Introduction

Tobacco control policies prohibiting smoking in public places are growing in popularity, particularly since the turn of the millennium. Although smokefree workplace restrictions have generally been well-received, the extension of smokefree workplace policies to bars often faces resistance, perhaps because of deeply established traditions of smoking in such settings. By strengthening the smokefree workplace law (AB13) in 1998, California became one of the first jurisdictions in the world to apply smokefree workplace protection to bar workers.Footnote 1 Following its implementation, the respiratory health of bartenders significantly improved [2]. This finding has also been evidenced internationally where tobacco control policies in bars have decreased exposure to second-hand smoke and improved bartender health [3–5]. AB13 has also received considerable public support in California. Tang et al.[6] surveyed randomly selected bar patrons to assess changes in attitudes towards AB13 and found that approval for the law increased from 59.8% to 73.2% between March 1998 and June 2000.

Although several smokefree bar ordinances, both in California and abroad, enjoy relatively high rates of compliance [7–9], bar workers in some communities continue to be exposed to second-hand smoke in their workplaces [10]. In previous research, the authors conducted 479 unobtrusive observations in 121 randomly sampled stand-alone bars in San Francisco, CA, and found that smoking in bars was significantly associated with patron ethnicity. Field staff observed smoking in 30% of bar observations and found that bars serving Asian and Irish patrons were more likely to be smoky. Bars serving Latino patrons, on the other hand, were less likely to be smoky [10].

As a result, the authors conducted a second study to closely examine sociocultural and contextual factors that might inhibit or promote the adoption of tobacco control policy in bars serving these diverse ethnic groups. In this paper, we present findings specifically on bars serving Asian patrons. Findings on bars serving Irish and Latino patrons will be presented elsewhere. By analyzing the implications of sociocultural and contextual factors on policy, we hope to provide a more nuanced understanding of the challenges involved in implementing tobacco control policy among ethnic minority populations.

Relatively few studies have focused on cross-cultural differences in the practice of public health policy. Robert Hayden [11] compared the impact of mandatory seatbelt laws in Yugoslavia and Illinois, considering differences in implementation and enforcement mechanisms. He found unexpected differences in adoption of the policy in the two regions. In Yugoslavia, compliance rates were low despite strong enforcement efforts. In Illinois, even though the law was nominally enforced and public opposition was pronounced, compliance rates were comparatively high. In contrast to Americans, Hayden argued that Yugoslavs did not perceive laws to be “normatively binding” and thus did not feel an obligation to comply. Hayden’s study suggests that cultural norms may well influence the effectiveness of policy.

Sociocultural norms and public health policies can be at odds. Some scholars suggest that the cultural norms of a group’s home country may influence how a group reacts to a particular policy [12]. Policies restricting smoking may be particularly difficult for some immigrant communities to adopt, because until recently, few countries around the world had implemented tobacco control policies. Kim and Nam [12] argue that identifying barriers that specific immigrant communities may face regarding tobacco control policies is necessary for effective implementation.

Bar staff and owners in most California bars we have studied frequently described their perception of lax enforcement of the smokefree workplace law and their initial concern that preventing customers from smoking would hurt their businesses. Nonetheless, the majority of California bars appear to comply with the law. The case of Asian bar communities is different in that those common issues of concern were combined with culturally framed behaviors, attitudes, and practices that made the adoption of the policy problematic.

Methods

Sample Selection and Data Collection

For the purposes of this study, we classified Asian bars as those with a predominantly Asian staff and serving a predominantly Asian clientele. From a census of all bars serving Asian patrons in San Francisco and Los Angeles counties in the state of California, we randomly sampled 50 stand-alone bars, 25 in each of the two counties. A stand-alone bar is one that is not connected to a hotel or restaurant and where drinking is the primary purpose of the bar. Because public health officials had noted the persistence of smoking within stand-alone bars [9], our study narrowed its focus on this particular category of bars. Pairs of trained field observers conducted three rounds of hour-long, unobtrusive observations in each bar between 2004 and 2005. We attempted to match observers to patrons on gender, ethnicity, and languages spoken. Each observer produced two types of observational data: survey data and descriptive field notes. The survey form included questions related to any evidence of smoking within the bars (e.g., ashtrays present or cigarette butts on the bar floor), the numbers of patrons or staff (both inside and outside) witnessed smoking, and the demographics of patrons. We also instructed observers to note what languages people spoke in the bar. Predominant use of a non-English language was used as a proxy measure for immigrant status, to be corroborated in interviews. Semi-structured field notes provided descriptive narratives of each observer’s experience within each bar. Elsewhere we have reported details of the study methodology [13] and sample frame [10].

Following the observation period, trained bilingual/bicultural field interviewers recruited bar patrons, staff, and owners from the sampled bars for confidential, semi-structured interviews. The interview guide included questions about the respondent’s background and personal smoking habits; observations on social relations within the bar and smoking in the bar; the respondent’s personal understanding of and reactions to the smokefree workplace ordinance; and perceptions of social relations within the ethnic community and of attitudes towards smoking and, if the respondent was born outside the USA, of smoking norms in the respondents’ home countries. The research staff instructed the interviewers in ethnographic techniques such as establishing rapport and using a variety of probes to encourage respondents to respond openly and to expand upon their answers.

The field interviewers purposively sought to recruit long-time and frequent, or “regular,” patrons and staff who were expected to have a breadth of knowledge about the bar. Therefore, eligibility consisted of working in, owning, or regularly frequenting the bars for at least one year. In total, staff conducted interviews with 9 bar staff, 7 bar owners, and 13 regular patrons for a total of 29 interviews. Interviewers offered all respondents the option of conducting the interview in English, Mandarin, Cantonese, or Korean, as these languages had been identified by observers as the main non-English languages spoken.Footnote 2 None of the respondents chose English. Professional translators, who were native speakers of Mandarin, Cantonese, or Korean, translated and back-translated for accuracy all protocols and materials for the interviews, including the interview guide and consent forms. All respondents received a $40 honorarium in thanks for their participation. With the respondent’s permission, interviewers digitally recorded the interviews. Professional translators, who were native speakers of Mandarin, Cantonese, or Korean, translated and transcribed the recorded interviews. The Institutional Review Board of the Pacific Institute for Research and Evaluation approved all data collection procedures for the protection of human subjects.

Analysis

Because this paper is concerned with addressing sociocultural features of the bar communities, we report primarily on qualitative findings from the semi-structured interviews, using the observational field notes to triangulate emergent themes. We have provided details of the analysis of the quantitative data elsewhere [10]. Briefly, we uploaded the survey data to SPSS for analysis. In order to identify the degree of smoking in bars, we generated a basic frequency of the variable “any smoking witnessed inside.”

The qualitative data consisted of the interview transcripts and observer field notes. A trained research assistant coded the qualitative data in ATLAS.ti, a qualitative data management software package [14], to index textual data for later retrieval and analysis. The initial codebook consisted of pre-defined codes informed by previous literature and the authors’ research. The research staff met periodically to review the coding and interview results and added new codes as topics emerged from reviews of the data, relevant literature, and conversations with the field staff. The research team identified key findings through an inductive, pattern-level analysis by considering the frequency of similar items across all interviews, corroboration of items between interviews and observational data, and congruence of items with theory. We also searched for disconfirming evidence in order to clarify and reformulate emergent themes [15].

Findings

Description of Sample Bars

Bars serving Asian patrons were distinct from other bars in our study. Data from field notes and interviews indicated that most patrons from the bars in our study had arrived in the USA relatively recently from China, Taiwan, and Korea. Observers noted that patrons and staff most often spoke in their native languages and that English was rarely spoken. Respondents confirmed that most of the patrons were foreign-born. Hostess-type bartenders and “bar girls,” responsible for entertaining a predominately male clientele, were found to be an integral part of the environment of these bars [16], a characteristic common in some bars in Asia [17].

Bars serving Asian patrons were also distinct from other bars in our study because of the frequency of smoking observed indoors. Field observers witnessed individuals smoking indoors on any observation in 82% of the 50 bars serving Asian patrons. Additionally, observers witnessed smoking on all three observations in over 50% percent of the sample bars. These findings are particularly intriguing when compared to the 24.4% rate of indoor smoking reported by Weber et al. [9] in their site inspections of randomly selected Los Angeles County stand-alone bars. We propose that sociocultural factors unique to bars serving patrons who are primarily foreign-born Korean and Chinese help to explain the pervasiveness of smoking in our sample bars.

Social Relationships

Interviews revealed that the interests of customers came first. A female bartender stated that “the customer is king” and that “if you tell (customers) not to smoke, they will get mad and take off.” One owner, who stated that she felt “guilty” because she allowed smoking, explained,

I wouldn’t say that they come here just to smoke, but I think they want to sit comfortably while drinking and smoking.

Bar owner

While it is clear from the rest of the interview that she recognized the potential repercussions (such as fines) associated with permitting smoking inside the bar, she chose to suppress her own discomfort and allow patrons to continue smoking inside.

The relationship between bar staff and patrons was more than just one of business; it was also one of friendship. Field observers commented on the strong relationships they observed between bar staff and patrons, noting that frequently bar staff and patrons appeared suspicious of their presence likely because they were outsiders and not a part of the regular bar community. One bartender described how the strength of interpersonal relationships took precedence over policy. He explained that the relationships between customers and bar staff were more important.

Easterners are not that strictly law abiding. They don’t like to follow the law. Take a look at the Americans: if you tell them not to smoke in here, do you see anyone taking a few puffs? They’re not even smoking slightly. However, us Easterners, we care too much about social relationships. [Patrons] feel that, “I come here regularly. I’ve been coming to your bar this long, it’s almost like we’re friends. What’s the deal?”

Bartender

Patrons, who enjoyed smoking in bars, might feel betrayed by their “friend” (i.e., the bartender) if asked to comply with AB13. As a result, bar staff allowed their “friends” to smoke inside in order to protect and maintain these commercial friendships [18].

Social Interactions

Field observers noted the presence of cigarettes in most bar social interactions, with patrons and bar staff constantly sharing and offering cigarettes to one another. Female bar staff, whose job was partly to direct and take part in bar social interactions, frequently took on the role of lighting patrons’ cigarettes for them.

The four staff members down liquor shots with the older Korean patron who sits near the bar, and share food with him that is brought in a white plastic bag from one of the staff member’s cars. Throughout our observation, the female bartenders light the patrons’ cigarettes for them.

Field notes

Interviews contextualized the observation data, suggesting that smoking was integral to nightlife and socializing, and therefore, AB13 conflicted with established norms. Many respondents in our studies noted that a person going outside to smoke interrupted his or her conversation with friends. This issue, however, was especially salient for Asian interview respondents. One patron explained that going outside for a smoke was “weird” because it conflicted with the expected flow of group interactions.

If we’re dancing and drinking and in the middle of drinking, you have to go wash the dishes, that feeling would be quite weird. In other words, you’re chatting, then when you want to smoke, you might say, “Ah, I’ll chat later, now I’m going to go have a smoke.” When you come back, you’re out of sync with the conversation, and you can’t catch up.

Patron

Patrons reported that going outside to smoke severed the social interaction such that one was not able to overcome the disruption to rejoin the group after smoking. By prohibiting smoking inside bars, AB13 may have created a situation that conflicted with culturally normative social interactions in bars.

Although respondents were well-aware of the harmful effects of second-hand smoke, smoking was considered such an essential part of sociability that it invariably took precedence over any concerns about health. One respondent, who said that smoking “causes diseases,” talked about the importance of smoking in social transactions in mainland China.

The Chinese born in the mainland, when they reached eleven or twelve, no matter where they go, they’ll be exposed to cigarettes. There are places called “san zai hok” where people hang out. When you want to do business in the mainland, you’ll have to get in touch with those people. If you don’t smoke, you can’t do business there. Smoking is very important.

Bartender

For the most part, interviews revealed that the negative health effects of smoking were disassociated from the social effects of smoking. However, some respondents reasoned that patrons who did not smoke and who were concerned about the effects of second-hand smoke did not have to frequent bars that allowed indoor smoking. The bar was considered a place for relaxation, entertainment, and networking with friends, and smoking was an essential component for creating that type of environment.

Avoiding Confrontation

AB13 requires that the bar staff and owners uphold the policy within their own establishments, or else the bar is held responsible for any infractions. Interviews revealed that bar staff and owners anticipated confrontations resulting from any direct implementation of the law and, therefore, chose to ignore AB13. As one bartender explained:

I think that bars condone smoking to stay afloat and that there are no bars that really encourage smoking. We let the patrons smoke because, if we make them stop, it will piss them off and it will decrease our revenues.

Bartender

Bar staff and owners were reluctant to engage in direct confrontations with their patrons over the smoke-free ordinance. Respondents from bars not serving a predominantly Asian clientele also talked about the many confrontations they faced when upholding AB13. Nevertheless, they more frequently upheld the policy. In bars serving Asian patrons, however, avoiding confrontation appeared to take precedence over the law, in part because of the importance of maintaining social relationships.

In Asian bars, where the interest of the collective is especially important, upholding the smoke-free law was more than just a legal action. Patrons perceived compliance with AB13 to be confrontational in that it placed the interests of the bar owner (individual) over the interests of the bar community (collective). One patron explained that prohibiting smoking inside could be construed as the bar owner disrespecting his or her own community. “If you ask people not to smoke inside, then it’s like you don’t want their business.” Because patrons might “get pissed off,” bar owners and staff preferred to avoid confrontation rather than jeopardize their relationships with patrons and their bar revenues.

Field staff observed few direct confrontations about smoking. However, attempts by bar staff to prevent patrons from smoking occurred a few times in locations where bar staff appeared suspicious of the field observers. In one location, where field staff believed that the bartenders considered them undercover police officers, the observers witnessed a bartender attempting to uphold AB13.

The middle-aged patron took out a cigarette to smoke, but the bartender told him not right now. When he asked why, she said that he just could not. He asked again, and she said that he should know he was not allowed to smoke. He said he did not care about the law, and she told him to go ahead and smoke.

Field notes

Discussion

Although comprehensive tobacco control policies in California are credited with contributing to a decrease in smoking among the general population [19], this reduction is not evidenced among Asians in California, which is home to over 4 million Asians or Asian-Americans [20]. Statistics from World Health Organization surveys in the late 1990s showed smoking in Asian countries to be much higher than in California [21]. In particular, research indicates that in China and Korea over 60% of men smoke [22]. Such data support the findings in this study that indicate that smoking in bars serving Asian patrons in California may be supported and reinforced by recent immigrants. In the absence of strong efforts to uphold the law by bar staff, Asian bar patrons would be expected to continue their well-established practice of smoking while drinking in bars.

Interview data from our study indicated that while Asian bar patrons and staff may be well aware of the law, multiple sociocultural factors contributed to their reluctance to uphold the smokefree workplace law. The importance of social relationships, the emphasis placed on normative social interactions, and an aversion to confrontation all complicated the implementation of AB13 in these bars. Scholars in anthropology, international business, and cross-cultural communication have addressed the saliency of these sociocultural norms in Korean and Chinese communities [23–26]. Bar staff in our study placed much value on the social relationships developed within the bar community, desiring to avoid conflict with patrons at all costs even if it meant breaking the law. Additionally, the cultural tendency to complete social interactions created expectations of a normative progression of social interactions and promoted cohesion between people, both which could be disrupted if someone had to leave the bar in order to smoke.

The situation of persistent smoking in these bars is also complicated by the socially integrative role of smoking. Other research has found that sharing cigarettes with friends, and even strangers, in China and Korea is quite common [27, 28], and smoking in groups appears to help establish and maintain social relationships. Pan and Hu [28] assert that “social smokers” emerged in China because of the of use cigarettes for social networking. People, who might not otherwise smoke, choose to use cigarettes socially to avoid jeopardizing their social relationships. Kim, Son, and Nam [29] found that Korean male smokers believed that the sociability created by smoking was smoking’s most attractive feature. Because bars serve sociability, and drinking and smoking appear to go hand in hand [30], the poor reception of tobacco control policies in bars serving Korean and Chinese patrons is further explained by recognizing the socially integrative role of smoking within these communities.

The sociocultural features operating within many bars serving Asian immigrants in California may keep the state’s multi-faceted tobacco control program from wielding as much influence with Asian bar-goers as it does with other bar communities, and also other non bar-going Asian communities. Zhu et al. [31] found that Asian immigrants have a high ratio of quitting once they are exposed to California smokefree norms and policies. However, in order for Asian bar communities to benefit, they need to be exposed to work and recreational environments in which the new smokefree norms are reinforced through the adoption of AB13.

Because of the importance placed on the collective interests of the bar community, increasing community-wide support for tobacco control policy may help to facilitate acceptance of AB13 among bar owners. Also, developing programs to foster and promote interpersonal relationships between patrons and staff, as well as to enhance the quality of service within the bar, might create added attractions for patrons which would outweigh any temporary inconvenience caused by banning indoor smoking. This might also prevent patrons from abandoning a smoke-free bar for a smoking bar. Finally, as we have shown elsewhere [32], all intervention campaigns are most likely to succeed when they are conducted in conjunction with an increase in the visibility of enforcement, so that bar staff and owners perceive a tangible consequence to permitting smoking inside of their establishments.

Although many workplace smoking ordinances, both in California and abroad, enjoy relatively high rates of compliance, some communities still suffer disproportionately from the effects of second-hand smoke. By gaining a more nuanced understanding of the many factors that impede the reception of policy in certain communities, enforcement officials may be better prepared to confront potential obstacles, and policy makers and legislators may consider these obstacles with reference to their own jurisdictions when drafting legislation.