How much progress has been made toward a smoke-free environment in the restaurants and bars of Japan? Limitations of partial bans and their enforcement Authors: Aoi Kataoka, Isao Muraki, Masakazu Nakamura, Yuri Ito Published: 2024 ABSTRACT AbstractBackgroundIn April 2020, the revised Health Promotion Act and Tokyo Metropolitan Ordinance to prevent second-hand smoke (SHS) exposure was fully enforced. The government has provided an exemption for small-scale, existing establishments by allowing them ‘unregulated’ status, and it is unclear to what extent indoor smoking bans have been applied in these small bars and restaurants. In this study, we report on indoor smoking status before/after the enforcement of the current legislation both regulated and unregulated restaurants and bars.MethodsSelf-questionnaire surveys were conducted in March 2020 and 2021. We applied the questionnaire to 6,000 restaurants and bars in Tokyo, Osaka and Aomori in the first survey and to 3609 restaurants in the second survey, including those that had responded to the first survey as well as 2800 additional establishments. We calculated the proportions of indoor smoking status before/after enforcement both regulated and unregulated establishments. In addition, we calculated the national estimate of indoor smoking status in all restaurants and bars in Japan after the enforcement using a restaurant guide website and the results of this study.ResultsThe responded establishments were 879 to the first (response rate: 14.7%) and 837 to the second (response rate: 23.2%). Of all the establishments, 67.0% kept “separate smoking or smoking” status, and 62.1% of unregulated establishments kept “separate smoking or smoking”. Although all the regulated establishments were required to change to “no smoking” by law, 78.4% that were “separate smoking or smoking” before the enforcement did not change their indoor smoking status. The national estimate of smoking status showed that the proportion of “no smoking” would change from 54.5% to 69.4% after the enforcement. If all regulated establishments changed smoking status to “no smoking”, the ideal proportion of “no smoking” would be 85.8% in Japan.ConclusionsMore unregulated restaurants and bars changed to “no smoking” than planned before the enforcement. In contrast, most of the regulated restaurants and bars continued to be “separate smoking or smoking” in non-compliance with the legislation. In order to prevent SHS exposure in all restaurants and bars, we need comprehensive legislation for indoor smoking bans in Japan.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12889-024-20765-6.Keywords: Second hand smoking, Passive smoking, Smoke-free legislation, Restaurants, Smoking Cessation, Smoking Prevention Introduction Tobacco kills more than eight million people a year worldwide, and around 1.2 million of those deaths are the result of non-smokers being exposed to secondhand smoke (SHS) [1]. SHS exposure causes many negative health outcomes including cardiovascular disease, cerebrovascular disease, and cancer [2, 3]. In addition, it has been reported that occupational exposure to SHS causes respiratory symptoms and asthma in restaurants and bars [2]. To protect the public from the harms of SHS, the World Health Organization Framework Convention on Tobacco Control (WHO FCTC) requires comprehensive smoking bans in all indoor public spaces [4]. In recent years, several countries have introduced smoke-free legislation in public spaces; however, although the framework was ratified by 195 countries, only 67 have the comprehensive smoke-free legislation FCTC calls for [5]. Although Japan has ratified FCTC, its smoke-free legislation for public spaces has not been comprehensive. While the revised Health Promotion Act of April 2020 strengthened smoke-free protections and prohibits smoking in public places such as schools, hospitals, and government offices, exceptions remain for restaurants and bars. The revised law allows restaurants and bars to set up indoor smoking-only rooms in which no eating or drinking takes place and to permit the use of heated tobacco products (HTPs) in smoking rooms where customers can eat and drink (Supplementary Fig. 1). However, it has been shown that separating smokers and non-smokers can lower elements of SHS exposure such as the nicotine concentration in non-smoking areas, but cannot eliminate it completely [3, 6]. Therefore, even if smoking rooms were established, the air was cleaned, and the building was ventilated, it would not adequately prevent non-smokers in non-smoking areas from exposure to SHS. In addition, restaurants and bars which have an area of less than 100 m2 or have capital of 50 million yen or less, and opened before April 2020 are except from the law, and can choose indoor smoking status by posting information signs [7]. Also in April 2020, the Tokyo Metropolitan Ordinance to prevent SHS exposure was enforced, and this was more restrictive than the revised law in that only restaurants and bars with no employees in addition to regulation of seating area, capital and year opening were allowed to choose indoor smoking by posting information signs [8]. The number of restaurants and bars not regulated by legislation was estimated at 20% in Tokyo and 50% in 46 other prefectures [9]. In these establishments, the owners decide whether or not to allow indoor smoking. Therefore, it is unclear how many have adopted smoke-free rules since the enforcement of the SHS exposure legislation. Also, as the COVID-19 pandemic occurred at the same time as the enforcement of the law, bar or restaurant owners might have hesitated to adopt a smoke-free policy for fear of reducing sales at a time when business was already drastically curtailed. We aim to examine the current status of indoor smoking in restaurants and bars before and after the enforcement of the anti-SHS legislation, and examine how much progress has been made with the implementation of smoking bans, especially in the unregulated restaurants and bars in Japan. Method Participants and data collection The surveys were conducted twice, before and after the enforcement. The first survey (before the enforcement) was conducted in February–March 2020, and questionnaires were applied to 6,000 restaurants and bars operating in 20 cities in Tokyo, Osaka, and Aomori prefectures (detailed questionnaires shown in Appendix 1). The second survey (after the enforcement) was conducted in March 2021. Of the 879 restaurants and bars that responded to the first survey, 809 were still running in 2021 and were included in the second survey. A further 2,800 establishments that were still in operation in 38 cities in Tokyo, Osaka, and Aomori prefectures were added, resulting in a total sample of 3,609 restaurants and bars (detailed questionnaires shown in Appendix 2 and 3). For both surveys, we mailed a self-administered questionnaire to respondents which they could return by mail or complete online. The sample prefectures were selected based on a combination of degree of urbanization and the status of enforcement of the anti-SHS legislation. Tokyo and Osaka prefectures were selected as the metropolitan samples, as the Tokyo Metropolitan Ordinance to prevent SHS exposure was enforced in April 2020 and the Osaka Prefectural Ordinance to prevent SHS exposure will be enforced in 2025. In contrast, as the rural sample, Aomori prefecture was selected because the ordinance has not been enforced and is currently under consideration. Furthermore, the adult smoking rate was the second highest among the 47 prefectures in 2019 [10], making smoking control a priority issue in health policy. The sample selection of food and beverage establishments was conducted using Tabelog.com (operated by Kakaku.com Inc., Tokyo), a popular online restaurant guide in which restaurants and bars from all over Japan are registered. First, we focused on restaurants and bars that were not regulated by law and selected them based on the definitions for small establishments: in Osaka and Aomori prefectures restaurants with fewer than 50 seats and bars which were less than 100 m2 in size and in the Tokyo metropolitan area, restaurants with fewer than 20 seats and bars with no employees. Second, we conducted random sampling to avoid bias among food and beverage service categories of restaurants and bars. The proportion of each service category was based on the number of establishments per service category registered on the website. To ensure the number of establishments in each service category would be approximately the same, the sampling was conducted with the following proportions as the target: 20% for Japanese food, 15% each for Japanese gastropubs, cafes, and ramen shops, 10% for western food, 4% for family restaurants, 12.5% for bars and pubs, 6% for Asian and international cuisine, and 2.5% for barbecue restaurants. These service categories have been adopted by the major restaurant guide websites in Japan and are defined as follows: Japanese food refers to a restaurant that serves traditional Japanese cuisine; Japanese gastro pubs, called Izakaya in Japanese, are informal Japanese bars that serve a variety of alcoholic drinks and accompanying food; cafes are small restaurants or coffeehouses that serve light meals, beverages, and desserts; ramen shops are restaurants that specialize in ramen dishes (wheat-flour Japanese noodles in broth); western food refers to restaurants that serve western cuisine (mainly European and American); family restaurants mainly target families and their menus include a wide variety of Japanese, Western and Chinese dishes at low prices; bars and pubs focus on serving alcoholic drinks; Asian and international cuisine refers to restaurants that serve Asian or multiethnic dishes; and barbecue restaurants, called Yakiniku in Japanese, are restaurants where customers can grill their meat themselves. Measures We recorded the indoor smoking status in restaurants and bars before/after the enforcement, and also, in the first survey, recorded plans for changing indoor smoking status after the enforcement. The indoor smoking status categories were “no smoking” and “separate smoking or smoking”; the latter comprised “complete smoking separation”, “area smoking separation”, “time limited smoking separation”, and “smoking”. In addition, we recorded the location (Metropolitan: Tokyo and Osaka, Rural: Aomori); seating area (less than 30 m2, 31–100 m2, more than 100 m2); year of opening (2010-March 2020: January-March 2020, 2015–2019, 2010–2014, before 2010, 2000–2009, before 1999); service category (serves mainly alcohol: Japanese gastropub, bars and pubs; serves mainly meals: Japanese food, cafes, ramen shops, western food, family restaurants, Asian and international cuisine, and barbecue restaurants); and the number of employees (zero–one employee, two or more employees). Data analysis We calculated the proportion of the indoor smoking options in operation before and after the enforcement and conducted a chi-square test to compare differences in restaurant and bar characteristics between indoor smoking status in the first and second surveys: “stayed no smoking”, “changed from separate smoking or smoking to no smoking”, and “stayed separate smoking or smoking”. We conducted these analyses for both the regulated establishments that had been required to change to “no smoking” by the legislation and the unregulated establishments that could choose “no smoking”, “separate smoking or smoking” indoor smoking status. To examine the impact of non-response on indoor smoking status, a sensitivity analysis was conducted. We focused on 86 nonresponding restaurants and bars about indoor smoking status in each 2019 and 2020, and we assumed the following for patterns for these 86 restaurants and bars, and calculated the proportion of no smoking for each of the patterns: 1) if all nonresponding restaurants and bars were no smoking in 2019 and 2020, 2) if nonresponding restaurants and bars in 2019 are no smoking and in 2020 are separate smoking or smoking, 3) if nonresponding restaurants and bars in 2019 are separate smoking or smoking and in 2020 are no smoking, 4) if all nonresponding restaurants and bars are separate smoking or smoking both in 2019 and 2020. In addition, we estimated the indoor smoking status of restaurants and bars before/after the enforcement both nationally and for establishments that mainly served meals and those that mainly served alcohol (Standardized Proportion of Indoor Smoking Status: SPISS) using the calculation (detailed methods shown in Appendix 4), where i indicates location: i = 0 is Tokyo and i = 1 is 46 prefectures other than Tokyo; j indicates the regulated status by law: j = 0 is not regulated and j = 1 is regulated; y indicates survey point in time: y = 0 is 2019, and y = 1 is 2020: s indicates indoor smoking status; s = 0 is no smoking, and s = 1 is separate smoking or smoking. SPISSys was the sum of pijys∗wij. pijys was the proportion of indoor smoking status observed in this survey, and wij was the distribution of restaurants and bars in Japan used as weighting. wij was calculated using the Tabelog database (T was the total number of restaurants and bars, Ti was the number of restaurants and bars in Tokyo and in 46 prefectures) and Rij was the expected proportion of regulated or unregulated establishments after the enforcement in Japan (the number of restaurants and bars not regulated by law was estimated at 20% in Tokyo and 50% in 46 other prefectures). All analyses were conducted in R version 3.6.1 [11]. “Ethical Guidelines for Medical and Health Research Involving Human Subjects” were not applicable as this study examined restaurants and bars not human subjects, and ethics approval was waved by institutional review board of Osaka University: Ethical Review Board Osaka University Hospital. Information on each restaurant and bar was handled on personal computers with appropriate security measures such as setting passwords and installing security software, so that no disadvantage could arise from this study. Results Of 6,000 restaurants and bars approached in the first survey, 879 responded to the questionnaire (response rate: 14.7%). Of 3,609 restaurants and bars, approached in the second survey, 837 responded (response rate: 23.2%). Of these, 707 were included in the analysis, after excluding those that did not answer questions about their indoor smoking status before the enforcement, seating areas, and the number of employees (Tokyo only), and those that opened after April 2020. Of the 707 establishments, 457 were unregulated with less than 100m2 of seating area and no employees (Tokyo only), and 250 were regulated (Fig. 1). Flow diagram of the selected restaurants and bars in this study Characteristics of the restaurants and bars In terms of characteristics, a higher proportion of establishments were in a metropolitan location, opened before 2010, had 31–100 m2 seating area, more than two employees, mainly served meals and did not allow use of HTPs. The significant differences in characteristics between the unregulated and regulated establishments were in location (metropolitan: 59.1% of the unregulated and 88.8% of the regulated establishments), seating area (31–100 m2: 60.6% of the unregulated establishments, 26.8% of the regulated establishments), number of employees (more than two employees: 48.6% in the unregulated, 68.4% in the regulated establishments) (Table 1). Table 1Differences in characteristics between regulated and unregulated restaurants and barsAll establishments (n = 707)Unregulated (n = 457)Regulated (n = 250)p-valuen%n%n%Location< 0.01Metropolitan: Tokyo, Osaka49269.627059.122288.8Rural: Aomori21530.418740.92811.2Year of opening0.86before 201036151.123551.412650.42010-March 202034648.922248.612449.6Seating area< 0.01less than 30m232646.118039.414658.431–100m234448.727760.66726.8more than 100m2375.200.03714.8Number of employees< 0.01zero–one employee29942.322148.47831.2two or more employees39355.622248.617168.4Service category0.10mainly served alcohol18225.710823.67429.6mainly served meals52574.334976.417670.4HTPs0.74allowance476.6286.1197.6not allowance45163.829263.915963.6p-value: chi-square test between unegulated and regulated establishmentsHTPs: heated tobacco productsUnregulated establishments: restaurants and bars not regulated by law and able to choose indoor smoking statusRegulated establishments: restaurants and bars regulated by law and required to implemennt smoking ban after the enforcement Differences in characteristics between regulated and unregulated restaurants and bars p-value: chi-square test between unegulated and regulated establishments HTPs: heated tobacco products Unregulated establishments: restaurants and bars not regulated by law and able to choose indoor smoking status Regulated establishments: restaurants and bars regulated by law and required to implemennt smoking ban after the enforcement Indoor smoking status before and after enforcement of the legislation Among the 707 establishments analyzed, the proportion of “no smoking” was 54.6% (386/707) before and 67.3% (476/707) after the enforcement (Supplementary Fig. 2). Results of the sensitivity analysis showed the following the proportion of “no smoking” before/after the enforcement: (1) if all nonresponding restaurants and bars are “no smoking” both in 2019 and 2020: 58.8% (458/779) before and 70.7% (551/779) after, (2) if nonresponding restaurants and bars in 2019 are “no smoking” and in 2020 are “separate smoking or smoking”: 58.8% (458/779) before and 65.3% (509/779) after, (3) if nonresponding restaurants and bars in 2019 are separate smoking or smoking and in 2020 are “no smoking”: 49.5% (386/779) before and 70.7% (551/779) after, (4) if all nonresponding restaurants and bars are “separate smoking or smoking” both in 2019 and 2020: 49.5% (386/779) before and 65.3% (509/779) after (Supplementary Table 1). In the unregulated establishments, the proportion of “no smoking” increased from 51.0% (233/457) to 67.4% (308/457). In particular, among the unregulated establishments which had “separate smoking or smoking” status before the enforcement, 34.8% (78/224) changed to “no smoking” (Fig. 2, with more detailed indoor smoking status is shown in supplementary Fig. 3). HTPs use was permitted in 76.0% (19/25) of the unregulated establishments that reported they were “staying separate smoking or smoking”, and in 2.7% (8/294) of those that were “staying no smoking” or “changing from separate smoking or smoking to no smoking” (Supplementary Table 2). In addition, 26% (33/127) of the unregulated establishments that were “separate smoking or smoking” planned to change their status to “no smoking” in the first survey; however, after the enforcement, 40.2% (51/127) of them actually changed their status to “no smoking” (Supplementary Fig. 4). Changing indoor smoking status for the unregulated restaurants and bars before/after enforcement of smoke-free legislation. We observed the actual indoor smoking status in the unregulated restaurants and bars that were able to choose whether to implement smoking bans or continue to allow smoking in the survey before and after the enforcement, and calculated the proportion of indoor smoking status Although all regulated establishments were required by law to change their indoor smoking status to “no smoking”, 78.4% (76/97) of those that were “separate smoking or smoking” before the enforcement did not change their status and were “staying separate smoking or smoking”. This was equivalent to 31.2% (78/250) of all the regulated establishments that were legally required to change their status (Fig. 3, with more detailed indoor smoking status shown in supplementary Fig. 5). HTPs use was allowed in 60.0% (12/20) of the regulated establishments that were “staying separate smoking or smoking”, and 4.4% (7/158) of those that were “staying no smoking” or “changing from separate smoking or smoking to no smoking” (Supplementary Table 3). Changing indoor smoking status for the regulated restaurants and bars before/after enforcement of smoke-free legislation. We observed the actual indoor smoking status in the regulated restaurants and bars that had to implement smoking bans in the survey before and after the enforcement, and calculated proportion of indoor smoking status Differences in the characteristics of restaurants and bars In the unregulated establishments, there were significant differences in characteristics between “changing from separate smoking or smoking to no smoking” and “staying separate or smoking” in number of employees (more than two employees; 60.3% “changing from separate or smoking to no smoking”, 43.9% “staying separate smoking or smoking”, p < 0.05), seating areas (31–100 m2; 76.9% “changing separate smoking or smoking to no smoking”, 39.6% “staying separate or smoking”, p < 0.01), and service categories (serves mainly meals; 79.5% in “changing separate smoking or smoking to no smoking”, 52.5% “staying separated or smoking”, p < 0.01) (Supplementary Table 2). In the regulated establishments, the only significant difference in characteristics between “staying no smoking”, “changing separate smoking or smoking to no smoking” and “staying separate or smoking” was in the service category (serves mainly alcohol: 59.0% “staying separate or smoking”, 16.7% “staying no smoking”, “changing from separate smoking or smoking to no smoking”, p < 0.01) (Supplementary Table 3). In terms of owner smoking status, among owners of unregulated establishments, more never or former smokers changed to “no smoking” than stayed “separate smoking or smoking”. Owners who were never or former smokers may be more likely to implement a change in indoor smoking status to a complete smoking ban (Supplementary Table 4). National estimate of the indoor smoking status of all restaurants and bars in Japan The estimated national proportion of “no smoking” in all establishments before the enforcement was 54.5% and 69.4% after. When we divided establishments by legal regulation status, 30% of regulated establishments continued to be “separate smoking or smoking” after the enforcement despite the legal requirement that they ban indoor smoking. If compliance with the legislation had been complete and all regulated establishments had changed to “no smoking” as an ideal situation, the estimated national proportion of “no smoking” would have been 85.8% after the enforcement (Table 2, Fig. 4, Supplementary Table 5, Supplementary Fig. 6). Table 2National estimate of indoor smoking status before/after the enforcement. National estimate of no smoking (%) and separate or smoking (%) before the enforcement (2019)LocationRegulation by lawNumber of establishments in Tabelog TiExpected % of regulated or unregulated establishments RjExpected number of establishments Ti*RjExpected % of establishments wij = Ti*Rj / TNo smoking (%) before the enforcement observed in this study (2019) pijysNational estimate of no smoking (%) before the enforcement (2019) SPISSysSeparate smoking or smoking (%) before the enforcement observed in this study (2019) pijysNational estimate of separate smoking or smoking (%) before the enforcement (2019) SPISSysTokyounregulatedT0: 129553R0: 0.2T0*R0: 25910.6w00: 0.03p0000: 0.65p0000*w00: 0.02p0001: 0.35p0001*w00: 0.01regulatedR1: 0.8T0*R1: 103642.4w01: 0.13p0100: 0.62p0100*w01: 0.08p0101: 0.38p0101*w01: 0.0546 prefectures other than TokyounregulatedT1: 694659R0: 0.5T1*R0: 347329.5w10: 0.42p1000: 0.50p1000*w10: 0.21p1001: 0.50p1001*w10: 0.21regulatedR1: 0.5T1*R1: 347329.5w11: 0.42p1100: 0.56p1100*w11: 0.24p1101: 0.44p1101*w11: 0.19TotalT: 8242121.00SPISS00: 0.55SPISS01: 0.45National estimate of no smoking (%) and separate or smoking (%) after the enforcement (2020)LocationRegulation by lawNumber of establishments in Tabelog TiExpected % of regulated or unregulated establishments RjExpected number of establishments Ti*RjExpected % of the establishments wij = Ti*Rj / TNo smoking (%) after the enforcement observed in this study (2020) pijysNational estimate of no smoking (%) after the enforcement (2020) SPISSysSeparate smoking or smoking (%) after the enforcement observed in this study (2020) pijysNational estimate of separate smoking or smoking (%) after the enforcement (2020) SPISSysTokyounregulatedT0: 129553R0: 0.2T0*R0: 25910.6w00: 0.03p0010: 0.76p0010*w00: 0.02p0011: 0.24p0011*w00: 0.01regulatedR1: 0.8T0*R1: 103642.4w01: 0.13p0110: 0.68p0110*w01: 0.09p0111: 0.32p0111*w01: 0.0446 prefectures other than TokyounregulatedT1: 694659R0: 0.5T1*R0: 347329.5w10: 0.42p1010: 0.67p1010*w10: 0.29p1011: 0.32p1011*w10: 0.13regulatedR1: 0.5T1*R1: 347329.5w11: 0.42p1110: 0.71p1110*w11: 0.30p1111: 0.29p1111*w11: 0.12TotalT: 8242121.00SPISS10: 0.69SPISS11: 0.31SPISSys was the estimated proportion of nationwide indoor smoking status in restaurants and bars in Japan before/after the enforcementi indicates location: i=0 is Tokyo and i=1 is 46 prefectures other than Tokyoj indicates the regulated status by law: j=0 is unregulated and j=1 is regulatedy indicates survey point in time: y=0 is 2019, and y=1 is 2020s indicates indoor smoking status: s=0 is no smoking and s=1 is separate smoking or smoking National estimate of indoor smoking status before/after the enforcement. National estimate of no smoking (%) and separate or smoking (%) before the enforcement (2019) SPISSys was the estimated proportion of nationwide indoor smoking status in restaurants and bars in Japan before/after the enforcement i indicates location: i=0 is Tokyo and i=1 is 46 prefectures other than Tokyo j indicates the regulated status by law: j=0 is unregulated and j=1 is regulated y indicates survey point in time: y=0 is 2019, and y=1 is 2020 s indicates indoor smoking status: s=0 is no smoking and s=1 is separate smoking or smoking National estimate of indoor smoking status in all restaurants and bars before/after the enforcement. Estimated proportion of indoor smoking status before/after the enforcement was calculated by summing up the value of multiplying “Proportion of number of establishments in the Tabelog” by “Proportion of establishments in this study that were observed as no smoking, separated or smoking” before/after the enforcement. Ideal estimated proportion of smoking status in the ideal situation after the enforcement was estimated using the proportion assuming all regulated establishments changed “No smoking” and the national estimated proportion of “No smoking” in the unregulated establishments In addition, the estimated national proportion of “no smoking” in establishments that mainly served meals changed from 59.1% to 75.3% after the enforcement, and the estimated ideal proportion of “no smoking” when full legal compliance was achieved was 90% (Supplementary Table 6, Supplementary Fig. 7-A, B). The estimated national proportion of “no smoking” in establishments that mainly served alcohol changed from 43.9% to 51.5%, and the estimated ideal national proportion of “no smoking” when full legal compliance was achieved was 73.4% (Supplementary Table 6, Supplementary Fig. 7-C, D). Discussion To our knowledge, this is the first study to report on indoor smoking status before and after the enforcement of the law to focus on both unregulated and regulated establishments in Japan. Among the unregulated establishments that were “separate smoking or smoking” before the enforcement, more actually changed to “no smoking” after the enforcement than had originally expressed an intention to change. In contrast, almost 80% of the regulated establishments that were “separate smoking or smoking” before the enforcement became illegal afterwards as they maintained “separate smoking or smoking” status. In this study, we found that smoking bans were increasingly in operation in unregulated restaurants and bars; in contrast, indoor smoking was continuing in regulated establishments. The fact that smoking is continuing after the enforcement in regulated restaurants and bars may be a serious issue for public health. These findings may indicate shortcomings in communication of the law to restaurants and bars, and inadequate implementation of the law, which might have been exacerbated by the COVID-19 pandemic. It is necessary to ascertain, in detail, the reasons for bar or restaurant owners not changing their indoor smoking status, the degree of understanding of the current legislation, and the extent to which smoking rooms are still included in restaurants and bars. These findings can then be used to assist in the enforcement of comprehensive anti SHS legislation in Japan. In our results, the proportion of all nonresponding restaurants and bars about indoor smoking status in 2019 and 2020 was 11.0% (86/779). If all non-respondents were “separate smoking or smoking”, the proportion of “no smoking” was 5.1% lower in 2019 and 2.0% lower in 2020 than the observed proportion. Therefore, the proportion of “no smoking” may have been overestimated compared with the actual proportion of “no smoking”. According to the national estimate of indoor smoking status in Japan, establishments that mainly served alcohol made less progress in changing to “no smoking”. The estimated national proportion of “no smoking” in all mainly alcohol establishments was lower than in all mainly meals establishments. There was 23.8% difference in “no smoking” between mainly meals and mainly alcohol establishments in the current situation after the enforcement, and 16.6% difference in “no smoking” in the ideal situation after the enforcement. Since unregulated establishments can freely select indoor smoking status, little progress has been made in terms of voluntary smoking bans in the mainly alcohol establishments but the proportion of “no smoking” after the enforcement was higher in the unregulated mainly meals establishments. In addition, although all regulated establishments were required to impose an indoor smoking ban by the legislation, many did not comply. The proportion of regulated establishments which did not change smoking status was higher in mainly alcohol than in mainly meals establishments. These results might suggest the decision to create a smoke-free environment depended on whether or not the establishment served alcohol. Furthermore, these results suggest that under the current Japanese law that permits establishments to remain unregulated, it is difficult to make progress in making all bars and restaurants “no smoking”. Previous studies indicated that never or former smoker owners or employees in restaurants and bars were more likely to have a positive attitude towards the smoke-free legislation [12, 13]. Hence, the smoking status of owners would affect with the decision regarding indoor smoking status. The promotion of “no smoking” in unregulated establishments may also be associated with the owner’s own smoking status or the service category of the establishment. We found more restaurants and bars which serve mainly alcohol in the “staying separate or smoking” category than in the “changing from separate smoking or smoking to no smoking” in this study. Previous studies have shown that smoking bans were more prevalent in the mainly meals than the mainly alcohol restaurants and bars [14, 15], and the present study found similar results. Indoor smoking is common in bars or pubs, and owners would be concerned about the possible reduction of customers or sales that would result from an indoor smoking ban [16]. In Japan, the mainly alcohol restaurants and bars have traditionally been regarded as places where smoking is socially acceptable [15]. It appears that voluntary smoking bans in unregulated, mainly alcohol establishments may be less aggressively promoted than other service categories. The reality that we observed was that almost 80% of the regulated restaurants and bars that were required to change to “no smoking” by law did not change. The restaurants and bars that kept “separate smoking” or “smoking” contrary to the law, were more likely to serve alcohol and allow HTPs in their seating areas than those restaurants and bars in “staying no smoking” or “staying separate or smoking”. It has been indicated that many owners of restaurants or bars where alcohol can be consumed did not comply with the law regarding smoking bans. Reluctance to comply with the ban has been evident in many countries; in the face of forthcoming smoking bans, bar and restaurant owners in Ireland, Canada, and United States used scaremongering tactics (including fear of breaking the law and loss of employment for hospitality sector workers) to try to prevent the ban [17]. Furthermore, the partial Japanese law allows the establishment of heated tobacco smoking rooms, where eating and drinking are allowed, even in restaurants and bars that are regulated by the law. Therefore, it is possible that these restaurants and bars have chosen not to change to a completely smoke-free environment. We found that establishments meeting regulation criteria are allowed to operate as a “smokers’ lounge” if their primary purpose is smoking and face-to-face sales of tobacco products, and if it is limited to serving light meals and alcohol (Supplementary Fig. 1). Therefore, some regulated establishments may have changed their business designation to smokers' lounge in order to continue operating even if the law requires them to change to an indoor smoking ban. Also, under the law, unregulated establishments can set up smoking rooms/areas [7]. However, regulated establishments may mistakenly believe that their establishments are unregulated by law and choose to have smoking rooms/areas without notifying the government. Second, almost 30% of owners of restaurants and bars reported not fully understanding the revised Health Act or Tokyo Metropolitan Ordinance to prevent SHS exposure in the survey in Tokyo [18] and Osaka prefectures [19]. In this study, almost 30% of all establishments that had to change to “no smoking” were non-compliant with the law, which echoed the results of previous surveys. Previous studies indicated that a lower level of knowledge about the smoke-free legislation was associated with maintaining separate indoor smoking areas in restaurants and bars [20]. Third, it is possible that owners of restaurants and bars are knowingly not complying. Minimal sanctions and inadequate regulation have been considered as impediments to widespread smoking bans in previous studies [21, 22]. In Japan, there are penalties for violation of the anti SHS legislation but these are basically warnings, not immediate business suspensions or fines. In addition, monitoring is carried out only through reports from the general public, and not by governments. Furthermore, Japan’s smoke-free legislation is partial, and this may be associated with the non-compliance. It has been reported that the prevalence of illegal in-bar smoking was higher in countries with partial smoke-free legislation than in countries with comprehensive legislation [23]. If the competing restaurants and bars that were not regulated by legislation continue to allow indoor smoking, the regulated establishments may not change their indoor smoking status, for fear of losing custom and sales [22]. Enforcement of the law in Japan coincided with the COVID-19 pandemic. Therefore, restaurants and bars that were negatively affected by the economic impact of COVID-19 may have been concerned that the smoking ban would further drive away customers. However, it is already clear that the implementation of the smoke-free law does not have any economic impact on restaurants and bars [24–26]. Therefore, now that four years have passed since the pandemic, it is necessary to communicate again with the regulated restaurants and bars and take action to allay their concerns about economic impact. Under the partial law in Japan, it may be difficult to achieve total smoking bans in restaurants and bars. It has been pointed out that the exceptions to legal regulations may cause confusion in understanding and complying with the law, which may eventually lead to noncompliance [27, 28]. Additionally, voluntary smoke-free policies based on the self-judgement of restaurants and bars that are exceptions to the law are not effective and the adoption of effective legislative, executive or administrative measures is necessary [6, 29]. Furthermore, the establishment of smoking rooms permitted by the law also prevents total smoking bans in restaurants and bars. The elimination of tobacco smoke must be absolute to create a comprehensively smoke-free environment. The smoking rooms, ventilation or air filtration advocated by the tobacco industry have not been effective, and have reduced the effectiveness of smoke-free areas in restaurants and bars. Japanese law has hampered the achievement of completely smoke-free restaurants and bars and in some cases allowed HTPs use in the same room as eating and drinking customers. It has been reported that second-hand aerosol exposure from HTPs is associated with both asthma attacks/asthma-like symptoms and persistent cough in non-smokers [30]. It has further been suggested that HTPs may cause health hazards similar to those of tobacco. The current Japanese law permitting the establishment of HTPs smoking rooms in restaurants and bars is unlikely to protect the health of employees, and strong regulation of HTPs use is needed. It has also been reported that countries with partial laws are less likely to achieve adult smoking cessation than those with comprehensive laws [24, 29, 31]. The fact that smoking rates have not declined may suggest lower support for, and compliance with, the law among smokers in Japan than in countries with comprehensive laws. It is already clear worldwide that the comprehensive law maximizes the effectiveness of smoking bans. Enforcement of the law has improved respiratory symptoms and lung function among bartenders, a job category vulnerable to SHS exposure [2]. While the observed nicotine concentration decreased in establishments that were made smoke-free by law, the observed concentration in bars and restaurants that were exempt from legal regulation was very high, indicating a very high risk of SHS exposure for those who work in these restaurants and bars [27]. A previous study conducted in six European countries revealed that SHS exposure in public places, including restaurants was lower in countries with comprehensive laws than in countries with partial laws [29]. It has been suggested that there are two approaches to the successful enforcement of such comprehensive laws: the first is to build voluntary compliance, which involves awareness-raising strategies such as educational campaigns; the second is a proactive approach, which involves monitoring compliance and imposing penalties and sanctions on violators [28, 32]. Japanese law, which has approaches such as the creation smoking rooms, is unlikely to prevent SHS exposure and smoking-related disease among employees, as shown in previous studies in other countries; as long as the law is partial, it is expected that it would be difficult to make progress in creating entirely smoke-free restaurants and bars. Following the example of Spain, where a comprehensive law was enforced sometime after a partial law which allowed smoking rooms in restaurants and bars, Japan needs to revise its law as soon as possible, to end legal exemptions which make it possible for restaurants and bars to allow indoor smoking. In addition, it is necessary not only to simply improve the law, but also to conduct awareness-raising activities, to monitor compliance, strengthen penalties by administrative intervention, and to enhance communication to gain policy support so that the law can be fully enforced and the compliance rate increased, as indicated as a strategy.” Strengths and limitations We have reported the impact of the revised Health Promotion Act or the Tokyo Metropolitan Ordinance to prevent SHS exposure on smoking bans in restaurants and bars in Japan for the first time. The revised law will be reviewed by the government in 2025. The results of this study indicate problems of smoking bans in restaurants and bars due to the limited regulation by the current law and provide important evidence for discussion with policy makers for a review of the law. However, there are several limitations to this study. First, the survey response rate was only 14.3% before the enforcement and 23.2% after. This may indicate that the responses came primarily from owners who were interested in smoking bans. If so, among the establishments that did not answer, there may be a considerable number of restaurants and bars that will continue to allow indoor smoking. Thus, the proportion of “no smoking” in unregulated restaurants and bars may be lower than found in this study. Second, there were potential errors due to the self-completion of the questionnaire which could affect the validity and reliability of the results, especially for indoor smoking status. Third, the revised laws or ordinances were implemented during the COVID-19 pandemic era. The impact of this pandemic on the determination of indoor smoking status in restaurants and bars would be difficult to examine, as it requires comparison with a hypothetical situation in which there was no COVID-19 pandemic. Fourth, the present study was based on self-administered questionnaire survey responses by owners only. The opinions of employees or customers regarding smoking bans in restaurants and bars could not be ascertained. Moreover, there is no objective assessment of the extent to which owners, employees, or customers are actually exposed to SHS such as levels of nicotine and particulates in the breath, and cotinine in saliva, urine, and hair. In order to clarify the impact of the partial law on people’s support for smoking bans and on the health on employees and others, it is necessary to conduct further surveys with a wide range of subjects and to evaluate SHS exposure from both subjective and objective perspectives in the future. Conclusion The current implementation of indoor smoking bans in restaurants and bars is progressing in a higher proportion of establishments than planned before the enforcement. On the other hand, we observed restaurants and bars that have not implemented a ban despite being legally required to do so. These findings indicate that the details of the law may not have been adequately communicated to restaurants and bars and that the legislation is inadequate and its enforcement insufficient. In order to progress the indoor smoking bans in all restaurants and bars, it may be important to communicate the details of the legislation more insistently and in easily understood language. Also, further examination of the detailed factors that advance or hinder indoor smoking bans in restaurants and bars is necessary. We need to continue to monitor indoor smoking status, and to discuss the necessity of implementing comprehensive regulation in Japan with policy makers, researchers, owners and employees in restaurants and bars. This approach will enable us to share the issues related to understanding or complying with the current legislation, and to consider effective revisions to the law or ordinance regarding SHS exposure in restaurants and bars with policy makers. Supplementary Information Additional file 1: Appendix 1. Survey on the Impact on Restaurants or Bars of Strengthened Passive Smoking Countermeasures. This file was the questionnaire used in the first survey conducted in February–March 2020. Additional file 2: Appendix 2. Survey on stores management and environmental improvements associated with COVID-19 pandemic. This file was the questionnaire that was used in the second survey conducted in March 2021. These questionnaires were sent to 809 restaurants and bars that responded to the first survey and were still running in 2021. Additional file 3: Appendix 3. Survey on stores management and environmental improvements associated with COVID-19 pandemic. This file was the questionnaire used in the second survey conducted in March 2021. These questionnaires were sent to 2,800 restaurants and bars that were still in operation in 38 cities in Tokyo, Osaka, and Aomori prefectures, which were newly added to the second survey conducted in March 2021. Additional file 4: Appendix 4. How did we estimate national indoor smoking status in restaurants and bars? In this file, we describe the detailed method of the estimation of national indoor smoking status in restaurants and bars in Japan. Additional file 5: Supplementary Fig. 1. Signs explaining indoor smoking status, especially establishments that do not choose to be completely smoke-free indoors. This figure explains signs about indoor smoking status that restaurants and bars post on their storefronts. Additional file 6: Supplementary Fig. 2. Changes in indoor smoking status for all restaurants and bars before and after the enforcement of the smoke-free legislation. This figure shows the changing indoor smoking status in all restaurants and bars in this survey before and after the enforcement. Additional file 7: Supplementary Fig. 3. Changes in indoor smoking status for the unregulated restaurants and bars before and after the enforcement of the smoke-free legislation. This figure shows the more detailed changing indoor smoking status in the unregulated restaurants and bars in this survey before and after the enforcement than Fig. 2. Additional file 8: Supplementary Fig. 4. Indoor smoking status of planned before the enforcement and of actual status after the enforcement. This figure shows the indoor smoking status that the separated or smoking restaurants and bars were considering before the enforcement of smoke-free legislation and the indoor smoking status that they actually implemented after the enforcement. Additional file 9: Supplementary Fig. 5. Changes in indoor smoking status for the regulated restaurants and bars before and after the enforcement of the smoke-free legislation. This figure shows the more detailed changing indoor smoking status in the regulated restaurants and bars in this survey before and after the enforcement than Fig. 3. Additional file 10: Supplementary Fig. 6. National estimate of indoor smoking status in all restaurants and bars by regulation status before/after the enforcement. This figure shows the national estimation of indoor smoking status in both the regulated and unregulated restaurants and bars before and after the enforcement in Japan. Additional file 11: Supplementary Fig. 7-A. National estimate of indoor smoking status in restaurants and bars that mainly served meals before/after the enforcement. Supplementary Fig. 7-B. National estimate of indoor smoking status in all restaurants and bars that mainly served meals by regulation status before/after the enforcement. Supplementary Fig. 7-C. National estimate of indoor smoking status in restaurants and bars that mainly served alcohol before/after the enforcement. Supplementary Fig. 7-D. National estimate of indoor smoking status in all restaurants and bars that mainly served alcohol by regulation status before/after the enforcement. These figures represent the national estimation of indoor smoking status in restaurants or bars, based on the service categories and the regulated or unregulated by law. Additional file 12: Supplementary Table 1-A. Proportion of no smoking in restaurants and bars observed in the present study and calculated in the sensitivity analysis. Supplementary Table 1-B. Proportion of changing indoor smoking status in restaurants and bars observed in the present study and calculated in the sensitivity analysis. This table shows the proportion of “no smoking” and the proportion of changing indoor smoking status in restaurants and bars observed in this study and calculated in the sensitivity analysis. Additional file 13: Supplementary Table 2. Differences in characteristics of the unregulated restaurants and bars regarding changes in indoor. This table shows the detailed characteristics of the unregulated restaurants and bars. Additional file 14: Supplementary Table 3 (.xlsx): Differences in characteristics of the regulated restaurants and bars regarding changes in indoor. This table shows the detailed characteristics of the regulated restaurants and bars based on the changing indoor smoking status. Additional file 15: Supplementary Table 4. Smoking status of owners before the enforcement and changing indoor smoking status before and after the enforcement in the unregulated restaurants and bars. This table shows the smoking status of owners in the unregulated restaurants and bars before the enforcement of the law based on the changing indoor smoking status. Additional file 16: Supplementary Table 5-A. National estimate of no smoking (%) and separate smoking or smoking (%) by regulation before the enforcement (2019). Supplementary Table 5-B. National estimate of no smoking (%) and separate smoking or smoking (%) by regulation before the enforcement (2020). These tables show the results of national estimation of indoor smoking status in Japan before and after the enforcment of the law. Additional file 17: Supplementary Table 6-A. National estimate of no smoking (%) and separate smoking or smoking (%) by service category before/after the enforcement. Supplementary Table 6-B. National estimate of no smoking (%) and separate smoking or smoking (%) by service category and regulation by law before/after the enforcement. These tables show the results of natonal estimation of indoor smoking status in Japan based on the service category and the regulated or unregulated by law. Acknowledgements All authors thank the restaurants and bars that responded to our surveys and cooperated in our research, and Dr Julia Mortimer for her English language editing. In addition, all authors would also like to thank the Tobacco Control Group of the Ministry of Health, Labour and Welfare for their valuable advice on this study's considerations. Abbreviations Secondhand smoke World Health Organization Framework Convention on Tobacco Control Heated tobacco products Authors’ contributions AK, IM, and YI conceived the study and collected the data, and MN supervised the project. AK analyzed the data and all authors contributed to the interpretation of the results. AK wrote the report supported by all authors, and all authors read and approved the final manuscript. Funding This project was supported by a Health Labor Sciences Research Grant FA Program grant number 19FA1005. Data availability The datasets used during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The “Ethical Guidelines for Medical and Health Research Involving Human Subjects” were not applicable to this study as the subjects were restaurants and bars not human subjects, and ethics approval and informed consent was waved by institutional review board of Osaka University: Ethical Review Board Osaka University Hospital. In the first page of the questionnaire, we explained the purpose of this study, stated that cooperation in the survey was voluntary and considered as agreement to participate of the survey. In addition, we included the contact information so that we could respond to inquiries and refusals to respond. Information on each restaurant and bar was handled on personal computers with appropriate security measures such as setting passwords and installing security software, so that no disadvantage would arise from this study. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Additional file 1: Appendix 1. Survey on the Impact on Restaurants or Bars of Strengthened Passive Smoking Countermeasures. This file was the questionnaire used in the first survey conducted in February–March 2020. Additional file 2: Appendix 2. Survey on stores management and environmental improvements associated with COVID-19 pandemic. This file was the questionnaire that was used in the second survey conducted in March 2021. These questionnaires were sent to 809 restaurants and bars that responded to the first survey and were still running in 2021. Additional file 3: Appendix 3. Survey on stores management and environmental improvements associated with COVID-19 pandemic. This file was the questionnaire used in the second survey conducted in March 2021. These questionnaires were sent to 2,800 restaurants and bars that were still in operation in 38 cities in Tokyo, Osaka, and Aomori prefectures, which were newly added to the second survey conducted in March 2021. Additional file 4: Appendix 4. How did we estimate national indoor smoking status in restaurants and bars? In this file, we describe the detailed method of the estimation of national indoor smoking status in restaurants and bars in Japan. Additional file 5: Supplementary Fig. 1. Signs explaining indoor smoking status, especially establishments that do not choose to be completely smoke-free indoors. This figure explains signs about indoor smoking status that restaurants and bars post on their storefronts. Additional file 6: Supplementary Fig. 2. Changes in indoor smoking status for all restaurants and bars before and after the enforcement of the smoke-free legislation. This figure shows the changing indoor smoking status in all restaurants and bars in this survey before and after the enforcement. Additional file 7: Supplementary Fig. 3. Changes in indoor smoking status for the unregulated restaurants and bars before and after the enforcement of the smoke-free legislation. This figure shows the more detailed changing indoor smoking status in the unregulated restaurants and bars in this survey before and after the enforcement than Fig. 2. Additional file 8: Supplementary Fig. 4. Indoor smoking status of planned before the enforcement and of actual status after the enforcement. This figure shows the indoor smoking status that the separated or smoking restaurants and bars were considering before the enforcement of smoke-free legislation and the indoor smoking status that they actually implemented after the enforcement. Additional file 9: Supplementary Fig. 5. Changes in indoor smoking status for the regulated restaurants and bars before and after the enforcement of the smoke-free legislation. This figure shows the more detailed changing indoor smoking status in the regulated restaurants and bars in this survey before and after the enforcement than Fig. 3. Additional file 10: Supplementary Fig. 6. National estimate of indoor smoking status in all restaurants and bars by regulation status before/after the enforcement. This figure shows the national estimation of indoor smoking status in both the regulated and unregulated restaurants and bars before and after the enforcement in Japan. Additional file 11: Supplementary Fig. 7-A. National estimate of indoor smoking status in restaurants and bars that mainly served meals before/after the enforcement. Supplementary Fig. 7-B. National estimate of indoor smoking status in all restaurants and bars that mainly served meals by regulation status before/after the enforcement. Supplementary Fig. 7-C. National estimate of indoor smoking status in restaurants and bars that mainly served alcohol before/after the enforcement. Supplementary Fig. 7-D. National estimate of indoor smoking status in all restaurants and bars that mainly served alcohol by regulation status before/after the enforcement. These figures represent the national estimation of indoor smoking status in restaurants or bars, based on the service categories and the regulated or unregulated by law. Additional file 12: Supplementary Table 1-A. Proportion of no smoking in restaurants and bars observed in the present study and calculated in the sensitivity analysis. Supplementary Table 1-B. Proportion of changing indoor smoking status in restaurants and bars observed in the present study and calculated in the sensitivity analysis. This table shows the proportion of “no smoking” and the proportion of changing indoor smoking status in restaurants and bars observed in this study and calculated in the sensitivity analysis. Additional file 13: Supplementary Table 2. Differences in characteristics of the unregulated restaurants and bars regarding changes in indoor. This table shows the detailed characteristics of the unregulated restaurants and bars. Additional file 14: Supplementary Table 3 (.xlsx): Differences in characteristics of the regulated restaurants and bars regarding changes in indoor. This table shows the detailed characteristics of the regulated restaurants and bars based on the changing indoor smoking status. Additional file 15: Supplementary Table 4. Smoking status of owners before the enforcement and changing indoor smoking status before and after the enforcement in the unregulated restaurants and bars. This table shows the smoking status of owners in the unregulated restaurants and bars before the enforcement of the law based on the changing indoor smoking status. Additional file 16: Supplementary Table 5-A. National estimate of no smoking (%) and separate smoking or smoking (%) by regulation before the enforcement (2019). Supplementary Table 5-B. National estimate of no smoking (%) and separate smoking or smoking (%) by regulation before the enforcement (2020). These tables show the results of national estimation of indoor smoking status in Japan before and after the enforcment of the law. Additional file 17: Supplementary Table 6-A. National estimate of no smoking (%) and separate smoking or smoking (%) by service category before/after the enforcement. Supplementary Table 6-B. National estimate of no smoking (%) and separate smoking or smoking (%) by service category and regulation by law before/after the enforcement. These tables show the results of natonal estimation of indoor smoking status in Japan based on the service category and the regulated or unregulated by law. Data Availability Statement The datasets used during the current study are available from the corresponding author on reasonable request.