Environmental Disease

: 2016  |  Volume : 1  |  Issue : 2  |  Page : 65--76

Secondhand smoke: A comparison of exposure to bar employees in a state with smoking bans and a state without smoking bans

Oluwapese M Akinbobola1, Mac Crawford1, Qinghua Sun2,  
1 Division of Environmental Health Sciences, College of Public Health, The Ohio State University, Columbus, Ohio, USA
2 Division of Environmental Health Sciences, College of Public Health, The Ohio State University, Columbus, Ohio, USA; Shanghai Key Laboratory of Meteorology and Health, Shanghai Meteorological Bureau, Shanghai, China

Correspondence Address:
Qinghua Sun
Division of Environmental Health Sciences, College of Public Health, The Ohio State University, 424 Cunz Hall, 1841 Neil Avenue, Columbus, Ohio, USA


Background: Only 28 states have comprehensive laws prohibiting smoking in public places including all bars and restaurants. This project studied secondhand smoke (SHS) in bars in Covington, KY, where it is still legal to smoke indoors versus air quality in Cincinnati, OH bars where it is not legal to smoke indoors. Previous studies took samples from bars before and after smoking bans were put in place, but outdoor air quality in general could have changed in the areas over time. Materials and Methods: Air samples were taken in eight bars, four in Ohio where smoking is prohibited and four in Kentucky where smoking is permitted. All samples were taken on the same evening spending 30 min in each location with the instrument as close to the center of the room as possible. Results: Environmental conditions in Cincinnati, OH, and Covington, KY were very similar on the sampling evening. This along with the fact that the two cities are adjacent across the state line means that the bars in question should be subjected to similar outdoor air influences on air quality. However, the overall average time-weighted average (TWA) in Ohio was found to be 0.019 while the average TWA in Kentucky was 16 times higher at 0.303 mg/m 3 . Conclusion: Smoking bans have been shown to reduce the levels of exposure of SHS to the employees of bars and restaurants along with the patrons of these establishments. This reduces the risk of disease that may result from such exposure.

How to cite this article:
Akinbobola OM, Crawford M, Sun Q. Secondhand smoke: A comparison of exposure to bar employees in a state with smoking bans and a state without smoking bans.Environ Dis 2016;1:65-76

How to cite this URL:
Akinbobola OM, Crawford M, Sun Q. Secondhand smoke: A comparison of exposure to bar employees in a state with smoking bans and a state without smoking bans. Environ Dis [serial online] 2016 [cited 2018 Jan 19 ];1:65-76
Available from: http://www.environmentmed.org/text.asp?2016/1/2/65/185301

Full Text


The food and beverage service industry employs over 7.7 million people/year in the United States. Of these, there are 2,289,010 waiters and waitresses and 512,230 bartenders. [1] Among bartenders, 207,740 work in full-service restaurants and 153,870 work in drinking places which predominantly serve alcohol. The remaining bartenders work for civil and social organizations, traveler accommodations, or other amusement and recreation industries. Employment in this field varies by state but is highest in California, Florida, New York, Texas, and Pennsylvania [Figure 1]. Of these five states, only California and New York have fully comprehensive smoking bans [Figure 2]. Those bartenders working in the remaining states have a notable occupational health hazard since they are at risk of being exposed to secondhand smoke (SHS). While there is no way to find the exact amount of SHS that every bartender is exposed to, this study aims to find out the general increase in exposure that may result from working in a state which allows smoking versus a state which does not allow smoking in bars and restaurants.{Figure 1}{Figure 2}

Tobacco smoking is responsible for an estimated 467,000 deaths/year making it the leading cause of preventable death in the United States. [2] Smoking causes cardiovascular diseases, cancers, and other respiratory diseases. SHS is the result of nonsmokers being exposed to toxic compounds from others' smoking. SHS is a combination of sidestream smoke from the burning end of cigarettes and mainstream smoke which is exhaled from smokers. While tobacco smoke contains more than 4000 chemical compounds, SHS has more than 7000; of these chemicals, 250 are harmful and over 60 are carcinogenic (the American Cancer Society, 2011). [3] It is well-established that SHS, as well as active smoking, increases the risk of cardiovascular disease, respiratory illness, and lung cancer.

Even though the adverse health effects of SHS are well known, only 28 states have comprehensive smoking laws prohibiting smoking in public places including all bars and restaurants. Of interest, in this study, while Ohio has banned smoking in bars and restaurants [Table 1] and [Figure 2], Kentucky does not have indoor air restrictions on smoking [Table 2] and [Figure 3]. [4] Ohio ranks 35 th among the states in terms of prevalence of adult cigarette smokers with 20.1% of this population being current smokers. [5] Among those aged 12-17, Ohio ranks 46 th with 12.9% of this population being current smokers. Kentucky ranks 49 th among the states in terms of current adult cigarette smokers, with 25.2% of their adult population falling into this category. Among youth aged 12-17, Kentucky is 51 st among the states with 15.9% of this population being current smokers. In general, those working in buildings such as hospitals and schools are subject to self-imposed indoor smoking bans. However, 10.7% of adults who work indoors in Kentucky report being exposed to smoking in their work area within the preceding 2 weeks, thus making Kentucky rank 48 th among the states for workplace SHS exposure. This ranking is due largely to the fact that Kentucky does not have any laws prohibiting smoking in public places.{Figure 3}{Table 1}{Table 2}

Inhalation of SHS is among the various occupational hazards associated with waiting tables and bartending. This is a hazard in places that still allow smoking either inside or in attached designated-smoking sections and patios. Several studies have been done to relate working in bars and restaurants to markers of SHS exposure. A study conducted in fifty bars in Greece associated the presence of particulate matter smaller than 2.5 μm (PM 2.5 ) within the bar to cotinine levels of nonsmoking employees. [6] Cotinine is the main metabolite of nicotine, and urinary cotinine levels varied directly with measurements of PM 2.5 . Both urinary cotinine and PM 2.5 were dependent on the number of cigarettes being smoked/100 m 3 of venue volume. All of the venues sampled had opened doors and windows allowing air exchange with the outdoors, but the association between smoke and cotinine levels was still measurable. This demonstrated that the open doors and windows did little to dissipate environmental tobacco smoke.

Many studies have been done to test the change in air quality of bars before and after smoking bans have been put in place. A study in Kentucky found PM 2.5 levels in 62 hospitality venues to be an average of 161 μg/m 3 , which is 6.4 times World Health Organization (WHO) guidelines. [7] They selected locations as far as possible away from direct puffs of smoke indicating that this smoke was diffusing throughout the building. The locations tested had either full or partial bans put in place, and the researchers found that places with partial bans still had 12.5 times the indoor air pollution than places with comprehensive laws.

Testing for particulate matter in locations allowing smoking versus locations without smoking has been done for 15 casinos in Reno, Nevada. [8] This found that the PM 2.5 levels in nonsmoking casinos was 16 μg/m 3 . While the levels they measured in smoking casinos were not much higher at 33 μg/m 3 , this is still more than double that of the nonsmoking casinos. They used the data they collected in collaboration with previous studies measuring particulate matter levels in casinos in Las Vegas, Atlantic City, California, and Pennsylvania. This showed levels of particulate matter ranging from 18.5 to 205 μg/m 3 . This is a wide range of data and was highly dependent on smoker density. Overall, they found that ventilation was an inadequate control method for reducing respirable particles. However, the reduction of smoker density was found to be effective in reducing PM 2.5 levels. Therefore, as much as a casino may try to dissipate the concentration of smoke in the air through ventilation, the only truly effective way to do so would be to decrease or eliminate the actual number of smokers in the building.

One study done in Boston using seven pubs found that while smoking was allowed, ventilation was inadequate to remove PM 2.5 from venues. [9] This is probably stemming from the fact that ventilation is not dependent on occupancy so as the number of people in a bar smoking increased the percent of total SHS removed decreased. As expected, respirable particles and carcinogen pollution were significantly reduced after bans were put in place and smoking in bars had ceased. They correlated this reduction with information about business in the bars and found that while smoking was eliminated, patronage at the bars was not decreased.

In Austin, TX, it was again found that while smoking was eradicated, occupancy levels were not decreased by the implantation of smoking bans. [10] Resistance to laws may come from establishments worried about losing patronage and revenue; however, the research has shown that this is not proven to occur. Smoking bans can be expected to improve the health of employees and customers without making a harmful economic impact on the food and beverage industry.

The only effective measure against SHS exposure would be the further use of bans where they presently do not exist. Currently, only 28 states have laws restricting smoking in all bars and restaurants [Figure 2]. Many states have exemptions, for example, Florida allows smoking in places that have food as <10% of their sales, which allows smoking in most bars and nightclubs. In some states, laws vary by city and country, and Native American territory is exempt from following smoking bans. More widespread bans are needed to protect the employees of restaurants and bars from the adverse effects of SHS.

The occupational hazards of bartending are a significant public health issue because they consist of a large sector of employment within the United States. Many of the hazards of bartending are also dangers to waiters, waitresses, and others employed in the food and beverage industry. With 7.7 million employees, this sector makes up about 2.5% of the US population. Environmental hazards, such as noise and smoke, within a bar affect patrons as well as employees. Workers are at increased risk because of their extended exposure, but the public should be concerned as well. Since a single exposure to a carcinogen can cause a gene mutation and lead to cancer, anybody who comes in contact with these hazards for any amount of time may potentially suffer long-term health consequences. There are almost 50,000 deaths each year in the United States because of environmental tobacco smoke. [11] Healthcare expenditures, morbidity, and mortality due to SHS exposure cost the United States more than $10 billion/year. [12]

Due to the immense health and economic impact of SHS, more strict laws should be established nationwide. In the service industry, smoking bans would benefit the employees as well as the patrons of such businesses. The goal of this study is to find the excess exposure to respirable particles that is experienced by employees working in bars that permit smoking versus bars that do not permit smoking. It is hypothesized that bars that permit smoking indoors will have remarkably higher levels of PM 2.5 than bars in which smoking is not authorized indoors.


Eight bars were chosen for sampling, four in Cincinnati, OH and four in Covington, KY. These two cities sit adjacent on the Ohio/Kentucky state line separated only by the Ohio River [bar locations shown in [Figure 4]. Bars were matched based on theme (Irish, British, Gay, and Dive) and relative size so that one of each type in Ohio was paired with one in Kentucky. This was done to make a reasonable comparison between working in similar locations in a state which prevent smoking in bars versus a state which does not have smoke-free laws. Sampling was performed on Friday, July 13, 2012. Start and end times were taken at each location. The total number of people present along with the number of people smoking was counted in each bar. In addition, various factors were noted which could change the air quality in a location such as the presence of fans or an open door in one bar.{Figure 4}

Particulate matter was measured using a Thermo MIE pDR-1000AN Personal DataRAM for Personal Data-logging Real-time Aerosol Monitor (Thermo Andersen Inc., Smyrna, GA). This model is a passive air sampler meaning that air reaches the sensing chamber through "convection, diffusion, and adventitious air motion"; [13] this process is adventitious in that there is no pump to actively suck air into the machine. Under the protective bumper, the air enters through two slot-shaped inlets, one on the front and one on the back. This instrument measures the respirable fraction of airborne dust, smoke, fumes, and mists of indoor environments termed PM 2.5 . The span of measurement for the instrument is wide ranging from 0.001 to 400 mg/m 3 .

The instrument was zeroed with particle-free air before each sample being taken using the hand-inflatable zero air pouch included with the device. The instrument has been factory calibrated, and zeroing enables comparison between these calibrated factory settings and the current optical background sensed during the process. The instrument was deemed ready for use once the screen read "Calibration OK," indicating that the instrument's zeroing agreed with the original factory settings. The instrument was placed upright at table height in each location so that the two air inlets were unobstructed. Every effort was made to place the instrument as close to the center of the room as possible. However, this was not possible in crowded bars where center tables were not available. In addition, an effort was made to setup the monitor away from air conditioner outlets and fans. Samples were taken in each bar for 30 min with automated readings every 20 s. The instrument also measured the time-weighted average (TWA) for the sampling period which is what the average reading would be for an 8 h period.


On July 13, 2012, the mean temperature in Cincinnati, OH was 76°F with a minimum of 70°F and a maximum of 84°F. [14] The dew point was 67°F, and there was no precipitation. Samples were taken in this area from 7:19 to 10:35 pm. The 5 min outdoor air sample gave a reading of 0.020 mg/m 3 . There were no people smoking inside any of the Ohio locations. TWAs for the 30 min sampling period ranged from 0.007 to 0.031 mg/m 3 [Figure 3]. The overall TWA for the four locations was 0.019 mg/m 3 . Summary statistics from the samples are shown in [Table 3]. The detailed results are shown in [Figure 5] and [Table 4].{Figure 5}{Table 3}{Table 4}

On July 13, 2012, the mean temperature in Covington, KY was 78°F with a minimum temperature of 70°F and maximum of 85°F. [15] The dew point was 65°F and there was no precipitation. Samples were taken in this area from 11:21 pm to 2:14 am. The 5 min outdoor air sample gave a reading of 0.031 mg/m 3 . [Table 3] gives the exact number of smokers in each location, but numbers varied from 2 to 9 people. These are the maximum number of lit cigarettes during the sampling period. At various points, there were no people smoking within these establishments. TWAs for these smoking locations ranged from 0.161 to 0.659 mg/m 3 [Figure 3]. This gave an overall average of 0.303 mg/m 3 .

The outdoor air samples taken differed by 0.011 mg/m 3 with the higher reading being in Kentucky. The overall TWAs for each state differed by 0.284 mg/m 3 with the TWA of Kentucky almost 16 times higher than that in Ohio. All of the Kentucky TWA measurements were higher than even the greatest reading in Ohio. There was a big difference in samples between the two states with the lowest reading in Kentucky (0.0161 mg/m 3 ) being more than five times higher than the highest reading in Ohio (0.031 mg/m 3 ).


Weather conditions, including temperature and dew point, for Cincinnati, OH and Covington, KY were similar on the night of sampling, July 13, 2012. The bars used for sampling are a maximum of 7.6 miles apart. Due to similar weather conditions and proximity, these sampling locations should be subjected to comparable levels of outdoor air contaminants. Therefore, differences in particulate matter levels indoors can be assumed to be the result of contaminants originating from within the establishments.

The differences in outdoor air samples of 0.011 mg/m 3 could be due to the fact that the sample in Ohio was taken with only one person in proximity to the instrument while the sample in Kentucky was taken on the side of the street with three people near the instrument and various other people walking by. In addition, the Ohio sample was taken near a park beside the first bar whereas the sample in Kentucky was taken in closer proximity to a street full of bars. Furthermore, these samples were taken for only a period of 5 min; further precision may have been found in taking outdoor samples for longer periods of time.

Upon entering the British pub and first sampling location in Kentucky, there were no people smoking. However, the readings were initially more than four times higher than in the comparable Ohio location. This demonstrates that the particulate matter levels remain high even when there are not people actively smoking in the venue. While in this bar, there were up to a maximum of two people smoking with not much variation in the PM 2.5 readings. This demonstrates that the particles remain stagnant. Therefore, people working in such a location are being exposed to elevated levels of PM 2.5 even when there no smokers present.

The Irish bar and third Kentucky location had an open door, open window, and two ceiling fans. However, the readings at 0.167 mg/m 3 were almost eight times higher than the readings in the comparable Ohio location which were at 0.021 mg/m 3 . The ceiling heights in both locations were similar. The open door and window should have allowed the particles to disperse. The Ohio Irish bar had two sets of doors and no windows that could be opened allowing minimal air exchange between the indoor and outdoor environments. In addition, the nonsmoking bar had many more people present (32) than the smoking location (19). The additional people would have contributed to higher levels of respirable particles. Even with these considerations of air exchange and number of people, the smoking location had much higher levels of respirable particles than the smoke-free location.

The dive bar in Ohio had overall the lowest TWA for all of the sampling locations at 0.007 mg/m 3 . The comparable dive bar in Kentucky had the second highest level of all of the bars at 0.226 mg/m 3 . These locations had similar ceiling heights and areas. They also had comparable numbers of people present in each. However, the smoking dive bar location had respirable particle levels more than 32 times higher than the smoke-free dive bar. This difference was present even though there was only a maximum of three people smoking while samples were being taken.

The gay bar in Kentucky was visibly the smokiest out of all of the bars visited. This was demonstrated by the fact that it had the highest levels of respirable particles with a TWA of 0.659 mg/m 3 . This is almost 44 times the levels in the comparable smoke-free gay bar which were at 0.015 mg/m 3 . The ceiling heights and areas of the bars were similar. However, there were many more people present in the smoking bar (68 people) than in the nonsmoking bar (19 people). In addition, this Kentucky location had the highest number of active smokers at nine. There were three ceiling fans in this location, but this did not seem to make any difference in the PM 2.5 levels.

When comparing the most populated smoking location, the Kentucky gay bar, with the most populated nonsmoking location, the Ohio British pub, both were of similar size and had ceiling fans. In addition, these two locations had the highest levels of respirable particles in their respective states. However, this smoking location still had levels over 21 times higher than this smoke-free location.

The WHO sets a 24-h exposure level of 25 μg/m 3 for PM 2.5 . [16] This level was exceeded in all four bars in which smoking was allowed and in one of the bars in which smoking was not allowed. However, in the nonsmoking bar in which this level was exceeded, if there was a background of 10 μg/m 3 to which the person was exposed while not working (i.e., from outdoor air particles) then an employee working an 8-h shift would only have a 24-h exposure level of 17 μg/m 3 ([31 μg/m 3 × 8 h + 10 μg/m 3 × 16 h]/24). Therefore, they would be under the WHO's limit for a typical day. However, when looking at the bar where smoking was permitted with the lowest levels of PM 2.5 and using the same daily parameters, there is a 24-h exposure level of 60 μg/m 3 ([161 μg/m 3 × 8 h + 10 μg/m 3 × 16 h]/24). This value is more than twice the WHO's limit. As a measure of comparison, the smokiest bar would be at 226 μg/m 3 ([659 μg/m 3 × 8 h + 10 μg/m 3 × 16 h]/24), which is more than nine times the WHO's limit. These numbers indicate that all of the smoking establishments put employees at an increased risk for smoking related illness.

A study done in bars and restaurants in Minnesota compared levels of particulate matter throughout the week and concluded that Friday evening had the highest levels. [17] Therefore, the samples taken may be higher than typical values in the venues. Therefore, these findings may give occupational exposures that are not standard for all employees. However, to be protective of health, the highest potential exposure levels should be taken into consideration.

In addition to respirable particles, working in bars and restaurants where smoking is permitted has been shown to increase employees' exposure to volatile organic compounds such as acetaldehyde, 1,4-dichlorobenzene, and perchloroethylene. [18] Exposure to acetaldehyde can cause degeneration of olfactory epithelium and has the potential to cause nasal squamous cell carcinoma and adenocarcinoma. [19] 1,4-dichlorobenzene has been shown to cause increased liver weights in males. Inhalation of perchloroethylene has been shown to cause neurotoxicity. It is also likely to be a human carcinogen producing hepatocellular adenomas and carcinomas. Therefore, whatever can be done to decrease the risk of such employment should be to protect the 2.5% of Americans which work in this industry.

One consideration in applying smoking bans in public places is revenue. However, bartenders in Ohio and Kentucky currently earn similar wages [Figure 6]. Therefore, those bartenders working on the Kentucky side of the state line are being exposed to harmful compounds without any monetary benefits. {Figure 6}Research done in California, the first state to implement smoke-free restaurant and bar laws, showed that the laws may have attracted more nonsmokers and only deterred a few smokers. [20] They came to this conclusion since bar revenues increased after the instatement of laws in both restaurants and bars. For [Figure 2] and [Figure 6], when viewed side by side, it can be seen that six of the seven top wage-earning states for bartenders have 100% smoke-free laws indicating that employees in this field are not being harmed by smoking bans and may even be helped financially through these laws. The topic of revenue was further explored in 10 cities in Minnesota. [21] The results of this study were similar to those in California in that there were slightly higher revenues in bars and restaurants subject to partial or full clean indoor air (CIA) policies than in bars and restaurants with no CIA policy. Revenues are important in law-making, but it has been shown that this should not be a concern when making smoking prohibitions.

The state of Florida currently has smoking bans in private workplaces and in restaurants. However, bars and the outdoor portions of restaurants do not fall under the law and are not smoke-free. [4] A study conducted in this state consisting of 1858 participants aged 18-24 assessed historical and current exposure to environmental tobacco smoke. [22] These data were then compared to self-reported respiratory symptoms that each person had experienced in the month before being interviewed. Fifteen percent of respondents reported being exposed to SHS at their places of employment. Sixty-four percent of participants reported being exposed to SHS at a bar, nightclub, or restaurant in the previous month. This indicates that almost two-third of the people in this age group are being exposed to the harmful chemicals in cigarettes just by going to locations that are not subject to smoking bans in this state. Even though Florida is considered a "clean" indoor air state by having partial bans on cigarette smoking, further laws could help appreciably reduce exposure to the toxic chemicals contained in cigarettes. Lee et al. also found that respiratory symptoms, such as shortness of breath, wheezing, repeated sneezing, and coughing episodes, increased with reported SHS exposure. [22] Additional smoking laws could lower these respiratory symptoms to people in Florida and other states which currently do not have comprehensive smoking bans.

Among a study of 4223 college students in North Carolina, 65% reported being exposed to environmental tobacco smoke in a bar or restaurant during the 7 days before being surveyed. [23] Exposure is not just a problem for employees of bars and restaurants which permit smoking; the patrons of these establishments are also at risk for the various adverse health effects that result from SHS.

Smoking regulations have been shown to reduce exposure to environmental tobacco smoke in Massachusetts. [24] This study found that areas with weak tobacco regulations had the lowest level of reported SHS exposure with 55.8%. Medium-level regulations yielded 70.2% being exposed. Areas with strong regulations found that 81.2% of respondents not being exposed to environmental tobacco smoke. This further supports the idea that smoking bans will not only benefit the health of the employees but of patrons as well.

Using the 1992-2002 National Health and Nutrition Examination Survey, it was found that 39.1% of all nonsmoking adults had SHS exposure. [25] This was dramatically decreased to 12.5% in counties with extensive smoke-free law coverage. Counties with limited coverage had exposure levels around 35.1%, and counties with no coverage had exposure levels at 45.9%. Overall, nonsmoking women living in counties with extensive smoke-free law coverage were 0.19 times as likely to be exposed to SHS as women living in counties without smoke-free laws. For men, the decrease in exposure was even greater with them being 0.10 times as likely to be exposed to SHS when residing in an area with smoking bans. Therefore, there is a significant correlation between the strength of smoke-free laws and SHS exposure among nonsmoking adults.

Support for restaurant and bar CIA laws has been demonstrated. A recent study to investigate The Oklahoma Tobacco Stops with Me campaign impact from 2007 to 2015 on SHS policy and risk attitudes found that Oklahomans demonstrated significant increases in supporting smoke-free bars (23.7-55%); reporting beliefs that SHS is harmful, causes heart disease, and causes sudden infant death; and reporting they are very likely to ask someone not to smoke nearby (45-52%), which indicates an increasing public perception of SHS with well-documented SHS risks, especially successful in women. [26] A study of 2044 young adults in Minnesota found that 66% supported such laws in restaurants and 40% supported these laws in bars and clubs. [27] Those people living and working in locations with current smoking bans were more likely to support the bans than those who were not subjected to the laws. Therefore, even though there may be some resistance to implementing such laws, people tend to support them once they are in place. This is further demonstrated in a study in El Paso, Texas which surveyed people before and after the implementation of a smoking ban in bars and restaurants. [28] A year after the smoking bans were put in place, 78.5% of those surveyed supported the ordinance, while only 10.9% opposed it. The remaining people said they had no opinion about the law. In addition to having the support of the majority of people living in the area, the laws also resulted in declines in adult smoking from 22.1% in 1996 to 17.3% in 2002. Evidence of smoking bans being effective is furthered by the fact that in this same El Paso, TX study, the number of employed waiters and waitressed increased from before the implementation of smoke-free laws to after, and there was no statistically significant change in bar and restaurant revenues after implementation.

Overall, various studies have shown support of smoking bans. Furthermore, research has shown that these bans do not impact revenue and jobs in the food and beverage industry. However, these bans do decrease exposure to SHS and are well-supported in the communities where they currently exist.


Many studies have been done that find a decrease in PM 2.5 after smoking bans have been put in place. This study furthers these results since it shows the vast difference in respirable particles in bars within the same metropolitan area which are subject to different smoking restrictions. Since bars subject to smoking restrictions and those without such regulation were tested on the same evening, outdoor influences were considered to be similar and to have comparable influences on the indoor air quality. Therefore, differences in particulate levels were assumed to originate from within the establishments.

The immense difference between the particulate levels in bars subject to smoking regulations to those that are not shows that such regulations do make a difference in indoor air quality. The patrons and employees of locations without comprehensive smoking bans are being put at risk of various adverse health effects simply because of the state or local they are employed in. With only 28 states having comprehensive laws preventing smoking in public places, more states need to enact smoking bans in order to protect the health of their employees.

Financial support and sponsorship

This work was supported by the NIH grant ES018900 to Dr. Sun.

Conflicts of interest

There are no conflicts of interest.


1Bureau of Labor Statistics, U.S. Department of Labor, Occupational Outlook Handbook, 2016-17 Edition, Food and Beverage Serving and Related Workers. Available from: [Last accessed on 2016 May 15].
2Danaei G, Ding EL, Mozaffarian D, Taylor B, Rehm J, Murray CJ, et al. The preventable causes of death in the United States: Comparative risk assessment of dietary, lifestyle, and metabolic risk factors. PLoS Med 2009;6:e1000058.
3American Cancer Society. Secondhand Smoke: What is secondhand smoke. Learn about cancer. Retrieved from: . [Last accessed on 2016 May 15].
4Centers for Disease Control and Prevention. State Tobacco Activities Tracking and Evaluation (STATE) System; 2012. Available from: https://www.apps.nccd.cdc.gov/statesystem. [Last accessed on 2016 May 15].
5Centers for Disease Control and Prevention. Tobacco control state highlights 2010. Atlanta: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2010.
6Karabela M, Vardavas CI, Tzatzarakis M, Tsatsakis A, Dockery D, Connolly GN, et al. The relationship between venue indoor air quality and urinary cotinine levels among semiopen-air café employees: What factors determine the level of exposure? J Aerosol Med Pulm Drug Deliv 2011;24:35-41.
7Lee K, Hahn EJ, Robertson HE, Lee S, Vogel SL, Travers MJ. Strength of smoke-free air laws and indoor air quality. Nicotine Tob Res 2009;11:381-6.
8Repace JL, Jiang RT, Acevedo-Bolton V, Cheng KC, Klepeis NE, Ott WR, et al. Fine particle air pollution and secondhand smoke exposures and risks inside 66 US casinos. Environ Res 2011;111:473-84.
9Repace JL, Hyde JN, Brugge D. Air pollution in Boston bars before and after a smoking ban. BMC Public Health 2006;6:266.
10Waring MS, Siegel JA. An evaluation of the indoor air quality in bars before and after a smoking ban in Austin, Texas. J Expo Sci Environ Epidemiol 2007;17:260-8.
11Centers for Disease Control and Prevention. Smoking & Tobacco Use; 2011. Available from: http://www.cdc.gov/tobacco/data_statistics/fact_sheets/fast_facts/. [Last accessed on 2016 May 15].
12Centers for Disease Control and Prevention. Smoking & Tobacco Use: Tobacco-related Mortality; 2011. Available from: http://www.cdc.gov/tobacco/data_statistics/fact_sheets/health_effects/tobacco_related_mortality/. [Last accessed on 2016 May 15].
13Thermo Electron Corporation. Model pDR-1000AN/1200 personalDATARAM Particulate Matter Instruction Manual; 2005. Available from: http://www.trs-environmental.com/Specs-Manuals/TEI_PDR_1000AN_Manual.pdf. [Last accessed on 2016 May 15].
14Weather Underground. History for Cincinnati, OH; 2012. Available from: http://www.wunderground.com/history/airport/KLUK/2012/7/13/DailyHistory.html. [Last retrieved on 2012 Jul 29].
15Weather Underground. History for Covington, KY; 2012. Available from: http://www.wunderground.com/history/airport/KCVG/2012/7/13/DailyHistory.html?req_city=NA & req_state=NA and req_statename=NA. [Last retrieved on 2012 Jul 29].
16World Health Organization. WHO Air quality guidelines for particulate matter, ozone, nitrogen dioxide and sulfur dioxide. Global update 2005. Summary of risk assessment. Geneva, Switzerland: World Health Organization; 2006.
17Bohac DL, Hewett MJ, Kapphahn KI, Grimsrud DT, Apte MG, Gundel LA. Change in indoor particle levels after a smoking ban in Minnesota bars and restaurants. Am J Prev Med 2010;39 6 Suppl 1:S3-9.
18Loh MM, Houseman EA, Levy JI, Spengler JD, Bennett DH. Contribution to volatile organic compound exposures from time spent in stores and restaurants and bars. J Expo Sci Environ Epidemiol 2009;19:660-73.
19United States Environmental Protection Agency. Integrated Risk Information System; 2012. Available from: http://www.cfpub.epa.gov/ncea/iris/index.cfm?fuseaction=iris.showSubstanceList and list_type=alpha. [Last accessed on 2016 May 15].
20Cowling DW, Bond P. Smoke-free laws and bar revenues in California - the last call. Health Econ 2005;14:1273-81.
21Collins NM, Shi Q, Forster JL, Erickson DJ, Toomey TL. Effects of clean indoor air laws on bar and restaurant revenue in Minnesota cities. Am J Prev Med 2010;39 6 Suppl 1:S10-5.
22Lee DJ, Dietz NA, Arheart KL, Wilkinson JD, Clark JD 3 rd , Caban-Martinez AJ. Respiratory effects of secondhand smoke exposure among young adults residing in a "clean" indoor air state. J Community Health 2008;33:117-25.
23Wolfson M, McCoy TP, Sutfin EL. College students′ exposure to secondhand smoke. Nicotine Tob Res 2009;11:977-84.
24Albers AB, Siegel M, Cheng DM, Rigotti NA, Biener L. Effects of restaurant and bar smoking regulations on exposure to environmental tobacco smoke among Massachusetts adults. Am J Public Health 2004;94:1959-64.
25Pickett MS, Schober SE, Brody DJ, Curtin LR, Giovino GA. Smoke-free laws and secondhand smoke exposure in US non-smoking adults, 1999-2002. Tob Control 2006;15:302-7.
26White AH, Brown-Johnson GG, Martinez SA, Paulson S, Beebe LA. Oklahoma "Tobacco Stops with Me" media campaign effects on attitudes toward secondhand smoke. J Okla State Med Assoc 2015;108:583-8.
27Bernat DH, Klein EG, Fabian LE, Forster JL. Young adult support for clean indoor air laws in restaurants and bars. J Adolesc Health 2009;45:102-4.
28Reynolds JH, Hobart RL, Ayala P, Eischen MH. Clean indoor air in El Paso, Texas: A case study. Prev Chronic Dis 2005;2:A22.