Smoking is generally not regulated in restaurants or bars in China, or the restrictions are not fully implemented if there are any, while the related hazard health effects are not recognized by the majority of the Chinese population.
This study aims to assess the excess health risks and mortality attributed to secondhand smoke (SHS) exposure in restaurants and bars for both servers and patrons to provide necessary evidence for advancing tobacco control in this microenvironment.
Two approaches were used for the assessment. One is a continuous approach based on existing field measurements and Repace and Lowrey’s dose-response model, and the other is a categorical approach based on exposure or not and epidemiological studies.
Based on the continuous approach, servers were estimated to have a lifetime excess risk (LER) of lung cancer death (LCD) of 730 to 1,831×10−6 for working five days a week for 45 years in smoking restaurants and 1,862 to 8,136×10−6 in smoking bars, and patrons could have a LER of LCD of 47 to 117×10−6 due to visiting smoking restaurants for an average of 13 minutes a day for 60 years, and 119 to 522×10−6 due to visiting smoking bars. The categorical approach estimated that SHS exposure in restaurants and bars alone caused 84 LCD and 57 ischemic heart disease (IHD) deaths among nonsmoking servers and 1,2419 LCDs and 1,689 IHD deaths among the nonsmoking patron population.
SHS exposure in restaurants and bars alone can impose high lifetime excess risks of lung cancer death and ischemic heart disease deaths to both servers and patrons, and can cause a significant number of deaths each year in China. These health risks and deaths can be prevented by banning smoking in restaurants and bars and effectively implementing these smoking bans.
Citation: Liu R, Jiang Y, Li Q, Hammond SK (2014) An Assessment of Health Risks and Mortality from Exposure to Secondhand Smoke in Chinese Restaurants and Bars. PLoS ONE 9(1): e84811. https://doi.org/10.1371/journal.pone.0084811
Editor: Alexander G. Obukhov, Indiana University School of Medicine, United States of America
Received: July 26, 2013; Accepted: November 26, 2013; Published: January 8, 2014
Copyright: © 2014 Liu et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research project was supported by Dr. William Cahan Distinguished Professor Award to SKH from the Flight Attendants Medical Research Institute. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
More than 600,000 deaths were caused by secondhand smoke exposure (SHSe) in 2004, corresponding to 1.0% of all deaths worldwide in that year . China is the largest consumer of tobacco in the world, with 301 million adults currently smoking (including 53% of men and 2.4% of women)  and 556 million (72%) nonsmoking adults exposed to SHS in 2010 . Gan et al. estimated that 130,000 lung cancer deaths (LCDs) and 169,600 ischemic heart disease (IHD) deaths were attributed to active smoking, and 22,200 LCDs and 33,800 IHD deaths were attributed to SHSe among nonsmoking adults in China in 2002 .
In China, where the tobacco industry is owned by the government , developing and implementing effective policies to reduce active and passive smoking is particularly challenging. China ratified the World Health Organization’s Framework Convention on Tobacco Control (WHO FCTC) in 2006. Health departments of the Chinese government and many health organizations have been working hard to implement the Convention and have made considerable progress. For example, the Beijing government restricted smoking in large restaurants within the city on May 1st 2008; several other big cities, such as Shanghai, Guangzhou, Hangzhou and Yinchuan, have also revised their local legislations to prohibit or restrict smoking in more public places; and the Chinese Ministry of Health revised and tightened the rules of restricting smoking in all indoor public places including restaurants and bars in May 2011; national legislation has been updated to regulate cigarette packaging and labeling and to adjust the consumption tax of tobacco products . However, a huge gap still exists between China’s current state of tobacco control and the FCTC requirements , . As reported by Liu, et al, Beijing Government’s smoking restrictions are poorly implemented and are ineffective in protecting people from the adverse health effects of SHSe .
Another major challenge for tobacco control in China is that SHSe is prevalent in various microenvironments, while the related hazardous health effects are not recognized by nearly 75% of the population . Thus, information about the magnitude of the health risks and disease burden due to SHSe is particularly important for policy makers to plan preventive strategies and for the general population to understand its right of enjoying smoke-free air.
In China, there were about 11 million restaurant or bar employees in 2004 , and 15% of people aged 15 years and older eat out at least every day . However, smoking is generally not regulated in hospitality venues, or the restrictions are not fully implemented if there are any , . To provide evidence necessary for advancing tobacco control in restaurants and bars, this study aims to assess the health risks and mortality of lung cancer and IHD attributed to SHSe in this specific microenvironment for both servers and patrons.
Both a continuous approach based on SHS concentrations and a categorical approach based on exposure or not and epidemiological studies were used for the assessment. Similar approaches have been applied by some other studies –.
Estimate of Life-time Risk of Lung Cancer Death Using a Continuous Approach
Sufficient evidence strongly indicates that SHSe can increase the risk of lung cancer among nonsmokers, and the magnitude of the excess risk is similar among populations with different races/ethnicity, regardless of exposure locations –. Repace and Lowrey  developed a model to predict the risk of lung cancer death (LCD) due to SHSe, based on the mortality data and SHSe data among the U.S. population. The model was validated by predicting epidemiologically derived observational data to within 5%, and has been widely used –. The model estimated that, for the U.S. population, the risk of LCD is 5×10−5 for exposure to 1 mg/day of SHS particulate matter (SHS PM) for one year and the lifetime excess risk (LER) of LCD can be estimated by Equation 1:(1)Where
Daily dose is the exposure dose of SHS PM in mg/day for an individual;Years is the number of years exposed to SHS;CSHS PM is the average concentration of SHS PM in mg/m3 during the period of exposure;BR is the breathing rate in m3/hr;Time is the average hours per day exposed to SHS;f is the adjustment factor, estimated by the number of days per week exposed to SHS divided by 7 days.
Since the long term survival rate of lung cancer is very low , the risk of lung cancer death was assumed to be similar to the risk of lung cancer incidence. Since the excess risk of lung cancer caused by SHSe is similar for different populations, the dose-response relationship indicated by Equation 1 was applied to the Chinese population. Servers’ and patrons’ exposures to SHS PM in Chinese restaurants and bars were estimated by existing field measurements of SHS concentrations using particulate matter as the tracer during peak-patronage times in restaurants and bars in some Chinese cities , –. Table 1 presents summaries of the results reported by these studies or re-analysis of the data from three of these studies , , . Three of these studies used a similar protocol and all adjusted the corresponding outdoor PM2.5 concentrations to estimate SHS PM concentrations , , . The other three studies reported indoor PM2.5 concentrations only without adjusting the corresponding outdoor levels, and thus were not included in this analysis. Smoking restrictions in restaurants and bars were rare in China before 2008 , and the sample sizes of venues restricting or prohibiting smoking in the two existing studies conducted before 2008 ,  were too small to be informative of SHS concentrations in these venues, thus servers’ and patrons’ LER of LCD was estimated based on measurements in venues allowing smoking only. The range of the means of SHS PM concentrations reported by these two studies was used to estimate servers’ and patrons’ exposure to SHS in restaurants and bars before 2008. Measurements conducted in Beijing restaurants and bars in 2008 and 2010 after the implementation of the governmental smoking restriction were used to estimate servers’ and patrons’ potential exposure under poor enforcement.
To estimate the daily dose of SHS PM exposure, servers were assumed to have a moderate activity level with a breathing rate of 1.6 m3/hour, and patrons were assumed to have a light activity level with a breathing rate of 1.0 m3/hour, based on recommendations by U.S. EPA . Servers’ exposure time was assumed to be four hours a day during peak patronage time for five days a week, and a working life of 45 years was also assumed. For patrons, an average adult was reported to spend 13 minutes per day (male 16 minutes, female 9 minutes) in restaurants and bars in 2008 . Based on these assumptions, f equals to 5days/7days for servers and one for patrons.
Estimate of Health Risk by Excess Risk Assessment Method (Exposed/Nonexposed)
This approach uses the population attributable fraction (PAF) to estimate the attributable deaths of lung cancer and ischaemic heart disease due to SHSe in restaurants and bars. PAF is referred as the proportional reduction in disease that would occur if the exposure was reduced to zero. This approach has been commonly used to assess disease burden attributed to SHSe worldwide and in some individual countries , , , . Equations 2 and 3 were used for the assessment.(2)(3)Where
B is the total number of deaths of a disease per year among the whole population;Bns is the number of deaths of a disease per year among nonsmokers;ABsm, is the attributable number of deaths of a disease per year among smokers due to smoking;ABSHS is the attributable number of deaths from a disease per year among nonsmokers due to SHSe;PAFsmk is the PAF of the total deaths of a disease among a population due to active smoking;PAFSHS is the PAF of the deaths of a disease due to SHSe among nonsmokers;psmk is the prevalence of smoking;pSHS is the prevalence of SHSe;RRsmk is the relative risk of a disease due to active smoking;RRSHS is the relative risk of a disease due to SHSe among nonsmokers;AERSHS is the average annual excess risk of death due to SHSe among nonsmokers;Prisk is the population at risk, that is, nonsmokers aged 35 years or older;LERSHS is the life time excess risk of death due to SHSe among nonsmokers;Years is the number of years of exposure during lifetime.
China made up of 82% of the population of WHO WPRO-B region (World Health Organization Western Pacific Regional Office countries with low child and adult mortality rates) in 2004, thus this factor was applied to the WPRO-B subregion data to estimate the age and gender specific population and mortality (B in the equations above) of lung cancer and ischemic heart disease (Table 2). Because of the latency of developing lung cancer or ischemic heart disease, the calculations were restricted to adults aged above 30 years.
There were 11,285,000 people employed by restaurants and bars in 2004 ; and gender and age distributions of restaurant and bar employees in urban areas (but not in rural areas) in 2004 could be obtained from the China Labor Statistical Yearbook  (Table 2). For this analysis, the gender and age distributions of all restaurant and bar employees were assumed to be the same as those of urban employees. The mortality rates of lung cancer and ischemic heart disease for this specific population was not available, thus, they were assumed the same as among the general population.
Age and gender specific smoking prevalence was obtained from the 1984, 1996 and 2002 national surveys of Chinese adult smoking behaviors – (Table 2). Detailed information on SHSe was available from the 1996 survey only; since the smoking prevalence of Chinese male adults did not change much from 1984 to 2002, SHSe prevalence in 1996 was used in this analysis. To account for the latency of lung cancer, the 1984 smoking prevalence and the 1996 SHSe prevalence of the 2004 population were used to calculate the PAF. For IHD death estimates, the average of the smoking prevalence in 1984, 1996 and 2002 was used to account for both the short-term and long-term effect of smoking on the disease. About 90% of restaurants and bars patrons reported that they were exposed to SHS in those places in 2010 , and this percentage was expected to be higher in earlier times, e.g. before 2004. For this analysis, 95% of restaurant and bar patrons as well as workers were assumed to be exposed to SHS in this microenvironment.
The relative risks (RRs) of lung cancer and ischaemic heart disease used to calculate the PAF due to active smoking and SHSe were the same as used by Gan et al . The relative risks of these two diseases for different ages were unavailable, and they were assumed the same across all age groups.
Homes, workplaces, and indoor public places, including restaurants and bars, were assumed to be the major microenvironments where people were exposed to SHS. To estimate the deaths of the two diseases due to SHSe in restaurants and bars, the number of total deaths attributed to SHSe in all microenvironments was multiplied by the percentage of SHS time in this microenvironment, which was estimated by weighting the proportion of the population of interest with SHSe in different microenvironments by their average time spent in each microenvironment. The proportion of males or females exposed to SHS in a specific microenvironment was estimated by multiplying the corresponding general prevalence of SHSe by the proportion of passive smokers who had SHSe in that specific environment as reported in 1996 . The average time spent in each microenvironment was estimated based on time activity surveys conducted in 10 provinces across China in 2008 . Restaurant and bar workers were assumed to spend similar time in homes, workplaces, and indoor public places as the general population, while their time spent in restaurants and bars as patrons was omitted. People were assumed to be exposed to SHS in residences only during waking hours. The percentage of SHS time for each type of microenvironment was calculated as.(4)Where
i is the ith microenvironment, which includes home, workplace, restaurants and bars, and other indoor public places;%SHS_timei is the fraction of time exposed to SHS in the ith microenvironment of the overall time exposed to SHS in all microenvironments by the population;SHS exposure prevalencei is the proportion of the nonsmoking population exposed to SHS in the ith microenvironment;Timei is the time spent in the ith microenvironment.
Sensitivity analysis of percentage of SHSe time in restaurants and bars was conducted by varying the smoking policy in different microenvironments. With this approach, the significance of SHSe in restaurants and bars was examined, in terms of the fraction of overall SHSe time for restaurant and bar servers and patrons.
Estimate of Life-time Risk of Lung Cancer Death Using a Continuous Approach
Based on field measurements in 368 smoking restaurants and 80 smoking bars in five Chinese cities in 2006 and 2007 , , servers were estimated to have an lifetime excess risk (LER) of lung cancer death (LCD) of 730 to1,831×10−6 for working five days a week for 45 years in smoking restaurants and 1,862 to 8,136×10−6 in smoking bars; and patrons were estimated to have LER of LCD of 47 to 117×10−6 for visiting smoking restaurants for an average of 13 minutes a day for 60 years, and 119 to 522×10−6 due to visiting smoking bars (Table 3). After the implementation of the governmental smoking restrictions in Beijing restaurants and bars in 2008, the mean SHS PM concentrations during peak-patronage time ranged from 57 to 84 µg/m3 in restaurants (n = 28) that allow smoking, approximately 78 µg/m3 in designated smoking sections (n = 13), 17 to 26 µg/m3 in designated nonsmoking sections (n = 30), and 36 to70 µg/m3 in nonsmoking restaurants (n = 110)  (Table 1). If future smoking restrictions are as poorly enforced as the 2008 Beijing governmental smoking restrictions, and servers and patrons are exposed to similar SHS concentrations as measured in Beijing restaurants in 2008 and in 2010 for a lifetime, servers’ lifetime excess risk of LCD could range from 370 to 720×10−6 due to work in nominal nonsmoking restaurants, and 175 to 802×10−6 due to work in restaurants nominally restrict smoking, and patrons could have lifetime excess risk ranged from 24 to 46×10−6 due to visit nominal nonsmoking venues only or from 11 to 17×10−6 due to visiting designated nonsmoking sections only for a lifetime.
Estimate of Health Risk by Excess Risk Assessment Method (Exposed/Nonexposed)
The time servers spend exposed to SHS in their work places (restaurants and bars) was estimated to account for 85% of their total SHS exposure time for males and 62% for females (Table 4), which corresponded to 28 LCDs and 19 IHD deaths for males and 56 LCDs and 38 IHD deaths for females in 2004 (Table 5). Patrons’ exposure to SHS due to visiting smoking restaurants and bars accounted for 9.6% of their total exposure time for males and 3.2% for females (Table 4), corresponding to 743 LCDs and 673 IHD deaths among males and 676 LCDs and 1,016 IHD deaths for females in 2004 (Table 5). A total of 3,249 deaths from lung cancer and IHD were attributed to SHSe in restaurants and bars for the servers and patrons together. The LER for nonsmokers were estimated to be 762×10−6 for LCD and 524×10−6 for IHD deaths among servers who work in restaurants and bars for 45 years and 200×10−6 for LCD and 240×10−6 for IHD among patrons who visit restaurants and bars for 60 years.
When 95% of restaurants and bars allow smoking, working in restaurants and bars accounts for almost all of their exposure time for servers living in smoke-free homes, and for more than half of their total exposure time for servers living in smoking homes. For patrons who are exposed to SHS at home, their exposure time in restaurants and bars accounted for less than 5% of their total exposure time, regardless of the smoking policy in their workplaces. However, for patrons who do not have any SHSe at homes or workplaces, their exposure time in restaurants and bars dominate their total exposure time even if only 50% of restaurants and bars allow smoking (Table 6).
Due to limited information available to assess the health risks of other diseases due to SHSe in China, including acute respiratory and sensory irritation, adult asthma initiation, breast cancer among young females, etc., only lung cancer deaths (LCDs) and ischemic heart disease (IHD) deaths were included in this analysis. Therefore, the total health risk and mortality due to SHSe in Chinese restaurants and bars is expected to exceed the estimates in this study.
Several studies have used the dose-response relationship reported by Repace and Lowrey  to estimate lifetime excess risk (LER) of LCD due to SHSe by hospitality workers. Lopez et al.  estimated a LER of LCD of 1,450×10−6 for hospitality workers working for 40 years in Spain; Siegel and Skeer  estimated a LER of 4,100×10−6 for U.S. bar workers for a working life of 45 years, and Liu et al.  estimated a LER of 802 (658–936)×10−6 for Minnesota restaurant and bar workers. This analysis estimated the LER of LCD for servers of Chinese restaurants and bars for the first time. The estimates for restaurants servers (a LER of 730–1,831×10−6 for servers working for 45 years in smoking restaurants and 1,862–8,136×10−6 in smoking bars) were similar to the estimates for Spain and U.S. hospitality workers ,  and higher than that for Minnesota restaurants and bars. All these estimated LCD risks for servers are much higher than the de manifestis risk of 3×10−4, a risk of obvious or evident concern, and one that U.S. public agencies will almost always regulate to control, or mitigate when recognized . Most estimated LCD risks are also higher than the significant risk of 1×10−3 defined by the U.S. Occupational Safety & Health Administration (OSHA).
Few studies in literature have assessed health risks for patrons. Liu et al.  and this study are the first to do so. Liu et al.  estimated a LER of LCD of 80 (66–95)×10−6 for Minnesota patrons visiting smoking restaurants and bars or sections only and 18 (13–23)×10−6 for those visiting designated nonsmoking sections of restaurants and bars only for 1.4 hours/week for 60 years; this analysis estimated a LER of 47 to117×10−6 due to visiting smoking restaurants for an average of 13 minutes a day (1.5 hours/week) for 60 years, and 119 to 522×10−6 due to visiting smoking bars. Analyses of the attendant LCD risk due to SHSe in Minnesota restaurants and bars were based on measurements of SHS concentrations in a representative sample, with restaurants and bars combined and exposure from smoking venues and designated smoking sections combined, while analyses in this study were based on SHS measurements in convenience samples of smoking venues. These estimated lung cancer death risks are all much higher than the acceptable risk or de minimis risk of 1×10−6 for large populations .
Based on the attributable risk assessment method, about 29,000 LCDs and 38,900 IHD deaths were attributable to SHS in 2004; about 30% and 15%, respectively, higher than estimates by Gan et al. , which were 22,200 LCDs and 33,800 IHD deaths. The reasons may be that the 2004 population of WHO WPRO-B was 1% higher than the 2002 population, and the total LCDs and IHD deaths in 2004 were 18% and 4%, respectively, higher than in 2002; furthermore, the population was divided into more subgroups in the analysis by Gan et al. than in this analysis, which may also lead to some discrepancy between the two analyses.
Jamrozik  estimated a total of 54 deaths of lung cancer, IHD and stroke among restaurant and bar workers due to their SHSe at work in the United Kingdom; Liu et al.  estimated a total of 3,200 deaths of lung cancer and ischaemic heart disease among hospitality workers and patrons due to SHSe in restaurants and bars in the U.S., based on the 1992–1994 U.S. National Human Activity Pattern Survey  that SHSe accounted about 9% of total SHSe in the U.S.; this analysis estimated a total of 3,249 deaths of the two diseases among hospitality servers and patrons caused by SHSe in restaurants and bars, accounting for about 5% of the total LCDs and IHD deaths attributed to SHS for the whole nonsmoking population. The total number of excess deaths of LCD and IHD caused by SHSe in restaurants and bars alone in China (3,249) is comparable to the total deaths from all sexually transmitted diseases, excluding HIV, among those aged ≥30 years, or to the total deaths from skin diseases (3,310 and 3,586, respectively, in the entire WHO WPRO-B region, with 82% estimated to be in China). While reducing the disease burden of the later two types of diseases is very expensive and challenging, eliminating SHSe in restaurants and bars, and thus preventing a similar number of deaths, is much more cost effective and practical.
About 20% of nonsmoking males are exposed to SHS at home and 21% in workplaces; about 47% of nonsmoking females are exposed to SHS at homes and 11% in workplaces in China. This leaves 63% of nonsmoking males and 47% of nonsmoking females who are not exposed to SHS in either their homes or workplaces. For this population, their time exposed to SHS in restaurants and bars accounts for more than 66% of their total exposure time, if 80% of the restaurants and bars allow smoking, which is true in most Chinese cities before 2011. Restaurants and bars are the most significant source of SHSe for more than half of the nonsmoking adult population. If smoking regulations are not fully implemented as happened with the 2008 Beijing governmental smoking restrictions (after which implementation, smoking was still observed in about half of nonsmoking venues or sections ), the time spent in restaurants and bars still dominates the total SHS exposure time for patrons who live in smoke-free homes and work in smoke-free environments, and for nonsmoking servers who live in smoke-free homes. Thus, passing smoke-free legislation is not enough, and effective implementation is necessary to protect against SHSe.
The lifetime excess risks and the number of LCD and IHD deaths caused by SHSe in restaurants and bars are probably underestimated for several reasons. Zheng et al.  reported that restaurant workers in Shanghai, China were exposed to SHS during working hours for 24.2 hours on average per week, while this analysis only assumed 20 hour/week of SHSe for workers when using the continuous dose-response approach. When using the attributable risk approach, exposure intensity (indicated by SHS concentrations) in different microenvironments was not weighted, while SHS concentrations are usually higher in restaurants and bars than in homes and most other work places: in the U.S. before 1999, mean airborne nicotine concentrations ranged from 2 to 6 µg/m3 in smoking offices, from 1 to 6 µg/m3 in smoking workplaces of blue-collar workers, from 1 to 3 µg/m3 in homes of smokers, and from 3 to 8 µg/m3 in restaurants , which have been considerably higher concentrations. In public places in some rural and urban areas in China, the medians of one-week time-weighted average concentrations were reported to range from 0.1 to 0.7 µg/m3 in schools, hospitals, governmental buildings, and transportations, and was 2.2 µg/m3 in restaurants ; a median of week-long nicotine concentrations of 1.2 µg/m3 in 14 office buildings in China was also reported ; the median of week-long nicotine concentrations in some Chinese homes was found to be about 0.2 µg/m3 . If these median nicotine concentrations in different environments in China reported in previous studies were used to weight the exposure time, the number of LCD and IHD deaths caused by SHSe in restaurants and bars would be 40% higher for servers and three times higher for patrons.
It should be noted that results presented in this paper should be interpreted with caution. Although the models used in this paper have been used by many other researchers, no validation exists for the Chinese worker and patron population in the hospitality industry; and the Repace-Lowrey dose-response relationship and the inhalation rates used in the analysis are based on U.S. data. However, the authors believe that the most relevant approaches and data are used in this analysis, and no efforts have been made to exaggerate the toll of diseases due to SHSe in Chinese restaurants and bars.
SHS exposure in restaurants and bars alone can impose high lifetime excess risks of lung cancer death and ischaemic heart disease deaths to both servers and patrons, and can cause a significant number of deaths each year in China. These health risks and deaths can be prevented by banning smoking in restaurants and bars and effectively implementing these smoking bans.
Conceived and designed the experiments: RL SKH. Performed the experiments: RL. Analyzed the data: RL. Contributed reagents/materials/analysis tools: RL QL YJ. Wrote the paper: RL SKH. Provided data: QL YJ. Data acquisition: QL YJ. Provided critical comments: QL YJ.
- 1. Oberg M, Jaakkola MS, Woodward A, Peruga A, Pruess-Ustuen A (2011) Worldwide burden of disease from exposure to second-hand smoke: a retrospective analysis of data from 192 countries. Lancet 377: 139–146.
- 2. Li Q, Hsia J, Yang G (2011) Prevalence of Smoking in China in 2010. New England Journal of Medicine 364: 2469–2470.
- 3. Xiao L, Yang Y, Li QA, Wang CX, Yang GH (2010) Population-Based Survey of Secondhand Smoke Exposure in China. Biomedical and Environmental Sciences 23: 430–436.
- 4. Gan Q, Smith KR, Hammond SK, Hu TW (2007) Disease burden of adult lung cancer and ischaemic heart disease from passive tobacco smoking in China. Tobacco Control 16: 417–422.
- 5. Chinese CDC (2011) Tobacco Control in China 2011 Report. Beijing: National Tobacco Control Office, Chinese CDC.
- 6. Yang G, Hu A, editors (2010) Tobacco Control and the Future in China. Beijing: Economic Daily Press.
- 7. Lv J, Su M, Hong ZH, Zhang T, Huang XM, et al. (2011) Implementation of the WHO Framework Convention on Tobacco Control in mainland China. Tobacco Control 20: 309–314.
- 8. Liu R, Jiang Y, Travers MJ, Li Q, Hammond SK (2013) Evaluating the efficacy of different smoking policies in restaurants and bars in Beijing, China: A four-year follow-up study. Int J Hyg Environ Health.
- 9. National Bureau of Statistics of China (2005) Bulletin of the First National Economy Census Key Findings (No.3).
- 10. Ma G, Hu X, Luan D, Li Y, Zhai F, et al. (2005) The Eating Practice of Peole in China (in Chinese). Acta Nutrimenta Sinica 27: 272–275.
- 11. Yang Y, Jiang Y, Wu X, Feng G (2008) Analysis of Policies on Banning Smoking in Public Places and Their Implementation in China. Chinese Journal of Health Education 24 657: 660.
- 12. Liu R, Bohac DL, Gundel LA, Hewett MJ, Apte MG, et al.. (2013) Assessment of risk for asthma initiation and cancer and heart disease deaths among patrons and servers due to secondhand smoke exposure in restaurants and bars. Tobacco Control: Published Online First.
- 13. El-Hougeiri N, El Fadel M (2002) Risk assessment of occupational exposure to environmental tobacco smoke emissions. Indoor and Built Environment 11: 83–94.
- 14. Hedley AJ, McGhee SM, Repace JL, Wong LC, Yu MYS, et al. (2006) Risks for heart disease and lung cancer from passive smoking by workers in the catering industry. Toxicological Sciences 90: 539–548.
- 15. Lopez MJ, Nebot M, Juarez O, Ariza C, Salles J, et al. (2006) Estimation of the excess of lung cancer mortality risk associated to environmental tobacco smoke exposure of hospitality workers. Medicina Clinica 126: 13–14.
- 16. Siegel M, Skeer M (2003) Exposure to secondhand smoke and excess lung cancer mortality risk among workers in the “5 B’s”: bars, bowling alleys, billiard halls, betting establishments, and bingo parlours. Tobacco Control 12: 333–338.
- 17. Steenland K (1999) Risk assessment for heart disease and workplace ETS exposure among nonsmokers. Environmental Health Perspectives 107: 859–863.
- 18. USDHHS (2006) U.S.Department of Health and Human Services. The Health Consequences of Involuntary Exposure to Tobacco Smoke: A Report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, coordinating Center for Health Promotion, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health.
- 19. IARC (2004) Tobacco Smoke and Involuntary Smoking (IARC Monograph 83). Lyon, France.
- 20. Cal/EPA (2005) Proposed Identification of Environmental Tobacco Smoke as a Toxic Air Contaminant. Sacramento, CA: California Environmental Protection Agency Air Resources Board and Office of Environmental Health Hazard Assessment.
- 21. Repace JL, Lowrey AH (1985) A Quantitative Estimate of Nonsmokers Lung-Cancer Risk from Passive Smoking. Environment International 11: 3–22.
- 22. Sugimura H, Yang P (2006) Long-term survivorship in lung cancer - A review. Chest 129: 1088–1097.
- 23. Cui X, Li Q, Niu P, Tian L, Gao A, et al. (2009) Air monitoring survey of PM2.5 in different places in Beijing. Journal of Cardiovascular & Pulmonary Diseases (in Chinese) 28: 50–52.
- 24. Kang J, Lin X, Yang Y, Nan Y, Li Z, et al. (2007) A Study on the level of tobacco-generated smoke in several restaurants and bars in Beijing, China. Chinese Journal of Epidemiology 28: 738–741.
- 25. Lee J, Lim S, Lee K, Guo XB, Kamath R, et al. (2010) Secondhand smoke exposures in indoor public places in seven Asian countries. International Journal of Hygiene and Environmental Health 213: 348–351.
- 26. Liu H, Lin Y, Qi F (2011) Tobacco Control Efforts and Secondhand Smoke Concentrations in Some Restaurants in Qingdao City, China in 2008. Preventive Medicine Tribune (in Chinese) 17: 210–212.
- 27. Liu RL, Yang Y, Travers MJ, Fong GT, O’Connor RJ, et al. (2011) A cross-sectional study on levels of secondhand smoke in restaurants and bars in five cities in China. Tobacco Control 20: 397–402.
- 28. U.S EPA (1997) Exposure Factors Handbook (Final Report). Washington, DC: U.S. Environmental Protection Agency. EPA/600/P-95/002F a–c EPA/600/P-95/002F a–c.
- 29. National Bureau of Statistics, Ministry of Labour and Social Security of China (2005) China Labour Statistical Yearbook 2005.
- 30. Zheng PP, Fu H, Li GY (2009) Smoke-free restaurants in Shanghai: Should it be mandatory and is it acceptable? Health Policy 89: 216–224.
- 31. National Bureau of Statistics of China (2008) Time Activity Patterns of Chinese People.
- 32. CDC (2008) Smoking-Attributable Mortality, Years of Potential Life Lost, and Productivity Losses – United States, 2000–2004. MMWR 57 1226–1228.
- 33. Jamrozik K (2005) Estimate of deaths attributable to passive smoking among UK adults: database analysis. British Medical Journal 330: 812–815.
- 34. National Bureau of Statistics of China (2009) Bulletin of the Third National Economy Census Key Findings (No.3).
- 35. Chen M (1997) The 1996 National Prevalence Survey of Smoking Patterns. Beijing: Chinese Academy of Preventive Medicine.
- 36. Weng X, Hong Z, Chen D, Chen B, Tian B (1986) A National Survey of Smoking Behaviors of 500 000 Adullts in China, 1984. Journal of Cardiovascular and Pulmonary Diseases (in Chinese) 02: 44–48.
- 37. Yang GH, Fan LX, Tan J, Qi GM, Zhang YF, et al. (1999) Smoking in China - Findings of the 1996 National Prevalence Survey. Jama-Journal of the American Medical Association 282: 1247–1253.
- 38. Yang GH, Ma JM, Liu N, Zhou L (2005) Smoking and Passive Smoking in China, 2002. Chinese Journal of Epidemilology 2: 77–83.
- 39. Liu R, Jiang Y, Travers MJ, Li Q, Hammond SK (2013) Evaluating the efficacy of different smoking policies in restaurants and bars in Beijing, China: A four-year follow-up study. International Journal of Hygiene and Environmental Health Accepted.
- 40. Crouch T, Milson R, Klema ED (1987) Cancer risk management: A review of 132 regulatory decisions. Environmental Science and Technology 21: 415–420.
- 41. Klepeis NE, Tsang AM, Behar JV (1996) Analysis of the National Human Activity Pattern Survey (NHAPS) respondents from a standpoint of exposure assessment. Washington, DC.
- 42. Hammond SK (1999) Exposure of US workers to environmental tobacco smoke. Environmental Health Perspectives 107: 329–340.
- 43. Stillman F, Navas-Acien A, Ma JM, Ma SJ, Avila-Tang E, et al. (2007) Second-hand tobacco smoke in public places in urban and rural China. Tobacco Control 16: 229–234.
- 44. Gan Q, Hammond SK, Jiang Y, Yang Y, Hu TW (2008) Effectiveness of a smoke-free policy in lowering secondhand smoke concentrations in offices in China. Journal of Occupational and Environmental Medicine 50: 570–575.
- 45. Hammond SK, Tian L, Li C (1999) Environmental tobacco smoke exposure in China; Edinburgh, Scotland. The International Academy of Indoor Air Sciences. 973–974.
- 46. Repace JL, Lowrey AH (1985) An indoor air quality standard for ambient tobacco smoke based on carcinogenic risk. N Y State J Med 85: 381–383.