Changes in air quality and second-hand smoke exposure in hospitality sector businesses after introduction of the English Smoke-free legislation

Transcription

1 Journal of Public Health Vol. 30, No. 4, pp doi: /pubmed/fdn062 Advance Access Publication 23 July 2008 Changes in air quality and second-hand smoke exposure in hospitality sector businesses after introduction of the English Smoke-free legislation N. K. Gotz 1,2, M. van Tongeren 3, H. Wareing 2, L. M. Wallace 1, S. Semple 4, L. MacCalman 3 1 Applied Research Centre in Health and Lifestyle Interventions, Coventry University, Coventry CV1 5FB, UK 2 Tobacco Control Collaborating Centre for England, Warwick CV34 4RA, UK 3 Institute of Occupational Medicine, Edinburgh EH14 4AP, UK 4 Department of Environmental and Occupational Medicine, University of Aberdeen, Aberdeen AB25 2ZP, UK Address correspondence to Nina Gotz, nina.gotz@coventry.ac.uk ABSTRACT Background To monitor and disseminate the short-term effects of the English Smoke-free legislation on air quality and employee exposure in businesses of the hospitality industry. Methods Indoor particle concentrations and salivary cotinine levels were measured in businesses in the hospitality sector and non-smoking employees one month before and after the implementation of the legislation. Results were immediately released to the media to announce the improvements in air quality and employee exposure to the wider public. Results Measurements were collected in 49 businesses and from 75 non-smoking individuals. Indoor PM 2.5 concentrations decreased by 95% from 217 mg/m 3 at baseline to 11 mg/m 3 at follow-up (P, 0.001). Salivary cotinine in employees was reduced by 75%, from 3.6 ng/ml at baseline to 0.9 ng/ml at follow-up (P, 0.001). The findings were presented to the public through press releases and interviews and were cited in over 20 media articles. Conclusion The project demonstrates the positive effects of the English Smoke-free legislation on air quality and second-hand smoke exposure in the hospitality industry sector. We believe that quick and positive feedback to the public on the effects of smoking restrictions is essential when introducing public health legislation such as the Smoke-free legislation. Keywords air pollution, public health, tobacco Introduction When the Smoke-free legislation was introduced in England on 1 July 2007, 1 restrictions on smoking in public places and workplaces had already been implemented in several countries around the world. 2 6 The available evidence suggested that legislating for smoke-free environments in these countries had led to substantial reductions in second-hand smoke (SHS) exposure, 7 11 improved health outcomes, and reductions in smoking prevalence Compliance with the law had generally been achieved, and trade for the hospitality and leisure industry seemed largely unaffected Health policy makers therefore expected that new, comprehensive legislation in England would reduce the health risks related to SHS exposure and make it easier for smokers to give up smoking. Reduced illness and mortality would result in substantial savings for the health service and the economy While the expert consensus was overwhelmingly in favour of the Smoke-free law, it was less clear what the public reaction would be. In other countries, initial resistance during the early planning stages had been followed by mostly positive attitudes once legislation was in place. 16,27,31,32 However, it was also noted that there was a time lag in feeding back the results of Smoke-free legislation to the general public. This may have created a vacuum which allowed negative N. K. Gotz, Research Associate M. van Tongeren, Head of Exposure Assessment H. Wareing, Director L. M. Wallace, Director of Research S. Semple, Senior Lecturer L. MacCalman, Statistician # The Author 2008, Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved. 421

2 422 JOURNAL OF PUBLIC HEALTH publicity to get a high profile. 33 It was therefore crucial to monitor the impact of the legislation during the first months following its implementation and to disseminate the findings while Smoke-free was still a high profile media topic. The project described here assessed air quality and employee exposure to SHS in hospitality and leisure venues one month before and one month after the legislation was introduced. Following data checking and analyses, results were immediately released to the media to demonstrate the changes that had taken place within the first two months after implementation. Methodology Baseline measures of indoor PM 2.5 levels and salivary cotinine of employees were carried out one month prior to the introduction of the legislation in June 2007 in the hospitality and leisure sector of six regions in England (East of England, East Midlands, North East, South East, South West and West Midlands). Follow-up measurements were taken one month after the introduction of the legislation in August The Tobacco Control Collaborating Centre (TCCC) worked collaboratively with Regional Tobacco Policy Managers of the six regions and officers from local services, such as smoking cessation services and Environmental Health departments. These agencies carried out site visits to businesses in the hospitality and leisure industry. The TCCC trained fieldworkers in the data collection procedures, provided technical support and collated the incoming data. Recruitment of venues and employees A convenience sample of establishments was selected for the project, including public houses, bars, clubs, bingo halls, private member clubs, cafés and betting shops. Fieldworkers recruited local businesses in person and by letter, and Regional Tobacco Policy Managers ensured that their samples included a diverse variety of locations, business types and sizes. At each site, two non-smoking employees who were present on the premises during the time of the visit were asked to provide saliva samples at baseline and follow-up. Only employees who reported not being exposed to any other sources of SHS (e.g. not living with a smoker) were tested. Salivary cotinine remains in the body for at least h, 34 therefore employee exposure to other sources of tobacco smoke was likely to influence the measurements. Assessment of air quality Site visits took place during peak business times, mainly on Wednesday, Thursday, Friday and Saturday evenings between 6 and 10 pm. Three types of monitors (TSI SidePak AM510 Personal Aerosol Monitor, Turnkey OSIRIS Dust and Particle Monitor and GRIMM Dust Monitor 1.105) were used to measure particulate matter of,2.5 mm in diameter (PM 2.5 ), a well-established surrogate for concentrations of SHS To ensure the comparability of readings between the three monitors, four side-by-side measurements were carried out in a fume cupboard with a lit cigarette; each experiment lasted at least 30 min. Monitors were placed in a central area on the premises where staff were working and away from the immediate vicinity of open doors, windows, ventilation ducts or fans, kitchen areas and smokers. Measurements were carried out over a time period of at least 30 min, with readings being recorded once every minute. In contrast to other studies 37,38 where covert measurements were carried out, staff were aware of when and where PM 2.5 measurements were conducted. Alongside the air quality and cotinine measurements, fieldworkers conducted an interviewer-administered survey with customers and staff and therefore announced their visits to the businesses. The recorded data were then downloaded to a personal computer and centrally collated at the TCCC. Assessment of SHS exposure Salivary cotinine, a metabolite of nicotine, was used as an indicator for employee exposure to SHS in the visited venues. Saliva samples were collected using citric-flavoured Salivettes (Sarstedt Ltd, Leicester, UK), which are plastic vials containing a cotton dental roll. Dental rolls were placed underneath employees tongues or between their teeth and cheek until they were fully saturated with saliva. Samples were then mailed to ABS Laboratories Ltd, London, UK for analysis using a previously published rapid gas liquid chromatographic technique. 39 In addition to self-reported smoking status, breath carbon monoxide (CO) monitors ( picoþ Smokerlyzer, Bedfont, UK) were used to assess CO concentrations in employees exhaled breath. Participants were excluded if they reported having smoked within the past year or if their CO concentrations were.10 p.p.m. or COHb.2%. Further exclusion criteria comprised self-reported SHS exposure outside work (Is the employee significantly exposed to SHS other than at work? Yes/No) and salivary cotinine levels.20 ng/ml. Ethics and consent The project methodology was reviewed by the Ethics Committee of Coventry University and approved on

3 CHANGES IN AIR QUALITY AND SECOND-HAND SMOKE EXPOSURE April 2007, reference S21/07. All participants signed informed consent forms prior to taking part in the project and were informed of the purpose, potential risks, rights of withdrawal and intended use of their data. Statistical methods Two types of analyses were carried out for PM 2.5 concentrations: repeated cross-sectional analyses with comparison of the overall mean results and matched analyses of those venues visited at both baseline and follow-up. For salivary cotinine measures, cross-sectional analyses were carried out for saliva samples provided by all employees satisfying the selection criteria. A cohort of employees who had participated pre- and post-legislation was also analysed across time. Results Comparison of monitors Four experiments were carried out to compare the performance of the monitors used in this study. The average concentrations during the four experiments ranged from 42 to 690 mg/m 3, which is comparable with the concentrations found in bars. The SidePak values were corrected using a previously established correction factor of ,38,40 and were used as the gold standard. The GRIMM monitor was found to overestimate the levels by 20% therefore an average correction factor of 0.82 was applied to the GRIMM data. The OSIRIS instrument underestimated the PM 2.5 levels by 70% on an average; a correction factor of 3.37 was applied to all the readings obtained with this monitor. Table 1 Venues at baseline and follow-up by region, location, type and size Visited at baseline (N ¼ 41) Visited at follow-up (N ¼ 43) n % n % Region East Midlands Northeast Southwest West Midlands Location City Centre Town Centre Urban Periurban a Rural Business type Bar Pub Club Bingo hall Betting shop Café Private members club Other Business size Micro (, employees) Small ( employees) Medium ( employees) a Defined as fringe urban setting including commuter villages and Venues Air quality was measured in a total of 49 different venues, 41 venues in June and 43 venues in August Due to the availability of fieldworkers and monitoring equipment, some of the venues were only tested once, either at baseline or at follow-up. The majority of businesses (77.5% at baseline and 64.3% at follow-up) were located in the city or town centres. The types of venues included bars, pubs, nightclubs, bingo halls, betting shops, cafes and private member clubs. All but two businesses were classified as micro or small businesses (i.e.,50 employees). Distributions by region, location, business type and business size are shown in Table 1. dormitory settlements. Indoor particle concentrations Results from the cross-sectional analyses show a large difference in average PM 2.5 levels between the two surveys, with an average PM 2.5 level of 217 mg/m 3 at baseline and an average PM 2.5 level of 11 mg/m 3 at the follow-up survey (Fig. 1). When results were matched on venue only (35 venues), the reduction showed a 96% decline from 212 mg/m 3 at baseline to 11 mg/m 3 at follow-up. When matched on venue, day of the week and time of the day (start time within 1 h), data were available for 20 venues. Results were again very similar to a 95% decline in average

4 424 JOURNAL OF PUBLIC HEALTH PM 2.5 level from 233 mg/m 3 pre-legislation to 14 mg/m 3 post-legislation. See Table 2 for further details. Salivary cotinine Eighty-three employees were tested for salivary cotinine, with 69 samples collected at baseline and 53 samples at follow-up. After removing participants with self-reported external exposure to SHS, CO levels.10 p.p.m. and cotinine levels.20 ng/ml at either phase, the results of 75 individuals from 42 different venues were retained for analysis. The pre- and post-legislation cotinine samples of 39 employees from 29 venues in five regions were available for the matched analysis. Distributions by region, gender and age group are shown in Table 3. As shown in Fig. 2, the average salivary cotinine level was 3.4 ng/ml at baseline and 0.8 ng/ml at follow-up in the repeated cross-sectional analysis, a difference of 2.6 ng/ml. Very similar results were obtained for the matched samples where the average cotinine level was reduced by 75% (2.7 ng/ml), from 3.6 ng/ml at baseline to 0.9 ng/ml at follow-up. Further details are shown in Table 4. Dissemination of results In contrast to other studies in this field, the results of the PM 2.5 and salivary cotinine measurements were fed back to all participating regions and the general public using press releases and interviews within weeks of the measurements. Findings were presented at the annual conference of the National Cancer Research Institute on 1 October 2007 and released to stakeholders and the media on the same day. At Table 2 Means and log-transformed t-test values for PM 2.5 concentrations at baseline and follow-up M GM GSD t (df) Cross-sectional Baseline mg/m Follow-up 11.3 mg/m (82)* Matched by venue Baseline mg/m Follow-up 11.1 mg/m (34)* Matched by venue, date and time Baseline mg/m Follow-up 14.0 mg/m (19)* *P, Table 3 Tested employees by region, gender and age group at baseline and follow-up Tested at baseline (N ¼ 66) Tested at follow-up (N ¼ 48) n % n % Region East Midlands Northeast East of England Southeast Southwest West Midlands Gender Male Female Age group (years) Fig. 1 Mean PM 2.5 concentrations at baseline and follow-up. least 20 different print and online newspapers and journals published news stories about the project, and interviews with representatives of the TCCC and Cancer Research UK were broadcast on two television channels and two radio stations.

5 CHANGES IN AIR QUALITY AND SECOND-HAND SMOKE EXPOSURE 425 Fig. 2 Mean salivary cotinine concentrations at baseline and follow-up. Table 4 Means and log-transformed t-test values for cotinine concentrations at baseline and follow-up Discussion M GM GSD t (df) Cross-sectional Baseline 3.4 ng/ml (87)* Follow-up 0.8 ng/ml Matched samples Baseline 3.6 ng/ml (38)* Follow-up 0.9 ng/ml *P, Main findings of this study We monitored air quality and employee exposure to SHS in businesses in the hospitality and leisure industry shortly before and after the introduction of the English Smoke-free legislation in July Baseline data were collected in June, one month before the legislation came into effect, and follow-up measures were taken in August, one month after implementation. Over this time period, PM 2.5 concentrations in the 35 venues that participated in both surveys reduced by 96%. In addition, salivary cotinine levels were reduced by 75%. Findings were immediately disseminated to the public through press releases and interviews in the regional and national media. What is already known on this topic No occupational exposure limits or indoor air limits are available for SHS or PM 2.5. However, others 37,38,40,41 have interpreted PM 2.5 levels in bars using the US Environmental Protection Agency s (US EPA) outdoor Air Quality Index which rates levels of 150 mg/m 3 as very unhealthy and levels of 250 mg/m 3 as hazardous to health 40,41 Although these limits are for outdoor particulate pollution, and are based on 24 h exposures, they provide some guidance to the potential hazardous nature of the indoor air quality in the venues. Based on US EPA guidelines, the air quality in many venues visited in June would have been considered to be very unhealthy at the time of the measurements, while in some venues the air quality would have been considered to be hazardous to health. In contrast, the concentrations measured in August were much lower and similar to outside ambient air levels in the UK. The findings suggest that the Smoke-free legislation was successfully implemented in the visited venues and most likely helped reduce employee exposure in the workplace to potentially hazardous levels of SHS. It is interesting to note that salivary cotinine levels at follow-up were higher than those measured after the introduction of smokefree laws in Ireland 42 and in Scotland. 37 Although the reasons for these differences are unclear, it should be acknowledged that employees are still exposed to SHS in non-enclosed, yet sheltered outside areas, such as fenced patios with umbrellas. What this study adds The study helped illustrate the positive effects of the English Smoke-free legislation by monitoring and reporting its short-term effects on a selection of hospitality and leisure businesses. Within just two months of the introduction of the legislation, air quality had significantly improved and employee exposure to SHS had substantially decreased in the participating venues. Measurements were collected in 49 businesses and from 75 non-smoking individuals in six of the nine regions in England. To our knowledge, no national air quality and exposure data have been released for England to this date. Quick and positive feedback to the public is essential when introducing public health legislation such as smoke-free laws. The findings were released to stakeholders and the media on 1 October 2007, just three months after the implementation of the Smoke-free legislation. They attracted a lot of attention by national press and were quoted in over 20 print and online media articles. In the future, researchers may wish to further investigate the role of the media and evaluate the impact of such information campaigns on public opinion and behaviours. Limitations of this study Site visits took place in six different regions and covered a variety of locations, types and sizes of venues. However, it cannot be assumed that our sample was representative of

6 426 JOURNAL OF PUBLIC HEALTH businesses in the hospitality and leisure industry nationwide, and results should therefore be generalized with caution. The recruitment procedure did not rule out a self-selection bias. It is conceivable that business owners who were already in favour of the legislation were more likely to participate in the project than owners who were opposed to the law change. Improvements in air quality and employee exposure might therefore have been greater in the visited venues than in other businesses. Measurements were taken in an overt manner and as a result, business owners were aware of the date and time of the visits. It could be argued that business owners knowledge of being tested might have influenced their compliance with the law, especially during the follow-up visits when failure to prevent smoking in enclosed premises had become a legal offence. Being involved in the project might have raised their awareness of the legislation and affected their general attitude towards compliance. Alternatively, one might suspect that business owners had only prevented smoking for the time of the visits when they were otherwise non-compliant. Although this is conceivable, the observed drop in cotinine concentrations suggests that virtually no smoking had taken place for at least two to three days before the site visits. 34 National compliance data furthermore support our findings; out of premises and vehicles inspected during August 2007, 98.7% were compliant in terms of no-smoking and 88.3% of premises and vehicles were compliant in terms of signage. 43 Moreover, timing and duration of the measurements might not have captured typical exposure levels in the visited premises. Fieldworkers were instructed to perform visits during peak business times and to operate particulate monitors over a period of at least 30 min. It is likely that these measurements either slightly over- or underestimated average exposure levels. To account for variations related to the timing of the visits, PM 2.5 concentrations of venues that had been monitored on the same day of the week and at the same time (within 1 h) at baseline and follow-up were analysed separately. The results confirm the observed 95% decline in PM 2.5 concentrations from June to August. Since the project aimed to monitor and report the shortterm impacts of the legislation, findings do not account for seasonal or other variations over time. A separate year-long study with several follow-up visits is currently underway led by the Department of Environmental and Occupational Medicine at the University of Aberdeen and their results will be released in autumn 2008 (Dr Semple S., Personal communication, 2008). Finally, almost 30 fieldworkers from a variety of local services were actively involved in the data collection, a likely source of variation when considering the reliability of the measurements. To ensure that the protocols were followed as closely as possible, we developed a Fieldworker Guide, containing detailed instructions for the planning and preparation of visits, the recruitment of participants, data collection, data handling and despatch (the guide is available as supplementary data at the Journal of Public Health online). We then organized four regional training workshops where fieldworkers learned how to prepare, maintain and use the monitoring and testing equipment. Supplementary data Supplementary data are available at the Journal of Public Health online. Acknowledgments We thank Elspeth Lee and Emma Gilgunn-Jones from Cancer Research UK for their helpful comments and advice throughout the project. We would also like to thank all fieldworkers who collected the measurements and Regional Tobacco Policy Managers for facilitating the data collection, as well as all the businesses and individuals who participated in the monitoring. Funding This work was supported by Cancer Research UK; grant number C24026/A9132. References 1 Great Britain. Health Act 2006 (ch. 28). London: HMSO, 2006: pdf (25 February 2008, date last accessed). 2 Haw SJ, Gruer L, Amos A et al. Legislation on smoking in enclosed public places in Scotland: how will we evaluate the impact? J Public Health 2006;28(1): New York State Department of Health. The Health and Economic Impact of New York s Clean Indoor Air Act. New York, United States, docs/ciaa_impact_report.pdf (25 February 2008, date last accessed). 4 Office of Tobacco Control. Smoke-free Workplaces in Ireland: A One Year Review. Clane, Co Kildare, Ireland, Uploads/1_Year_Report_FA.pdf (25 February 2008, date last accessed). 5 Lund M. Smoke-free Bars and Restaurants in Norway. SIRUS, National Institute for Alcohol and Drug Research. Oslo, Norway, Norway.pdf (25 February 2008, date last accessed). 6 New Zealand Ministry of Health. After the Smoke has Cleared: Evaluation of the Impact of a New Smokefree Law. Wellington, New

7 CHANGES IN AIR QUALITY AND SECOND-HAND SMOKE EXPOSURE 427 Zealand, $File/smokefree-evaluation-report-with-appendices-dec06.pdf (25 February 2008, date last accessed). 7 Haw SJ, Gruer L. Changes in exposure of adult non-smokers to secondhand smoke after implementation of smoke-free legislation in Scotland: national cross sectional survey. Br Med J 2007;335(7619): Semple S, Creely KS, Naji A et al. Secondhand smoke levels in Scottish pubs: the effect of smoke-free legislation. Tob Control 2007;16(2): Allwright S, Paul G, Greiner B et al. Legislation for smoke-free workplaces and health of bar workers in Ireland: before and after study. Br Med J 2005;331(7525): Ellingsen DG, Fladseth G, Daae HL et al. Airborne exposure and biological monitoring of bar and restaurant workers before and after the introduction of a smoking ban. J Environ Monit 2006;8(3): Farrelly MC, Nonnemaker JM, Chou R et al. Changes in hospitality workers exposure to secondhand smoke following the implementation of New York s smoke-free law. Tob Control 2005;14(4): Sargent RP, Shepard RM, Glantz SA. Reduced incidence of admissions for myocardial infarction associated with public smoking ban: before and after study. Br Med J 2004;328(7446): Barone-Adesi F, Vizzini L, Merletti F et al. Short-term effects of Italian smoking regulation on rates of hospital admission for acute myocardial infarction. Eur Heart J 2006;27(20): Eagan TML, Hetland J, Aarǿ LE. Decline in respiratory symptoms in service workers five months after a public smoking ban. Tob Control 2006;15(3): Shields M. Smoking bans: influence on smoking prevalence. Health Rep 2007;18(3): Fong GT, Hyland A, Borland R et al. Reductions in tobacco smoke pollution and increases in support for smoke-free public places following the implementation of comprehensive smokefree workplace legislation in the Republic of Ireland: findings from the ITC Ireland/UK Survey. Tob Control 2006;15(Suppl. 3): Wilson N, Sertsou G, Edwards R et al. A new national smokefree law increased calls to a national quitline. BMC Public Health 2007;7:75, doi: / Biener L, Garrett CA, Skeer M et al. The effects on smokers of Boston s smoke-free bar ordinance: a longitudinal analysis of changes in compliance, patronage, policy support, and smoking at home. J Public Health Manag Pract 2007;13(6): Fernando D, Fowles J, Woodward A et al. Legislation reduces exposure to second-hand tobacco smoke in New Zealand bars by about 90%. Tob Control 2007;16(4): Weber MD, Bagwell DAS, Fielding JE et al. Long term compliance with California s Smoke-Free Workplace Law among bars and restaurants in Los Angeles County. Tob Control 2003;12(3): Borland R, Yong HH, Siahpush M et al. Support for and reported compliance with smoke-free restaurants and bars by smokers in four countries: findings from the International Tobacco Control (ITC) Four Country Survey. Tob Control 2006;15(Suppl. 3): Skeer M, Land ML, Cheng DM et al. Smoking in Boston bars before and after a 100% smoke-free regulation: an assessment of early compliance. J Public Health Manag Pract 2004;10(6): Luk R, Ferrence R, Gmel G. The economic impact of a smoke-free bylaw on restaurant and bar sales in Ottawa, Canada. Addiction 2006;101(5): Alpert HR, Carpenter CM, Travers MJ et al. Environmental and economic evaluation of the Massachusetts Smoke-Free Workplace Law. J Community Health 2007;32(4): Stolzenberg L, D Alessio SJ. Is nonsmoking dangerous to the health of restaurants? The effect of California s indoor smoking ban on restaurant revenues. Eval Rev 2007;31(1): McCaffrey M, Goodman PG, Kelleher K et al. Smoking, occupancy and staffing levels in a selection of Dublin pubs pre and post a national smoking ban, lessons for all. Ir J Med Sci 2006;175(2): Thomson G, Wilson N. One year of smokefree bars and restaurants in New Zealand: impacts and responses. BMC Public Health 2006;6:64, doi: / Department of Health. White Paper: Smoking Kills. London, England, cm41/4177/contents.htm (26 February 2008, date last accessed). 29 Department of Health. Public Health White Paper: Making Healthier Choices Easier. London, England, Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/ DH_ (26 February 2008, date last accessed). 30 Department of Health. Partial Regulatory Impact Assessment- Smokefree Aspects of the Health Bill. London, England, en/publicationsandstatistics/legislation/regulatoryimpactassessment/ DH_ (26 February 2008, date last accessed). 31 Hilton S, Semple S, Miller BG et al. Expectations and changing attitudes of bar workers before and after the implementation of smoke-free legislation in Scotland. BMC Public Health 2007;7:206, doi: / Friis RH, Safer AM. Analysis of responses of Long Beach, California residents to the Smoke-free Bars Law. Public Health 2005;119(12): Haw SJ. Evaluating Smoke-free Legislation: Lessons from Scotland. National Smokefree Conference, Liverpool, Etzel RA. A review of the use of saliva cotinine as a marker of tobacco smoke exposure. Prev Med 1990;19(2): Di Giusto E, Eckhard I. Some properties of saliva cotinine measurements in indicating exposure to tobacco smoking. Am J Public Health 1986;76(10): Benowitz NL. Cotinine as a biomarker of environmental tobacco smoke exposure. Epidemiol Rev 1996;18(2): Semple S, Maccalman L, Naji AA et al. Bar workers exposure to second-hand smoke: the effect of scottish smoke-free legislation on occupational exposure. Ann Occup Hyg 2007;mem Edwards R, Hasselholdt CP, Hargreaves K et al. Levels of second-hand smoke in pubs and bars by deprivation and food serving status: a cross-sectional study from North West England. BMC Public Health 2006;6:42 doi: /

8 428 JOURNAL OF PUBLIC HEALTH 39 Feyerabend C, Russell MA. A rapid gas-liquid chromatographic method for the determination of cotinine and nicotine in biological fluids. J Pharm Pharmacol 1990;42(6): Repace J. Air Pollution in Virginia s Hospitality Industry: Repace Associates Inc., (March 2008, date last accessed). 41 Repace JL, Hyde JN, Brugge D. Air pollution in Boston bars before and after a smoking ban. BMC Public Health 2006;6:266, doi: / Allwright S, Paul G, Greiner B et al. Legislation for smoke-free workplaces and health of bar workers in Ireland: before and after study. Br Med J 2005;331: Local Government Analysis and Research. Smokefree Legislation Compliance Data: Report Covering August 2007 (Period 3), smokefree-report-period-3-v2.pdf (5 March 2008, date last accessed).