Article Text

Download PDFPDF

Smoking reduced in urban restaurants: the effect of Beijing Smoking Control Regulation
  1. Lin Xiao1,
  2. Yuan Jiang1,
  3. Xiurong Liu2,
  4. Yuqin Li2,
  5. Quan Gan3,
  6. Fan Liu1
  1. 1Tobacco Control Office, Chinese Center for Disease Control and Prevention, Beijing, China
  2. 2Health Education Institution, Beijing Centers for Disease Control and Prevention, Beijing, China
  3. 3China Office, The International Union Against Tuberculosis and Lung Disease, Beijing, China
  1. Correspondence to Dr Lin Xiao, Tobacco Control Office, Chinese Center for Disease and Prevention, No 27 Nanwei Road, Xicheng District, Beijing 100050, China; xiaolin201304{at}


Objective To evaluate the effectiveness of Beijing Smoking Control Regulation, occurrence of smoking in restaurants was compared before and after the law took effect.

Methods A cohort study design was used in a randomly selected sample of 176 restaurants in two districts of Beijing. Undercover visits were paid by investigators to the same restaurants at lunch or dinner time 5 months before the law took effect and 1-month after. The occurrence of smoking and presence of no-smoking signs were observed.

Results Much less smoking was observed (14.8%) in restaurants compared to that before the law took effect (40.3%). The drop in smoking occurrence was more evident in open dining areas (from 32.4% to 5.1%) compared to the men's restrooms of the restaurants (23.8% to 18.8%). No intervention from restaurant staff was observed whenever smoking occurred. Posting of no-smoking signage increased considerably after the law came into effect (from 52.6% to 82.4%), but very few no-smoking signs included the symptom hotline number (38.5%) or the amount of penalty (5.6%).

Conclusions The Beijing Smoking Control Regulation achieved one of its intended goals of reducing smoking occurrences in restaurants, but further effort of strengthening implementation is still needed and should focus on boosting compliance with no-smoking sign requirements, reducing smoking in restrooms of the restaurants and mobilising the restaurant staff to intervene in case of violations.

  • Secondhand smoke
  • Public policy
  • Prevention
  • Surveillance and monitoring

Statistics from

Request Permissions

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.


Tobacco smoke contains hundreds of toxic chemicals, among which at least 69 are carcinogenic.1 Scientific studies have shown that secondhand smoke (SHS) exposure can lead to cancer, cardiovascular and respiratory diseases.2–5 It is harmful for non-smokers to be exposed to SHS, regardless of the level exposed and the length of exposure.6 Article 8 of the WHO Framework Convention on Tobacco Control (WHO FCTC) requires that ‘each party adopt and implement effective legislative, executive, administrative and/or other measures providing for protection from exposure to tobacco smoke in indoor workplaces, public transport, indoor public places and, as appropriate, other public places’. In recent years, more and more jurisdictions have passed laws to ban smoking in public places to protect the public from harms of SHS.7 Such 100% smoke-free policies have been shown to be effective in the protection of public health by lowering the incidence of diseases caused by SHS, in particular heart disease.8–11

Since the WHO FCTC became effective in China in 2006, Beijing has implemented a number of smoke-free initiatives in hospitals, schools, taxis and government office buildings. During the Olympic Games in 2008, the Beijing government issued a provisional policy on smoke-free Olympics, which was supported by massive media campaigns.

In November 2014, the Beijing People's Congress passed the Beijing Smoking Control Regulation, effective from 1 June 2015. The regulation bans smoking in all indoor areas of public places, workplaces and public transport, as well as outdoor areas including kindergartens, schools and child and maternal hospital campuses. This complete ban of indoor smoking makes the Beijing law compliant with Article 8 of the WHO FCTC. Violators face a penalty of ¥50–200 (US$8–32) as an individual or ¥2000–10 000 (US$322–1611) as a venue owner or manager. No-smoking signage and the telephone number for complaint hotline are required to be posted in prominent locations inside the smoke-free venues. Venue owners and managers also have the responsibility to discourage people from smoking on their premises, and not provide ashtrays or display tobacco advertising.

To evaluate the effect of the Smoking Control Regulation, we conducted undercover visits to some restaurants in Beijing both before and after the law came into force and made observations on the occurrences of smoking and the presence of no-smoking signage.


Study design and sample

The survey was conducted in District A and B of Beijing and the sample included large restaurants (500 m2 or more, or 250 seats or more), medium-sized restaurants (between 150 and 500 m2, or between 75 and 250 seats), small restaurants (150 m2 or under, or 75 seats or less), fast food restaurants and snack bars. Fast food restaurants refer to restaurants where food is processed centrally, distributed and served on-site. Snack bars mainly serve pastry and snacks. In this study, fast food restaurants and snack bars are grouped together. If the area and the number of seats of a specific restaurant fall into two different categories, the higher category is picked.

Two districts were randomly selected from eight urban districts of Beijing. Four communities (or Jiedao in Chinese) were selected from each district using the probability proportionate to size sampling method; size in the method refers to the number of restaurants in the community. Twenty-five restaurants were selected randomly from each community. The proportion of each restaurant category among the 25 restaurants is consistent with those in the entire district. In total 200 restaurants were selected.

A cohort study design was used to evaluate the changes in smoking occurrences. The first round of undercover visits was made 5 months before the Smoking Control Regulation was implemented and the investigators were able to make their observations in 199 restaurants. The second round of undercover visits was made to the same restaurants 1-month after the Smoking Control Regulation came into force, and this time 23 restaurants were found to be out of business. The sample used for analysis includes the remaining 176 restaurants.

Survey process

All undercover visits were paid either at lunch time (11:30–13:00) or at dinner time (18:00–19:30). Trained investigators dined at the restaurants as customers and stayed for at least 30 min in the open dining area. The investigators (one male and one female as a group) also checked both the men's and women's restrooms at the restaurant. No smoking was found in women's restrooms either before or after the law took effect, and only the findings from the men's restrooms are reported.


The investigators were required to check no-smoking signage both at the entrance and inside the restaurants. Observations on the occurrence of smoking were also made on the basis of the following three parameters in the open dining area and the restrooms: (1) whether someone was found to be smoking; (2) whether there was smell of cigarette smoke; and (3) whether cigarette butts were found. If any of the three parameters were found to be positive, then the restaurant would be identified as ‘smoking occurrence observed’. When someone was found to be smoking inside the restaurant, the investigator also took note of whether there was intervention from the restaurant staff.

Data analysis

EpiData3.1 was used to set up a database for double data entry verification, and SPSS V.22.0 was used to analyse the data. Percentage was used for descriptive statistics. χ2 was used to compare difference between groups at the same time point such as before the Regulation was implemented or 1-month after the Regulation was implemented. McNemar test was used to compare the changes before and after the Regulation came into force. A p value <0.05 was considered statistically significant.


Among the 176 restaurants included in the analysis, 24 (13.6%) were large restaurants, 61 (34.7%) were mid-sized restaurants, 48 (27.3%) were small restaurants and 43 (24.4%) were fast food restaurants or snack bars. See table 1 for details.

Table 1

Characteristics of restaurants observed before and after Beijing Smoking Control Regulation implementation

Presence of no-smoking signage

Compared to before the law was effective, the proportion of restaurants posting no-smoking signage increased considerably across all four groups, up from 52.6% to 82.4% among all (p<0.001) the restaurants. The changes were most significant in small restaurants, up from 41.7% to 83.3% (figure 1). However, only 5.6% of the no-smoking signs displayed the amount of penalty, and only 38.5% of the no-smoking signs had the symptom hotline number printed. The complaint hotline number is required by law to be printed on the no-smoking sign while amount of penalty is not. Posting of signage was lowest among fast food restaurants and snack bars compared to the other three categories of restaurants. There was no significant difference for posting the no-smoking signage (p>0.05) between District A and District B.

Figure 1

No-smoking sign posting status among 176 restaurants before and after Beijing Smoking Control Regulation implementation.

Observed smoking occurrence

The observed smoking occurrence decreased considerably from 40.3% to 14.8% (p<0.001) after the law came into force. The decline was more evident in the open dining areas (from 32.4% to 5.1%) than in the men's restrooms (from 23.8% to 18.8%) (table 2). For all the ‘live’ smoking observed in this study, no restaurant staff was found to intervene.

Table 2

Smoking occurrence among 176 restaurants before and after Beijing Smoking Control Regulation Implementation

The observed smoking occurrence varied with the type of restaurants (figure 2). The lowest smoking occurrence was observed in fast food restaurants and snack bars and the largest decline was observed in large size restaurants. Mid-sized restaurants had the lowest observed smoking occurrence before the law came into force but the highest occurrence after the law came into force.

Figure 2

Observed smoking occurrence in 176 restaurants by scale before and after Beijing Smoking Control Regulation implementation.

Voluntary partial smoking ban before the law became effective does not seem to have an effect on the observed smoking occurrence. The observed smoking occurrence was 48.6% among restaurants that did not ban smoking and 53.8% among restaurants that had a partial ban (table 2). After the law became effective, smoking occurrence dropped across all three groups (p<0.05). Smoking occurrence decreased in District A and District B after the law came into force (p<0.05) and the difference in decrease was not statistically significant between the two districts (p>0.05). It does not matter whether the visit was paid at lunch time or dinner time (p>0.05).


It has been shown before that with proper implementation, comprehensive ban on smoking can reduce the occurrence of smoking in indoor public places and workplaces, including restaurants and bars.12–14 We observed much less smoking and much more no-smoking signages in restaurants in Beijing after the law took effect, the evidence of early success of the implementation of the Beijing Smoking Control Regulation.

According to previous tobacco surveys among adults in China, restaurants have one of the highest smoking occurrences among all public places and workplaces.15 Restaurants are places where people in China socialise during and after meals in the same way people in many other countries socialise in bars. Whether the smoke-free law is well complied with in restaurants they always receive the greatest public attention and are key in building public confidence in the law. The considerable decline of smoking in restaurants at 1-month after the Beijing Smoking Control Regulation took effect shows substantial progress in curbing smoking in public places and protecting non-smokers' health in Beijing.

Our study records initial success in implementing the Smoking Control Regulation in Beijing; improvement is still needed in a few areas to ensure long-term success. First, much less smoking was observed in restaurants compared to that before the law came into effect, but wherever smoking was observed, no restaurant staff intervened. This indicates that restaurant staff may not be fully aware of their responsibility to keep the venue smoke-free or may be reluctant to tell customers to stop smoking to avoid conflict or loss of business. Further education effort targeting the restaurant owners and managers might still be needed. Second, smoking in men's restrooms did not decrease much compared to that before the law came into effect and we still found smoking in 18.8% of men's restrooms. Smoke in restrooms can drift out and into the open dining areas. Targeted education in men's restrooms on the health effects of SHS is probably needed and restrooms should not remain a blind spot in enforcement of law. Third, overall posting of no-smoking signage is high (>80%) and increased considerably compared to that before the law took effect, but few signages bear the information such as the symptom hotline number and the amount of penalty. Although the amount of penalty is not required by law to appear on the no-smoking signs, such information is helpful in deterring smokers from smoking in smoke-free venues. Fourth, we found that mid-sized restaurants had the lowest smoking occurrence before the law came into effect but the highest after the law came into effect, indicating lax of enforcement during the first month of implementation in these specific types of venues and need targeted enforcement later.

Article 8 of the WHO FCTC requires that authorities pass legislations to protect their citizens from exposure to SHS in indoor workplaces, public transport and public places. Since the implementation of the WHO FCTC in China in 2006, 18 cities including Beijing have passed local smoke-free laws. However, it is still far from enough as two-thirds of these cities only have partial smoking bans and all these 18 cities cover only about 10% of the national population. Nationally 72.4% of the adult non-smokers are still regularly exposed to SHS.15 About the same percentage of middle school students (72.9%) are also regularly exposed.16 Annually at least 100 000 people die in China from exposure to SHS.17

To ride on the subnational smoke-free momentum with the most recent success in Beijing, and to address the still serious widespread exposure to SHS in the country, it is time for China to realise its obligation as a party to the WHO FCTC and pass a national smoke-free law to ensure that all its citizens are able to enjoy the benefits of living in a smoke-free environment.


This study randomly selected two districts from the eight urban districts in Beijing. The level of implementation of the law may vary across the districts, therefore the study findings may not apply to all the other six urban districts. There are 10 other suburban districts in Beijing and they were not considered in this study.

Many restaurants in Beijing have private rooms and access to these rooms is restricted to guests of the rooms and restaurant staff. It is possible that there was more smoking in private rooms, but investigators did not have access to these rooms and only made the observations in public open dining areas.

The post survey was conducted in summer. It is possible that smokers are more willing to go outside to smoke during summer time. Another survey in winter time after the implementation of the law could provide more insight into whether the smoke-free momentum generated would be maintained.

What this paper adds

  • This study is the first evaluation of the effect of the Beijing Smoking Control Regulation, which completely bans smoking in enclosed public spaces.

  • This study compares smoking in restaurants before and after the introduction of the Beijing Smoking Control Regulation and shows that the law has led to a sharp decrease of smoking occurrence in restaurants (from 40.3% to 14.8%).

  • The success of the Beijing Smoking Control Regulation, particularly in the restaurants where there used to be heavy smoking, provides strong evidence for other cities in China that such law can succeed in achieving the protection of public health from the harms of secondhand smoke.


The authors would like to acknowledge the local representatives of the Centers for Disease Control and Prevention for their contribution in data collection.



  • Contributors LX conducted literature reviews and wrote the first draft of the manuscript and coordinated inputs from other authors. FL analysed data. YJ, XL, YL and QG were part of the discussion group and revised the subsequent drafts of the paper. All authors contributed to and approved the final version of the manuscript.

  • Funding This study was supported by Campaign for Tobacco Free Kids (CHINA-RIA-01) and the Chinese Center for Disease Control and Prevention.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.