Who smokes in smoke-free public places in China? Findings from a 21 city survey

Abstract

Efforts toward controlling secondhand smoke in public places have been made throughout China. However, in contrast to the western world, significant challenges remain for effectively implementing smoke-free regulations. This study explores individual and regional factors which influence smoking in smoke-free public places. Participants included 16 866 urban residents, who were identified through multi-stage sampling conducted in 21 Chinese cities. The reported smoking prevalence in smoke-free public places was 41.2%. Of those who smoked in smoke-free public places, 45.9% had been advised to stop smoking. Participants stated that no-smoking warnings/signs with ‘please’ in the statement had a better likelihood of gaining compliance and preventing smoking in public spaces. Multilevel logistic regression analysis showed that ethnicity, education, occupation, type of smoking, age of smoking initiation, smoking situation, stress, household smoking restrictions and city population were all associated with smoking in smoke-free public places. Interestingly local smoke-free regulations were not associated with smoking in public places. The findings underscore that efforts to restrict smoking in public places in China should emphasize strong enforcement, while simultaneously raising public awareness of the perils of second hand smoke.

Introduction

Globally, the tobacco-smoking pandemic accounts for approximately 5.4 million deaths annually including the deaths of more than 600 000 nonsmokers [1]. The International Labor Organization (ILO) estimates that at least 200 000 workers die every year from exposure to secondhand tobacco smoke (SHS) [2]. China leads the world in tobacco consumption, and approximately one million Chinese die annually from tobacco-related diseases [3]. High smoking prevalence in public spaces contributes to the risks of SHS for many Chinese citizens, and smoking routinely occurs in most Chinese restaurants, schools, hospitals, government buildings and train stations [4, 5].

Many studies have shown that public smoking bans are an effective way to reduce exposure to SHS [6, 7] and many countries have approved legislation for smoke-free public places [7, 8]. More than 739 million people worldwide, an increase of more than 385 million since 2008, are now protected by comprehensive, national smoke-free laws [9]. In China some moves to combat SHS exposure were made in the 1990s, particularly in schools and on public transportation [10], but it was not until the Chinese National People’s Congress ratified the World Health Organization Framework Convention on Tobacco Control (WHO FCTC) in 2005, that regulating SHS became a more formal part of tobacco control policy [11]. To meet this objective, efforts were made to expand the number of smoke-free places throughout China. Although China still lacks a comprehensive smoke-free law, several national laws and policies regulate smoking in public places. On May 1, 2011, the Chinese Ministry of Health released the ‘Guidelines on the Regulatory Measures of the Sanitary Administration in Public Places,’ an action which strengthened control measures on SHS in public places [12, 13]. More recently, the Ministries of Health and Education have developed guidelines for making schools and hospitals smoke-free. While the tobacco control measures were intended to cover the entire nation, their impact has been modest. Unmet public expectations have motivated many local governments to initiate their own smoke-free policies or regulations in public places [14]. Consequently, by the end of 2010 nearly half of Chinese mid-size and large cities had implemented smoking bans [12, 13]. Currently some new advances are taking place. Harbin, Shenzhen and Beijing’s smoke-free regulations (SFRs) in public places were implemented respectively in 2012, 2014 and 2015, which were thought to be close to the FCTC requirements, and the best among all local SFRs [15, 16], A national comprehensive smoke-free law is currently under process and negotiation.

However, in contrast to the western world, where increased regulation has led to the stigmatization and ‘denormalisation’ of smoking [17], in China the implementation of smoking bans has been challenged or largely ignored [12, 13]. Consequently due to low compliance with the existing smoke-free regulations, exposure to SHS, with the partial exception of hospitals and schools, is still common in many public places [18, 19]. While a number of studies have determined the extent of exposure to, and who is most affected by, SHS [20, 21], there has been a scarcity of work on the enforcement of smoking ordinances at the local level and how such ordinances have been responded to by local populations. A number of studies have suggested that support for smoke-free policies is strong especially amongst higher educated people, non-smokers [22] and those aware of the harms of SHS [18, 15], yet there has been little attention to who actually smokes in smoke-free public places. This is surprising since the operationalization of SFRs should be based on a clearer knowledge of who is ignoring local smoke-free regulations and why. Better knowledge of this issue is especially important in China where SHS initiatives have largely been city-based ‘bottom-up’ approaches leading to considerable geographic variation in the strength and degree of enforcement of SFRs [14].

The aim of this study, therefore, is to focus on how smokers have responded to the presence of smoke-free regulations in different urban environments in China. The study addresses three questions: (i) who is least likely to abide by smoke-free regulations? (ii) to what extent do city-level contextual factors, including environmental restrictions, also influence smoking prevalence in public places (SPP)? and; (iii) are household and workplace smoking restrictions associated with higher or lower levels of smoking in public places?

With respect to the first question, we seek to understand who is least likely to abide by smoke-free regulations and what causes individuals to ignore smoke-free regulations. Certain groups of individuals, such as those who are most stressed [23, 24], or the most nicotine dependent, may continue to smoke regardless of the presence or absence of SFRs. Smoking in smoke-free public places will also be influenced by environmental factors, however these have seldom been considered [5]. Key among these will be the extent to which anti-smoking norms are more prevalent in some urban environments (e.g. more affluent cities) than others and the degree to which local SFR’s exist and, more importantly, are enforced [25]. While China and some of its cities and provinces have attempted to restrict smoking, smoking in smoke-free public places has neither come under strong societal condemnation nor faced strong penalties, so smoking continues to be a common phenomenon in these venues. Thirdly, we seek to determine whether smoking restrictions in households and workplaces are associated with smoking in public places. A number of studies have suggested that workplace restrictions on smoking are connected to restrictions at home [26. 27], so the question arises whether both types restrictions has led to a displacement of smokers into smoke-free public places.

Methods

Study area and participants

This study employed a cross-sectional multistage sampling design. In Stage 1, 21 cities were selected from across China and differentiated by regional location. Stages 2 and 3 involved the selection of two residential districts within each city and the random selection of four communities within each residential district. In Stage 4, a family household registration (‘hukou’) list was used to randomly sample households in each community. Individuals aged 15 years and older, who had lived in their home for at least a year, were identified within each household. Finally, one participant was randomly selected from each family with eligibility determined by birthdate being closest to the contact date.

Data collection

A structured self-administered questionnaire was used to collect data. A face-to-face interview was scheduled with selected individuals who agreed to participate in the study. The survey was administered privately to participants in their home or at a designated meeting place, such as a backyard or community park. Interviews were conducted on Saturdays, Sundays, during the evening, or at other times when the participants were available. Participants were asked to fill out a survey questionnaire of approximately 30 min duration, following instruction from an interviewer. Participants were requested to resolve any omissions, as appropriate, and were given a small token of appreciation following questionnaire completion.

The same research protocol was utilized across all 21 cities to ensure homogeneity of interview and data collection methods. The data were collected between March and July in 2011.The study was approved by the Ethics Committee at the Medical Center, Zhejiang University, and verbal consent was obtained from all participants prior to data collection. The study methods have been extensively employed in smoking research in China, and they possess acceptable validity [28, 29].

Measures

Individual level variables

Sociodemographics, smoking status and mental stress

Individual level characteristics, namely age, gender, ethnicity, educational level, occupation and income demographics, which have been associated with smoking prevalence [28], were collected for this study. In addition measures of smoking status, smoking history, smoking situation and mental stress were also included and identified through self-report. Smoking status was defined in terms of whether people were occasional or daily smokers. Smoking history was measured in terms of the age at which respondents first started smoking, while smoking situation recorded whether people tended to smoke with others or smoke alone. For these measures, we employed standard methods used in previous work [29, 30]. Mental stress was measured using the Chinese Perceived Stress Scale (CPSS) [31, 32]. This scale is comprised of 14 items that addressed each individual’s self-perceptions of stress over the month prior to the survey. Items were rated on a 5-point Likert scale and ranged from 0 (never) to 4 (very often). Item scores were summed to yield a total, with higher scores indicating higher perceived levels of stress. In accord with previous research, severe stress was operationalized with a score ≥25.

Smoking prevalence in public places

SPP was measured by several questions. First, participants reported whether they had been to public places, including indoor shops, hospitals, and restaurants in their city during the last six months (yes/no). Next they were asked whether they had observed any no smoking signs in these public places (yes/no). Participants who answered ‘yes’ to these two questions, and who indicated they smoked in such places, were defined as SPP respondents. Those responding ‘no’ to any one of these questions were defined as non-SPP respondents.

Environmental smoking restrictions

Two measures of home and workplace restrictions on smoking were also included. Respondents were asked whether they faced any restrictions and the extent of these. In both cases the options were coded dichotomously as 1 = no restrictions or partial restrictions and 2 = where a total smoking ban occurred [5].

Contextual city-level variables

The extent to which people smoke in smoke-free places will also reflect city-level characteristics. Four measures were used; the presence of municipal smoke-free regulations, GDP per capita, city size, and the extent to which smoking was restricted in households and workplaces. The presence or absence of smoke-free regulations was obtained from the SFR press release for each of the 21 cities in the sample [33].

The impact of SFRs on smoking in smoke-free public places was measured as follows. First, participants were asked whether they had been advised to stop smoking in these venues and what their response had been. The former was measured by the question, ‘did you receive advice to stop smoking at these venues?’ Response options were yes/no. The latter was assessed with the question; ‘if you were advised to stop smoking, what was your response?’ Response options, in this case, were grouped into three categories: those who immediately stopped smoking, those who stopped after smoking some more, and those who stopped after finishing the cigarette. Participants were also asked to choose the top three warning signs that they believed would be useful to promote adherence to smoking restrictions in public places. Some warning signs were collected from these cities, while others were created for the purpose of this study. Overall eight different warning signs were presented to respondents (see Table I).

Table I.

No smoking warnings and response

Group	Smokers
	N	Weighted %	Rank
1. Please do not smoke	3043	49.2	1
2. Please do not smoke to keep the air clean	2642	45.4	3
3. Do not smoke because secondhand smoke endangers health	2541	45.8	2
4. No smoking	2210	41.2	5
5. Do not cause smoke pollution	1099	19.8	8
6. Resist smoke to improve your quality of life	1563	29.3	6
7. Breathing clean air is everyone's right	1388	24.5	7
8. Keep the air clean to protect women and children's health	2318	43.1	4

In addition to the presence of local SFRs, GDP per capita (<40 000, 40 000–49 999, 50 000+ yuan) was used as an indicator of urban wealth. Given that higher income Chinese persons are less likely to smoke [34] then it might be expected that SPP levels would be less in more affluent cities. The third contextual variable, city size (<5 million, 5.0–9.9 million and 10 million and over), was included because larger cities, in addition to being more affluent with changed social norms, usually have more financial resources and technology, and higher social service levels. Consequently they are more likely to have implemented more comprehensive SFRs. Information on GDP per capita and city size was obtained from the National Bureau of Statistics [35]. Finally, we also constructed contextual variables pertaining to environmental smoking restrictions in the home and workplace from the aggregation of individual responses. More specifically for the household variable in each city we grouped the proportion of households facing total or partial smoking restrictions into three categories (<25%, 25–34.9% and 35% and over), while a similar categorization applied to workplace restrictions (<30%, 30–39.9% and 40% and over).

Data analysis

All data were entered into a database using Microsoft Excel. The dataset was then imported into SAS version 9.3 for statistical analyses. Descriptive statistics were calculated for smoking prevalence. χ² analyses were conducted using the SAS 9.3 survey procedure when determining individual and regional-variable differences in SPP. Associations were confirmed through the application of a multilevel logistic regression model using the SAS NLMIXED procedure [36, 37].

Two models were utilized in the multilevel analysis. First the Null Model, a two-level model with random intercepts, was developed. In this base model, all demographic and regional level variables were entered as fixed main effects to form the full model to evaluate the separate impact of all individual-level and regional-level variables on SPP. For this analysis, the dependent variable, SPP, was operationalized as a binary response (1= no SPP and 2 = SPP). The independent variables in this analysis were those emerging as statistically significant (P < 0.1) in the χ² tests. These results are presented in Table II. Model fitting was assessed by the likelihood of a change in the −2log. Significance of the random parameter variance estimates was assessed using the Wald joint χ² test statistic [36, 38].

Table II.

Smoking prevalence in smoke free public places

Group	N (% of sample)	Prevalence (%) χ2 P-value
Individual level
Age (years)		6.29 0.1779
16–25	798 (16.2)	45.6
25–34	1192 (20.9)	45.7
35–44	1213 (21.0)	37.8
45–54	997 (19.0)	40.8
55+	828 (23.0)	41.6
Gender		0.72 0.3978
Male	4562 (90.1)	41.0
Female	466 (9.9)	52.6
Ethnicity		9.53 0.0020
Han	4849 (97.5)	41.7
Other	179 (2.5)	60.5
Education		19.25 0.0002
Elementary school or less	1208 (24.5)	65.6
Junior high school	2635 (50.1)	31.0
High school	626 (14.2)	37.6
Junior college or college	559 (11.2)	43.8
Occupation		36.16 <0.0001
Managers and clerks	925 (16.2)	79.0
Professionals	657 (11.5)	23.4
Commerce and service	1574 (30.7)	26.4
Operations	650 (11.5)	53.7
Students	285 (5.3)	44.6
Retired	363 (12.2)	38.0
Other	574 (12.5)	42.6
Income/person/year Yuan		11.20 0.0037
<10 000	2248 (43.7)	53.1
10 000–19 999	1771 (33.0)	29.5
20000 and over	1009 (23.3)	39.5
Types of smokers		5.02 0.0250
Occasional smokers	1698 (32.6)	23.8
Daily smokers	3330 (67.4)	51.0
Age of smoking initiation	727 (14.9)	28.10 <0.0001
< 16 years		91.8
16 years and over	4301 (85.1)	33.5
Smoking situation	3406 (64.8)	15.58 <0.0001
Smoke with others	1622 (35.2)	57.2
Smoke alone		14.4
Stress		11.30 0.0008
Low score	3066 (67.5)	32.3
High score	1962 (32.5)	62.6
Restrict smoking in households		8.36 0.0038
None or partial	2643 (53.1)	30.1
Restriction	2385 (46.9)	55.8
Restrict smoking in work places		0.80 0.3710
None or partial	1925 (39.7)	38.9
Restriction	2103 (60.3)	44.3
Regional Level
Local Municipal SFR	1168 (18.9)	3.18 0.0742
Present	3860 (39.3)	54.2
Absent		39.3
GDP per capita (yuan)		0.13 0.9351
<40 000	1880 (13.0)	46.0
40 000–49 999	1906 (45.8)	39.0
50 000 and over	1874 (41.2)	42.2
City Size (millions)		21.77 <0.0001
<5.0	2083 (23.7)	57.5
5–9.9	1922 (34.9)	46.1
10 and over	1023 (41.5)	30.1
Smoking restricted in households		12.16 0.0023
<20%	990 (21.8)	32.8
25%–34.9	2984 (46.3)	38.5
35% and over	1054 (16.9)	54.1
Smoking restricted in work places		2.12 0.3474
<30%	1467 (27.5)	37.0
30–39.9	2661 (55.8)	42.8
40% and over	900 (15.7)	49.0

All analyses were weighted to ensure results were representative of SPP in these cities [39]. Weights included sampling weights, the non-participation weight and post-stratification weights [34]. The final overall weights were computed as the product of the above three weights. χ² analyses were weighted using the overall participant-level weights, and the multilevel analysis was weighted using sampling weight at the city level and subject-level weights with non-participation and post-stratification weights, respectively.

Results

A total of 18 310 individuals were identified as potential subjects for this study, among which 17 424 (95.2%) attended an interview and agreed to complete the questionnaire. Of those completing the questionnaire, 16 866 (96.8%) submitted fully complete and valid questionnaires that were used in the study.

Of those completing the questionnaire, 5660 (32.5%) were current smokers, of which 5447 (95.9%) had been in public places in the last six months. Of these public places, 5028 (92.6%) had implemented a ban on smoking. Of the smokers who had been in public places where smoking was banned, 41.2% had chosen to smoke in spite of the ban. Almost half (45.9%) of these individuals had been asked to stop smoking. In response to the advice to stop smoking, participants reported the following: most (71.9%) immediately stopped, 19.4% stopped after smoking some more and 8.7% stopped after finishing their cigarette. Smokers also ranked the most effective no-smoking/warning signs that they thought would promote greater adherence to no smoking restrictions (Table I). The three highest rated warnings were those that included ‘please’ in the request and offered a particular reason, such as secondhand smoke endangers health or that women and children should be protected from the harms of SHS.

In terms of who was least likely to abide by smoke-free regulations, there were significant differences in SPP prevalence by ethnicity, education, occupation, income, types of smokers, age of smoking initiation, smoking situation, stress level, and by the presence or absence of household smoking restrictions (Table II). Among the contextual characteristics, only city size and household smoking restrictions were significant, although the presence of local SFRs approached significance. Surprisingly urban wealth had no relationship to SPP and this was also true of the city-level contextual variable relating to workplace restrictions.

When all variables were considered in a multi-level logistic regression model similar patterns emerged. Again ethnicity, education, occupation, type of smoking, year of smoking initiation, smoking situation, stress, household smoking restrictions and city size were associated with SPP. Most notably smoking in public places was more likely for those of other ethnicity, for daily smokers, for those who had started smoking at a younger age, for people suffering greater stress and for those who faced home smoking restrictions. On the other hand, SPP was less likely for commercial and service workers, for those attending Junior High School, and among residents of larger cities (Table III).

Table III.

Results of multilevel analyses

Group	Null model	Full model (OR, 95% C.I)
Individual level
Ethnicity
Han		1.00
Other		1.96 (1.37–2.80)**
Education
Elementary school or less		1.00
Junior high school		0.56 (0.35,0.89)**
High school		0.80 (0.41,1.58)
Junior college or college		0.85 (0.36,2.02)
Occupation
Managers and clerks		1.00
Professionals		0.81 (0.25, 2.61)
Commerce and service		0.23 (0.09, 0.59)**
Operations		0.36 (0.13, 1.01)
Students		0.57 (0.23, 1.44)
Retired		0.66 (0.21, 2.07)
Other		0.50 (0.18, 1.38)
Types of smokers
Occasional smokers		1.00
Daily smokers		2.78 (1.02,7.69)*
Age of smoking initiation
<16 years		1.00
16 years and over		0.08 (0.03, 0.28)**
Smoking situation
Smoking with others		1.00
Alone		0.16 (0.06, 0.47)**
Stress
Low score		1.00
High score		2.19 (1.05, 4.55)**
Smoking restricted in households
None or partial		1.00
Restriction		3.41 (1.45, 10.17)**
Regional level
City size (millions)
<5.0		1.00
5.0–9.99		0.47 (0.17, 1.26)
10 and over		0.27 (0.09, 0.77)**
Random parameters between regions	0.2202 (0.0748)**	0.1477 (0.0644) **
Fixed parameters	1.4595 (0.0335)**	1.2858 (0.0383) **

*P < 0.05 **P < 0.01.

Discussion

A significant challenge remains in China to effectively implement smoke-free regulations in public places. This study is the first systematic analysis of SPP within China. In this sample most public places were smoke-free venues, but 41.2% of smokers still smoked in these areas. When participants smoked in public smoke-free places less than half were advised to stop, which indicates that many people are still accepting of public smoking and that the enforcement of public smoking bans has been limited and ineffectual.

The first aim of the study addressed the question of who is most likely to smoke in public smoke-free places? Some striking correlates of SPP were noted in this study, where distinct variations by ethnicity, education and occupation emerged. SPP was more frequent among minority groups than among Han smokers, which is consistent with the findings of Lock et al. [40]. While higher educational groups were less likely to smoke in smoke-free public places, educational differences were significant only in the case of those with a junior high education. There were no distinct trends by occupation except for commercial and service workers who, compared with managers and clerks, were much less likely to smoke in places with smoking bans. This finding stands in contrast to Thomas et al. [41] who reported that higher paid occupational groups were more likely to comply with smoking restrictions. SPP was also higher in daily smokers and among those who had initiated smoking at a younger age. Long-term smokers may have higher nicotine dependence, which contributes to their difficulty in adhering to non-smoking rules in public places [42, 43]. Individuals who smoked with others also had higher SPP prevalence, suggesting that group smoking tends to inhibit compliance to regulations. Finally those who indicated that they suffered from mental stress were more than twice as likely to ignore local smoking bans. Higher mental stress levels make it more difficult for people to control their smoking [23, 44] and to desist even when restrictions are present.

City-level contextual factors also had some influence, above and beyond individual characteristics, on levels of SPP. The only city-level factor to emerge was city size. City-level variations in household smoking restrictions, while significant in the univariate analysis, were not important in the final model. It is plausible that higher levels of urban management and more resources in larger cities leads to stronger tobacco control regulations and greater compliance on the part of local populations. The recent introduction of new tougher anti-smoking regulations in Beijing is one example [45], but it remains to see whether the new laws are more strictly enforced than before. Larger cities, also have more affluent populations, who are less likely to smoke and thus changed social norms may have also been a factor.

In this study it was interesting that local SFRs were not associated with SPP. Further analysis also showed that SFR was not associated with the prevalence of no smoking signs. This suggests that the impact of local SFRs is very limited, which is consistent with other research which suggests that while many low and middle income countries have, as a result of the WHO FCTC, introduced smoking bans, they remain unenforced [46]. In China there are numerous problems in local SFRs; they tend to apply to limited public venues, lack specific enforcement guidelines, and evoke weak penalties that do not serve as a deterrent [12, 15, 47]. It should also be mentioned that most local SFRs included in this study were issued many years ago, and there are limitations in the regulations themselves and their supervision and management. New SFRs in Harbin, Shenzhen and Beijing, for example, implemented in 2012, 2014 and 2015, have made some efforts in both meeting the FCTC requirements and emphasizing supervision, and management.

Not surprisingly, previous studies have found that SFRs have failed to change tobacco related beliefs, awareness, attitudes and behaviors among residents [13]. While public education has the potential to advance the understanding of why smoking should be banned in public places and encourage support for better tobacco control efforts [48], changing longstanding Chinese cultural norms about smoking is complex and challenging. This is particularly true as smoking is connected to work and social interactions and where cigarette gifting is an important form of social capital. Further, prohibiting smoking and complying with smoke-free regulations may mean a loss of face.

With respect to the third aim of the study, the findings demonstrate that restrictions on smoking in households, but not workplaces, were positively associated with SPP. Where home smoking bans occurred smokers were three times more likely to smoke in smoke-free public places. Since many Chinese women oppose smoking in their households [49], such restrictions appear to have resulted in a certain spatial displacement of smoking from private to public places. This stands in contrast to much western research where public smoking bans appear to have increased the adoption of smoke-free homes especially where children are involved [1, 50, 51]. Such findings are of significance to China in that they suggest that home smoking bans are more likely to be enforced in countries which have implemented comprehensive smoke-free legislation.

The study has several limitations. First, the cross-sectional study design precluded causal inference. However, our findings describing the characteristics of SPP may be useful for consideration in tobacco control policy and managing the implementation of smoke free public places. Second, we assessed SPP through self-report. Smoking behavior may be a sensitive issue in some cultural contexts resulting in information bias. Smoking is, however, a normative behavior for adults in China and most participants did not consider the act of smoking in smoke-free public places as shameful. Therefore, the potential information bias was minimized. The third limitation was that SPP was measured by participant recall during the past six month period; therefore, it is possible the study findings were influenced by recall bias. In addition, the published literature in this field is limited and this may have influenced the depth of discussion in this study. All these limitations confirm the need for future research. Finally, the largely quantitative approach of this study could be complemented by a qualitative approach which explores why Chinese smokers continue to ignore smoking bans in different kinds of public places and under what circumstances their behavior is likely to change. The fact that this study found that ‘no smoking’ warnings/signs that included ‘please’ or which had a rational reason as to why people should not smoke in particular venues, suggests that Chinese smokers can respond positively to anti-smoking cues.

Conclusion

A significant challenge remains for effectively implementing smoke-free regulations in public places. This study has addressed a significant gap in the literature and the findings underscore that efforts to restrict smoking in public places in China should emphasize clear, concise well-written policy along with strong and consistent enforcement, while simultaneously raising public awareness of expected social norms and the perils of SHS. Better enforcement is also important given that a number of studies, both in China [34, 52, 53] and elsewhere [54, 55], indicate that smoking bans can be effective in decreasing tobacco consumption and encouraging cessation. Given that there is limited research on the effective implementation of smoke-free policy [56], future research needs not only to consider resistance to smoking bans but also the effectiveness of enforcement in different geographic locales. It could be argued that the pro-smoking culture of China and resistance to tobacco control occurs precisely because pro-smoking norms are supported by a lack of enforcement. Although a number of cities have made notable accomplishments in terms of implementing smoking bans [14] these have been limited in terms of their impact to selected workplaces and public spaces. National, as well as local, leadership is needed to implement and enforce comprehensive smoke-free legislation. Only then will policy changes be more effective in helping change social norms, levels of tobacco consumption and in reducing the adverse effects of SHS in China.

Local collaborators and PI

Angui Medical University (Yang Jinxia), Baotou Medical College (An Jinagang), Chengdu Medical College (Cheng Jian), Chongqing Medical University (Miao Qing), Dalian Medical University (Li Xiaofeng), Guangxi Medical University (Cheng Zhiping), Guiyang Medical College (Yan Zheng), Hainan Medical College (Yang Jianjun), Hebei Medical University (Tan Fengzhu), Huazhong University of Science and Technology (Du Kaiyu), Jilin University (Li Jinhua), Kunming Medical College (Cui Wenlong), Lanzhou University (Bai Yanai), Nanhua University (Long Dingxin), Nanchang University (Zhou Xiaojun), Qinghai University (Mao Huiqing), Shandong University (Li Jie), Shanghai Jiaotong University (Bao Yong), Tian jin Medical University (Ma Jun), Xi An Medical College (Cao Ping), Xinjiang Medical University (Liu Jiwen), Xuzhou Medical College (Lu Zhaojun), Xiamen University (Fang Ya), Zhengzhou University (Fen Qingyun).

Funding

This study was partly funded by the National Nature Science Foundation of China (71473221), Ministry of Health Tobacco Control Project (WJF: 081216), Center for Tobacco Control research, Zhejiang University School of Medicine. The affiliated universities for the ‘Building advocacy capacity for tobacco control in medical universities in China’ (supported by UNION) project, local government and Center for Disease Control and Prevention also partly funded the survey.

Conflict of interest statement

None declared.