not possible (i.e., insufficient sample). 83 non-smokers from the pre-legislation survey and 19 from the post-legislation survey were excluded, because they had cotinine concentrations consistent with active smoking (>10 ng/mL). Therefore, the final sample for analysis included a total of 1602 non-smokers; 724 (58.2% of those interviewed) before the legislation and 878 (67.2% of those interviewed) after the legislation (
Footnote to
The prevalence of self-reported exposure to SHS in any setting fell from 75.7% in 2004–05 to 56.7% in 2011–12 (relative reduction −25.1, p<0.001) (
Self-reported exposure to secondhand smoke | n | % of non-smokers exposed (95% CI) | Prevalence ratio |
Before the legislation | 720 | 75.7 (72.6–78.8) | 1 |
After the legislation | 871 | 56.7 (53.4–60.0) | 0.46 (0.40 to 0.54) |
Before the legislation | 721 | 32.5 (29.1–35.9) | 1 |
After the legislation | 878 | 27.6 (24.6–30.6) | 0.78 (0.65 to 0.94) |
Before the legislation | 364 | 42.9 (37.8–48.0) | 1 |
After the legislation | 507 | 37.5 (33.3–41.7) | 0.79 (0.63 to 0.98) |
Before the legislation | 723 | 61.3 (57.7–64.9) | 1 |
After the legislation | 872 | 38.9 (35.7–42.1) | 0.38 (0.32 to 0.44) |
Before the legislation | 626 | 12.3 (9.7–14.9) | 1 |
After the legislation | 669 | 3.7 (2.3–5.1) | 0.26 (0.16 to 0.41) |
Before the legislation | 585 | 9.4 (7.0–11.8) | 1 |
After the legislation | 616 | 10.7 (8.3–13.1) | 0.97 (0.67 to 1.41) |
*Based on multivariate log-binomial models, adjusted for sex, age, and educational level.
**The figures do not sum the total because of missing values.
Before legislation | After legislation | Percentage of change |
|||
N | GM (GSD) (ng/mL) | N | GM (GSD) (ng/mL) | ||
724 | 0.93 (4.01) | 878 | 0.12 (3.12) | 87.6 (76.7–102.0) | |
Men | 296 | 1.11 (3.65) | 380 | 0.12 (2.91) | 89.4 (80.6–102.1) |
Women | 428 | 0.82 (4.22) | 498 | 0.12 (3.28) | 86.1 (74.4–102.7) |
16–44 | 236 | 1.00 (3.66) | 361 | 0.12 (3.09) | 88.0 (78.1–102.7) |
45–64 | 234 | 0.82 (4.17) | 254 | 0.13 (3.18) | 85.4 (73.9–104.1) |
≥65 | 251 | 0.98 (4.19) | 263 | 0.11 (3.10) | 89.2 (80.6–102.9) |
Less than primary and primary | 342 | 0.87 (4.16) | 236 | 0.12 (3.27) | 86.1 (79.4–103.5) |
Secondary | 132 | 0.97 (3.95) | 341 | 0.14 (3.28) | 85.2 (73.7–104.3) |
University | 249 | 0.98 (3.83) | 300 | 0.10 (2.75) | 90.2 (82.2–102.1) |
GM: Geometric mean.
GSD: Geometric standard deviation.
*Based on the adjusted geometric mean derived from a generalized linear model that included all the variables in the table.
**The figures do not sum the total because of missing values.
This was the first study to evaluate using both self-reports and a personal biomarker of exposure to SHS the impact of the stepped Spanish smoke-free legislation (laws 28/2005 and 42/2010) on SHS exposure in different settings among adult non-smokers from the general population. We found that self-reported exposure to SHS and salivary cotinine levels significantly decreased after the implementation of the legislation. This reduction was observed at workplaces, during leisure time, and even in settings not regulated by the law, like in the home and public transportation.
The reduction in SHS exposure between 2004–05 and 2011–12 was greater for women than men and for individuals aged 45 to 64 compared with other age groups. Haw and Gruer
Data from another study in Spain showed that both airborne nicotine and PM2.5 decreased by more than 90% in bars and restaurants after the implementation of law 42/2010
The proportion of non-smokers that had undetectable cotinine concentrations increased from 7.3% before the 28/2005 law to 53.2% after the implementation of the 42/2010 law. Our results confirmed the positive impact of smoke-free laws on SHS exposure at the population level. For example, after legislation, in New York, Bauer et al.
In addition to this shift in the distribution of the non-smoking population towards lower levels of cotinine, the mean concentration declined from 0.93 ng/mL to 0.12 ng/mL (adjusted reduction of 87.6%). This reduction in cotinine concentration was greater than those obtained after the implementation of smoke-free legislation in New York
One potential limitation of the study was an information bias derived from the use of a questionnaire. Self-reported, adult non-smokers represented 70.6% of the participants interviewed in the pre-legislation survey and 72.5% in the post-legislation survey. These prevalences were consistent with data from the 2006 and 2011 Spanish National Health Interview Surveys (Ministerio de Sanidad y Consumo: Encuesta Nacional de Salud 2006, 2013). This limitation was reduced by using an objective, specific biomarker of SHS exposure, and by asking the participants about their exposure in both private and public places, including the home, work/educational venues, leisure venues, and transportation vehicles. Thus, we covered the primary settings where SHS exposure can occur.
Another limitation is that we did not have data after the first law and previous to the second law, thus preventing us to elucidate the separate effects of both laws, as would have been of great interest given the stepped nature of the Spanish smoke-free legislation. However, the interpretation of our results together with the previous studies focused on the first law allows to globally evaluating the effects of the Spanish smoke-free laws.
This was a repeated cross-sectional study, which was potentially more likely to be biased than a longitudinal study. However, longitudinal studies can be subject to some bias, due to the loss of participants in the follow-up, which reduces its advantages. Nevertheless, repeated cross-sectional surveys that include a biological marker have been shown to be a valid method for evaluating smoke-free legislation
This study included representative, random samples of the population of Barcelona (Spain) and it evaluated the impact of smoke free legislation on exposure to SHS with a combination of self-reported exposure and cotinine as an objective biomarker of SHS exposure. To minimize differences between the two collection periods, we used the same strategy in collecting the pre and post legislation data. Additionally, the fieldwork was performed during different days of the week, including weekends, and in different months to avoid systematic biases due to potential seasonal and timing aspects of data collection. The method for analyzing cotinine in the post legislation survey was more sensitive and had a lower limit of quantification than that used in the pre legislation survey. However, we reanalyzed the samples in the pre-legislation survey with the new method, and found satisfactory agreement in the results. Individuals that declined to participate were replaced at random with individuals with the same characteristics to prevent problems with sample size and selection biases. Although we had a high percentage of substitutions in both surveys, we obtained a high percentage of non-smokers that provided saliva samples in the pre- and post- legislation surveys (92.9% and 94.9%, respectively); this proportion was higher than those observed in similar assessments in Scotland (64.8% and 63.1%, respectively)
This study showed that the implementation of a stepped smoke-free legislation (laws 28/2005 and 42/2010) was accompanied by a large reduction in SHS, both self-reported and assessed by means of salivary cotinine levels, in the adult non-smoking population in Barcelona, Spain. The strategy of strengthening Law 28/2005 to hospitality venues without exceptions was clearly effective. We observed a high reduction in SHS exposure during leisure time, and a reduction in SHS exposure at home contrary to the speculative tobacco industry hypothesis of displacement of smoking from public to private places. Based on the results of this study, comprehensive tobacco control policies were effective in reducing SHS exposure. Thus, over time, the law will result in a reduction in morbidity and mortality among nonsmoking adults.
(DOCX)
(DOCX)
The authors would like to thank Antonio Agudo, Anna Schiaffino, and Jorge Twose for participating in the pre-legislation study. In addition, we would like to thank Montse Ferré and Lucía Baranda for coordinating the fieldwork in the second study. The authors would also like to thank the field workers for their collaboration, and Chupa-Chups Spain for providing the SmintR candies.