Article Text

Thirdhand smoke and exposure in California hotels: non-smoking rooms fail to protect non-smoking hotel guests from tobacco smoke exposure
  1. Georg E Matt1,
  2. Penelope J E Quintana2,
  3. Addie L Fortmann3,
  4. Joy M Zakarian4,
  5. Vanessa E Galaviz5,
  6. Dale A Chatfield6,
  7. Eunha Hoh2,
  8. Melbourne F Hovell7,
  9. Carl Winston8
  1. 1Department of Psychology, San Diego State University, San Diego, California, USA
  2. 2Graduate School of Public Health, San Diego State University, San Diego, California, USA
  3. 3San Diego Joint Doctoral Program in Clinical Psychology, San Diego State University/University of California, San Diego, California, USA
  4. 4San Diego State University Research Foundation, San Diego, California, USA
  5. 5Department of Environmental and Occupational Health Sciences, University of Washington, Seattle, Washington, USA
  6. 6Department of Chemistry, San Diego State University, San Diego, California, USA
  7. 7Center for Behavioral Epidemiology and Community Health, Graduate School of Public Health, San Diego State University, San Diego, California, USA
  8. 8Department of Hotel and Tourism Management, San Diego State University, San Diego, California, USA
  1. Correspondence to Dr Georg E Matt, Department of Psychology, San Diego State University, San Diego CA 92182-4611, USA; gmatt{at}mail.sdsu.edu.

Abstract

Introduction This study examined tobacco smoke pollution (also known as thirdhand smoke, THS) in hotels with and without complete smoking bans and investigated whether non-smoking guests staying overnight in these hotels were exposed to tobacco smoke pollutants.

Methods A stratified random sample of hotels with (n=10) and without (n=30) complete smoking bans was examined. Surfaces and air were analysed for tobacco smoke pollutants (ie, nicotine and 3-ethynylpyridine, 3EP). Non-smoking confederates who stayed overnight in guestrooms provided urine and finger wipe samples to determine exposure to nicotine and the tobacco-specific carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone as measured by their metabolites cotinine and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL), respectively.

Findings Compared with hotels with complete smoking bans, surface nicotine and air 3EP were elevated in non-smoking and smoking rooms of hotels that allowed smoking. Air nicotine levels in smoking rooms were significantly higher than those in non-smoking rooms of hotels with and without complete smoking bans. Hallway surfaces outside of smoking rooms also showed higher levels of nicotine than those outside of non-smoking rooms. Non-smoking confederates staying in hotels without complete smoking bans showed higher levels of finger nicotine and urine cotinine than those staying in hotels with complete smoking bans. Confederates showed significant elevations in urinary NNAL after staying in the 10 most polluted rooms.

Conclusions Partial smoking bans in hotels do not protect non-smoking guests from exposure to tobacco smoke and tobacco-specific carcinogens. Non-smokers are advised to stay in hotels with complete smoking bans. Existing policies exempting hotels from complete smoking bans are ineffective.

  • Secondhand smoke
  • Public policy
  • Environment

Statistics from Altmetric.com

Introduction

In response to the preferences of the consumer and the well-documented harmful effects of secondhand smoke (SHS) exposure,1 ,2 a majority of states in the USA and a growing number of countries worldwide have completely banned smoking in restaurants and bars.3 As of 2 January 2013, there are 30 states in the USA with laws banning smoking in restaurants and bars.4 Similarly, strict smoking bans are in place for restaurants and bars in Canada, Uruguay, Australia, New Zealand, Thailand and several European countries.5

Complete smoking bans for all hotel rooms have been legislated in several countries, including Bermuda, Brazil, China, Hong Kong, Indonesia, Kuwait and South Korea. In contrast, complete smoking bans in hotels and motels are uncommon in the USA—even in states that ban smoking in restaurants and bars. As of 2 January 2013,4 only Indiana, Michigan, North Dakota and Wisconsin prohibited smoking in hotel and motel rooms; all other states allowed hotels and motels to designate a certain percentage of guestrooms as smoking rooms. Legislation banning smoking in enclosed workplaces in California (but exempting a majority of hotel guestrooms) was enacted in 1995.6 Specifically California allows smoking in up to 65% of the guestrooms of hotels and motels, in up to 50% of the lobby area and in meeting and banquet rooms except during functions that involve food or beverage service.

Although public policies in most states of the USA continue to exempt hotels from complete indoor smoking bans, two prominent hotel brands and numerous independent hotels have implemented comprehensive smoking bans for their properties. In 2006, Westin became the first hotel chain to ban smoking in all their properties in the USA, Canada and the Caribbean,7 but not in their overseas properties. Shortly thereafter, Marriot brand hotels in the USA and Canada adopted complete smoking bans.8 Subsequently, Walt Disney, Sheraton, Wyndham Hotels and Resorts and Comfort Suites enacted similar policies.9 However, Hilton, InterContinental Hotel Group and Starwood (parent company of Westin, Sheraton and Starwood) and many other hotels continue to offer designated smoking rooms. According to data collected by the American Automobile Association on 31 000 hotels, motels and other lodgings, 12 900 properties (42%) were smoke free as of February 2011. This represents a 51% increase from November 2008.9 ,10

Notably, research has shown that non-smokers may be exposed to tobacco smoke toxicants even if a smoker is not in the same room or apartment.11 The migration of tobacco smoke has been studied in detached homes and multi-unit housing but not in hotels.12–14 Also known as ‘neighbour smoke’, in multi-unit housing, tobacco smoke easily moves throughout an entire building via doors, hallways, ventilation systems, plumbing and electric ducts and windows.12–16

Research has also demonstrated that non-smokers may be exposed to tobacco smoke toxicants even if the last cigarette smoked in a room was extinguished weeks or months prior.11 Dubbed thirdhand smoke (THS), the long-term pollution of indoor environments in which cigarettes have been regularly smoked has been studied in controlled laboratory environments,17–19 private homes,20 ,21 private cars22 ,23 and rental cars,24 but it has not been studied in hotels. Although SHS is a mixture of the sidestream smoke (ie, smoke emitted from the burning cigarette, pipe or cigar) and the mainstream smoke exhaled from the lungs of smokers,7 THS consists of tobacco smoke constituents that (1) remain on surfaces and in dust, or (2) are re-emitted back into the gas phase from sorbed reservoirs, or (3) are re-suspended particles from dust deposits or (4) react with oxidants and other compounds to yield secondary pollutants.7 SHS and THS coexist in the air during the early period of THS formation and in contaminated environments in which smoking takes place.

In combination, these research findings suggest that the existing smoke-free exemptions in California hotels make it virtually impossible to protect a non-smoking guest who stays in a designated smoking room from tobacco smoke exposure—even if no one smokes during their stay. This is because smoking hotel rooms become reservoirs of tobacco smoke toxicants that accumulate in carpets, dust, upholstery, mattresses, curtains and furniture, penetrate wallpaper and paint, and are even stored in drywall.17–19 Existing exemptions even make it difficult to protect non-smoking guests who stay in a non-smoking room of a hotel that allows smoking in other rooms. This is because similar to a multi-unit housing building, tobacco smoke cannot be confined to a hotel room but may spread to adjacent and more distant non-smoking rooms, hallways, ventilation systems, windows and utility ducts.

Existing research on tobacco control policies and tobacco smoke exposure in hotels is limited. Bialous et al25 have shown how the tobacco industry in Brazil established partnerships with hotel associations to prevent creating 100% smoke-free environments. Hotel guests surveyed in Australia expressed concern that their health was adversely affected by other people's smoke in the hotel,26 and a survey of physicians in Sweden showed that 82% ask for smoke-free hotel rooms.27 Based on the findings of the INTERHEAT study, Peters recommends eliminating the hotel exemption from smoking prohibition in the Netherlands as an important step to further reduce myocardial infarction caused by tobacco smoke.28 Although there is a large body of research on SHS exposure in restaurants and bars,29–36 and there is small body of research on exposure of hotel and other hospitality workers,35 ,37–45 there is no research on THS pollution of hotel rooms and THS exposure of hotel guests.

This study examined THS pollution of guestrooms and hallways of hotels with and without smoking bans, using validated chemical markers of tobacco smoke pollutants (ie, nicotine and 3 ethynylpyridine (3EP)) on surfaces and in the air of hotel rooms. To determine whether non-smokers staying overnight in hotels without complete smoking bans (ie, non-smoking or smoking guestrooms) were exposed to THS, we examined two validated biomarkers of tobacco smoke exposure: cotinine, the major metabolite of nicotine, and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL), a metabolite of the tobacco-specific carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK). Finally, we examined the extent to which tobacco smoke pollutants present in the air and on surfaces in guestrooms were associated with biological exposure measures.

Methods

Sample

A listing of 600 randomly selected lodging properties in San Diego County was obtained from a trade association (California Hotel and Lodging Association) in April 2009. Properties were excluded if they were located more than 20 miles from the study office, were not hotels or motels available for nightly rentals to the general public (eg, a naval base property and properties requiring >1 night minimum stay). Following Expedia.com's rating system, hotels were classified as ‘budget’ if they were rated with 1, 1.5 or 2 stars and ‘midscale’ if they were rated with 2.5 or 3 stars. For budgetary reasons, we excluded upscale and luxury hotels rated with more than 3 stars by Expedia.com. Hotels with fewer than 55 rooms were classified as ‘small’ and those with 55 or more rooms were classified as ‘large’. Research assistants examined the hotels’ websites and called the properties when necessary to determine whether or not they offered designated smoking guestrooms, thus classifying each as a hotel with or without a complete indoor smoking ban. Table 1 shows the total number of guestrooms and the percentage of smoking rooms of the hotels examined in this study.

Table 1

Hotel characteristics

To assure that we included a broad range of the most common types of hotels, we randomly selected hotels from four strata: budget small, budget large, midscale small and midscale large. The sample of 30 smoking hotels consisted of 8 budget small, 8 budget large, 7 midscale small and 7 midscale large hotels. The sample of 10 non-smoking hotels consisted of 3 budget small, 3 budget large, 2 midscale small and 2 midscale large. All hotel visits were conducted from March 2009 to February 2010. One hotel was visited each week; three smoking hotels and one non-smoking hotel were sampled each month. The week for sampling the non-smoking hotel each month was randomly determined. Prior to the scheduled visit, a research assistant telephoned to reserve the rooms. If a hotel could not confirm a reservation for the needed room type, then a different hotel was selected. At each smoking hotel, reservations were made for one smoking room and one non-smoking room. At each non-smoking hotel, a reservation was made for one non-smoking room. The institutional review board at San Diego State University approved the study.

Measures

All measures were collected by a research assistant who checked into the hotel, set up the air monitoring equipment, conducted the surface wipe sampling and collected observational measures in the afternoon before the confederates (see below) entered the hotel rooms. The research assistant returned to the hotel in the morning to retrieve the equipment and return it to the research office.

Environmental measures of tobacco smoke pollutants

Air nicotine and 3EP in guestrooms

Nicotine and 3EP in the air of a guestroom in which no active smoking is taking place are markers of drifting tobacco smoke (ie, SHS) or re-emitted tobacco smoke (ie, THS).17 ,24 ,46 ,47 Air samples were collected overnight in  hotel guestrooms with a sorbent tube (SKC West 226-93) connected to a sampling pump (SKC Airchek Model 224). Sampling time ranged from 16 to 20 h. Pumps were calibrated to 1.5 lpm before and after use, and samples where pump flow rates changed by >10% were discarded. A field blank was collected at each site. Tubes were transported, cooled and stored at −20°C until extraction, when contents were removed, spiked with deuterated internal standard (3EP-d4 (3-vinylpyridine-d4; TRC Chemical, North York, Ontario, Canada), deuterated nicotine-pyridinal-d4 (nicotine-d4) (66148-15-0, 69980-24-1; CDN Isotopes, Point-Claire, Quebec, Canada)) and extracted in methanol. Samples were analysed by liquid chromatography/mass spectrometry/mass spectrometry (LC-MS/MS) as detailed below. The limit of detection varied with sampling time but was approximately 1.5 ng/m3 nicotine and 1 ng/m3 3EP for the shortest sample. Field blank values were subtracted from the sample values before reporting results.

Surface nicotine and 3EP in guestrooms and hallways

Nicotine and 3EP collected by wipe sampling of surfaces are markers of tobacco smoke that has been sorbed or deposited on the surfaces (ie, THS).17 ,21 ,46 Pre-screened cotton rounds (100% cotton cosmetic rounds) were wetted with 1–2 ml each of 1% ascorbic acid solution and used to wipe a 100 cm2 area on a nightstand in the hotel room and separately on an outside door panel in the hall. An adjacent 100 cm2 area (ie, next to but not overlapping the first area) was wiped with either 0.1% ascorbic acid as the wetting agent or distilled water, as a methodological check. At least one field blank was collected per room. All procedures were performed wearing fresh gloves. Samples were stored at −20°C in the dark until analysis. Cotton rounds were analysed for nicotine and 3EP. Nicotine-d4 was added as an internal standard, then 10 ml of 1 M KOH (aqueous) was added and mixed, and the cotton rounds were removed from the solution. Then, concentrated formic acid was added until acidic, vortexed and 2 ml was transferred to a pre-cleaned solid phase extraction column (Isolute C8, International Sorbent Technologies, Hengoed, UK). The column was washed with water, then the nicotine eluted with acetonitrile:pH4 20 mm ammonium acetate buffer (70:30 v/v) into an amber autosampler vial.

The method of analysis was by LC-MS/MS using positive ion electrospray ionisation (ESI) on a Thermo-Finnigan TSQ Quantum Mass Spectrometer (Thermo Fisher Scientific, Waltham, Massachusetts, USA). Isotope dilution mass spectrometry (IDMS) techniques used nicotine-d4 to quantify the nicotine concentrations. Prepared samples were injected (1–5 µl) from the instrument autosampler onto a silica column (Hypersil Silica, 50×2.1 mm, 3 µm) and separated in a hydrophilic interaction liquid chromatography mode using acetonitrile:pH4 20 mM ammonium acetate buffer of 70 : 30 (v/v) at 150 µl/min. Selected reaction monitoring of the MS–MS transitions at 16V collision-induced dissociation of m/z 163.1 to  117.1 and 130.1 and m/z 167.1 to 121.1 and m/z 134.1 was used for nicotine and the deuterated analogue, respectively. Standard calibration curves (m/z 117.1: m/z 121.1 and m/z 130.1: m/z 134.1) were linear over the concentration range studied, 0.1–1000 ng/ml with R2>0.997. Each batch of 24 samples was run with a blank and a quality control sample. Limits of detection were approximately 0.1–1 μg nicotine/m2 for wipe samples. As IDMS techniques correct for sample losses in extraction and processing, the percent recovery is not important unless maximum sensitivity is required (not necessary in this case). Thus, nicotine concentrations are reported relative to the amount of internal standard added to the samples. Details of percent recovery and other method details are reported elsewhere.48

Biological measures of exposure to tobacco smoke pollutants

We measured exposure to tobacco smoke pollutants in hotel rooms with the assistance of two non-smoking, female confederates (22 years old) who lived with non-smokers and were not exposed to SHS. One confederate stayed overnight in each hotel room. The confederates’ assignment to room type was rotated so that neither stayed in a smoking room more than twice  in 1 month. The confederates checked into the hotel rooms between 19:00 and 20:00, after the research assistant had set up the air monitoring equipment and collected the surface wipe samples. The confederates checked out between 9:00 and 10:00. the following day. During the time (approximately 14 h) they were in the room, confederates were instructed to behave as they would normally, such as using the air conditioning and heating and washing their hands, with the exception that they were instructed to keep the windows in the room closed and not to leave the room or have any visitors other than one pre-approved, screened non-smoker each.

Nicotine and 3EP on fingers

Nicotine and 3EP collected by wipe sampling of non-smokers’ fingers are markers of tobacco smoke exposure through touching polluted objects or through  transfer from the polluted air.21 Finger wipe samples were collected from confederates before and after their stay in the hotel. A baseline sample was obtained at the research office in the evening of day 1 prior to the confederate leaving for the hotel. The research assistant wiped the confederate's dominant hand index finger with a pre-wetted cotton round. A second sample from the same finger was obtained in the rented hotel room on the morning of day 2, before showering and after touching each of the following for at least 30 s: TV remote control, alarm clock (including dials), nightstands (opening and shutting drawers and doors), each chair in the room (including seat, arm rests and picking up chair to move it slightly) and fabric drapes (closing with pull rope if present, and opening by holding just the fabric). The finger wipe samples were analysed using the same methods as for the other wipe samples described earlier.

Urine cotinine

Cotinine is a major metabolite of nicotine and a sensitive and specific biomarker of SHS49 and THS exposure in non-smokers.20 ,21 A baseline urine sample was obtained on day 1 from confederates at the research office prior to their leaving for the hotel.  On day 2, from 6:00  to 19:00, the confederates collected a sample of each urine void, and samples were combined in equal amounts to yield a single day 2 sample. Samples were frozen at −20°C until analysis for cotinine concentration by LC-MS/MS using ESI. Our methods have been previously described.21

Urine NNAL

NNAL is a metabolite of the tobacco-specific carcinogen NNK.50 In non-smokers, urinary NNAL is a carcinogen biomarker of exposure to SHS or THS.51 For the 10 most polluted hotel rooms according to surface nicotine levels, confederates’ urine samples were analysed for NNAL by LC-MS/MS at the Division of Clinical Pharmacology and Experimental Therapeutics, University of California, San Francisco.52

Statistical analyses

Statistical analyses were performed using Stata IC V.12.53 To control for non-normal distributions and heterogeneous error variances, we applied logarithmic transformations to all quantitative variables and report geometric means, medians and quartiles. Tobit regression analyses for left-censored data (ie, nicotine and 3EP levels below the level of detection) were used to test the hypotheses about differences between hotels with and without complete smoking bans and between designated smoking rooms and non-smoking rooms. As the smoking and non-smoking guestrooms of a hotel without a complete smoking ban are not independent, we treated each hotel as a cluster with two observations (ie, non-smoking and smoking rooms). Spearman's r was used to calculate bivariate associations. The type I error rate was set at α=0.05.

Results

THS pollution of non-smoking guestrooms in hotels with and without complete smoking bans

Detailed descriptive statistics for surface and air nicotine and 3EP levels in hotel guestrooms are presented in table 2.

Table 2

Levels of thirdhand smoke pollutants in the air and on surfaces of hotel guestrooms

Surface and air nicotine

In guestrooms of hotels with complete smoking bans, geometric means of surface and air nicotine levels were 1.4 μg/m2 and 20.5 ng/m3 respectively. In non-smoking rooms of hotels without complete smoking bans (ie, hotels that also offer designated smoking rooms), geometric means of surface and air nicotine levels were consistently higher. Surface nicotine levels were more than twice as high as those of non-smoking guestrooms of hotels with complete smoking bans (geometric means: 3.7 vs 1.4 μg/m2), and this difference was statistically significant (p=0.048). Air nicotine levels were approximately 40% higher (28.9 vs 20.5 ng/m3) and this difference was not statistically significant (p>0.20).

Surface and air 3EP

The geometric mean level of 3EP in air was 0.8 ng/m3 in non-smoking rooms of hotels with complete smoking bans and was more than seven times higher in the air of non-smoking rooms of smoking hotels (6.4 ng/m3). This difference was statistically significant (p=0.049). Surface levels of 3EP were below the level of detection in a majority of guestrooms, and their geometric mean levels did not differ from each other (p>0.20).

THS pollution of smoking guestrooms of hotels without complete smoking bans

Surface and air nicotine

In smoking guestrooms, the geometric means of surface and air nicotine levels were 51.8 μg/m2 and 452.4 ng/m3, respectively. That is, when non-smoking confederates stayed overnight in guestrooms where previous guests were permitted to smoke, surface and air nicotine levels were 35 and 22 times higher than those of non-smoking rooms of hotels with complete smoking bans, respectively (p<0.001). Surface and air nicotine levels in smoking guestrooms were 13 and 15 times higher than those of non-smoking rooms of hotels without complete smoking bans, respectively (p<0.001). Note that the third quartiles and maxima in hotels without complete smoking bans reached substantially higher levels than the median levels.

Surface and air 3EP

The mean level of 3EP in the air was 63.6 ng/m3 in smoking rooms; this is about 10 times higher than mean levels in non-smoking rooms of hotels without complete smoking bans (p<0.001) and more than 75 times higher than those in the rooms of hotels with complete smoking bans (p=0.003). Surface levels of 3EP were below the level of detection in a majority of smoking guestrooms, and geometric mean levels did not differ from those of non-smoking rooms (p>0.20).

THS pollution in hallways outside of guestrooms

Detailed descriptive statistics for surface nicotine and 3EP levels in hallways outside hotel guestrooms are presented in table 3. Hallways outside of smoking hotel rooms showed significantly higher levels of surface nicotine (9.3 μg/m2) than those found in hallways of non-smoking rooms of hotels with (1.2 μg/m2; p=0.008) and without (2.8 μg/m2; p=0.003) complete smoking bans. Although levels in hallways outside of non-smoking rooms in hotels without complete smoking bans were more than twice as high as those in hotels with complete smoking bans, this difference was not statistically significant (p>0.20).

Table 3

Levels of thirdhand smoke pollutants on surfaces in hotel hallways outside of guestrooms

Geometric means of surface nicotine in hallways outside of non-smoking guestrooms showed similar levels as those found inside the guestrooms reported above (1.2 vs 1.5 μg/m2 and 2.8 vs 3.7 μg/m2 for hotels with and without complete smoking bans, respectively). Neither of these differences were statistically significant, nor was the difference between hallway nicotine levels outside non-smoking rooms of smoking and non-smoking hotels (all p values>0.20). Similarly to surfaces in guestrooms, 3EP was not detectable on a majority of hallway surfaces. The third quartiles and maxima indicate that some hotel hallways were substantially more polluted with THS than were the average hallways.

Exposure to THS

Table 4 provides detailed descriptive statistics for finger nicotine and urine cotinine levels of non-smoking confederates staying overnight in the hotel rooms. There were no differences at baseline between confederates assigned to staying in different hotel rooms in finger nicotine and urine cotinine with median levels equivalent to those found after the overnight stay in hotels with complete smoking bans (all p values >0.20).

Table 4

Exposure to thirdhand smoke among non-smoking confederates after staying in a hotel guestroom

Nicotine on fingers of non-smoking confederates

Non-smoking confederates staying in non-smoking rooms of hotels without complete smoking bans showed significantly higher levels of nicotine on their fingers than after staying in hotels with complete smoking bans (2.4 vs 11.9 ng/wipe; p=0.029). Confederates staying in smoking rooms accumulated the highest concentrations of nicotine on their fingers (60.3 ng/wipe); this level was significantly higher than those levels found after they stayed in non-smoking rooms of hotels with and without complete smoking bans (p<0.001 and p=0.001, respectively). Note that some hotel rooms contributed to finger nicotine levels many times higher than the median or mean levels (see third quartile and maxima in table 4).

Urine cotinine levels of non-smoking confederates

Non-smoking confederates staying in smoking rooms showed significantly elevated levels of cotinine in their urine following their hotel stay, with geometric mean levels 5–6 times higher compared with confederates staying in non-smoking hotel rooms of hotels with and without complete smoking bans (p<0.001). Geometric mean urine cotinine levels did not differ between non-smoking confederates staying in non-smoking rooms of smoke-free and smoking hotels. Note that some hotel rooms contributed to urine cotinine levels many times higher than the median or mean levels (see third quartile and maxima in table 4).

NNAL of non-smoking confederates staying in smoking rooms

For the 10 most polluted rooms based on surface nicotine levels, we determined urinary NNAL levels in confederates before and after their overnight stay. NNAL levels significantly increased after spending 12–14 h in these rooms, with an average increase of 0.39 pg NNAL/mg creatinine (p=0.003). Geometric mean levels were 0.86 pg NNAL/mg creatinine before staying in the rooms and 1.24 pg NNAL/mg creatinine after spending a night in the rooms.

Association between THS pollution of hotel rooms and confederates’ exposure

Table 5 shows bivariate associations between nicotine and 3EP levels found in hotel guestrooms and finger nicotine and urine cotinine levels found in non-smoking confederates after their overnight hotel stay. Surface nicotine levels were significantly and positively associated with air and finger nicotine levels and with urine cotinine levels.

Table 5

Spearman's correlations between measures of THS pollution in hotel rooms and THS exposure of non-smoking confederates staying in hotel rooms

Air nicotine levels were positively and significantly associated with air 3EP levels, finger nicotine and urine cotinine. Surface 3EP was not significantly associated with any of the other measures.

Urine cotinine showed positive and significant associations with surface nicotine, air nicotine, air 3EP and finger nicotine levels. When examined in a Tobit regression model controlling for baseline urine cotinine levels, air nicotine (p=0.003), finger nicotine (p=0.017) and surface nicotine (p=0.039) each independently accounted for variance in urine cotinine levels (F3,58=11.64, p<0.001).

For the subsample of the 10 most polluted smoking rooms based on surface nicotine levels, the change in urinary NNAL levels before and after sleeping in these rooms was significantly positively associated with the change in urinary cotinine in the same samples (Spearman's r=0.73, p=0.025).

Discussion

This is the first study to evaluate the effectiveness of partial smoking bans in hotels (ie, smoking is banned in designated non-smoking rooms and allowed in designated smoking rooms), in protecting non-smoking guests from tobacco smoke exposure. We examined how smoking in hotels caused persistent pollution of designated smoking and non-smoking rooms (ie, THS pollution) and biological exposure to tobacco smoke pollutants of non-smoking confederates staying overnight in these rooms (ie, THS exposure). Across multiple measures of tobacco smoke pollution and biological exposure to tobacco smoke, a consistent picture emerges to illustrate the consequences of allowing smoking in hotels. A partial smoking ban did not protect either the non-smoking rooms from tobacco smoke pollution or the confederates staying in these rooms from exposure to tobacco smoke. Guests who wish to protect themselves from exposure to tobacco smoke should avoid hotels that permit smoking and instead stay in completely smoke-free hotels. Non-smoking rooms of hotels without complete smoking bans show higher levels of tobacco smoke pollutants on surfaces (nicotine) and in the air (nicotine and 3EP) than non-smoking rooms in hotels with complete smoking bans. Consistent with higher THS pollution levels in hallways outside of smoking rooms, THS in non-smoking guestrooms is likely caused by smoke drifting from smoking rooms to non-smoking rooms via hallways, windows, utility ducts and ventilation systems.12–14 ,54 THS in non-smoking rooms may be caused by guests smoking who may not know their room is a non-smoking room or those who disregard policy. Designated smoking rooms showed the highest levels of THS pollution, comparable to the levels found in private homes of active smokers.24

Our findings demonstrate that non-smoking confederates were exposed to tobacco smoke pollutants that collected on their fingers after spending 12–14 h in a smoking or non-smoking room of a hotel without a complete smoking ban. The hypothesis that the source of nicotine on confederates’ fingers was surface nicotine in the guestrooms is further supported by the positive correlation between nicotine on guestroom surfaces and nicotine found on confederates’ fingers the morning after their stay. These findings show that fingers are important sampling tools for environmental pollutants residing on surfaces. Nicotine on fingers also may lead to hand-to-mouth, hand-to-nose and hand-to-eye exposure routes independent of or in addition to inhaling re-emitted volatile gases or re-suspended respirable particles. Findings from the Tobit regression model are consistent with the hypothesis that exposure to THS may occur through inhalation, ingestion and dermal transfer.

A closer examination of the distribution of THS pollution and exposure levels points to important variability between hotels and limitations of 3EP as a marker of THS. Our findings illustrate that some hotels with complete smoking bans show evidence of THS, and guests are exposed to tobacco smoke pollutants. This is possibly the consequence of some guests ignoring the existing smoking policies and of THS pollutants that accumulated before a hotel adopted a complete smoking ban. Our findings also demonstrate that some non-smoking guestrooms in smoking hotels are as polluted with THS as are some smoking rooms. Moreover, non-smoking guests staying in smoking rooms may be exposed to tobacco smoke pollutants at levels found among non-smokers exposed to SHS. Of our measures of THS pollution, 3EP on surfaces was the least sensitive. This suggests that in contrast to nicotine, 3EP does not accumulate much on surfaces and is a better measure of SHS than THS pollution. It is also possible that 3EP detected in the air may represent drifting SHS tobacco smoke rather than re-emitted THS.

It is not uncommon for non-smokers to be offered designated smoking rooms when a hotel has a shortage of designated non-smoking rooms. When non-smokers are assigned to designated smoking rooms, a one-night stay leads to significantly higher exposure to nicotine as measured by its metabolite, cotinine, found in urine collected during the following day. Findings from this study suggest that this practice should be abandoned because designated smoking rooms are highly polluted with THS and lead to tobacco smoke exposure, including exposure to the potent tobacco-specific lung carcinogen NNK (assessed through measuring its metabolite NNAL in urine). This is the first study to show increased levels of NNAL among non-smoking confederates staying in THS-polluted smoking rooms. Our findings are consistent with a growing body of evidence demonstrating exposure to NNK among non-smokers exposed to SHS in hospitality venues (eg, casinos, inside and outside restaurants and bars).55–57

It is not well understood how to clean up a room in which tobacco has been smoked consistently, periodically or occasionally. Sleiman et al58 reported that nicotine, a THS compound readily available on indoor surfaces in smoking hotels, reacts with atmospheric nitrous acid to form three carcinogenic tobacco-specific nitrosamines, one of which (NNA) is not present in freshly emitted tobacco smoke. In an effort to remove the unpleasant odour of stale tobacco smoke, hotels sometimes employ ‘air purifiers’ which are intended to trap small airborne particles with filters. These do not affect volatile and semi-volatile compounds in their gas phase, and their effectiveness depends on the type of filter used and on the maintenance of equipment. Other methods focus on neutralising unpleasant odour (eg, ozone machines) and are designed to induce chemical reactions that change odorants into odourless compounds. However, ozone and related atmospheric oxidants undergo chemical reactions with THS components that may create new and additional irritants and toxicants (including ultrafine particles, formaldehyde, N-methyl formamide and myosmine).59–61 Thus, it is not yet clear how best to decontaminate polluted homes or hotel rooms. Until effective technology is developed, contaminated rooms should be confirmed by similar assays, disclosed and avoided.

This study has several notable limitations. First, our sampling approach for hotels was not designed to achieve a representative probability sample of hotels. Instead, it was designed to yield a broad range of common hotel classes. Second, we did not investigate the history of smoking policies of the hotels or of the specific hotel rooms we studied. Therefore, we could not determine whether THS pollutants in non-smoking rooms were deposited via migrating tobacco smoke from other smoking rooms in the same hotel or from previous tobacco use. Finally, we could not determine whether and what kinds of remediation efforts a hotel may have employed to neutralise the odour of stale tobacco smoke or to remove THS pollutants from the room.

Different from smoking bans in restaurants and bars, few countries have adopted public policies that completely ban smoking in hotels, and the most common public policy and practice are to offer guests a choice of designated non-smoking and smoking rooms. The physical separation of smoking and non-smoking rooms, however, has been proven ineffective in workplaces, private homes, restaurants and bars.30 ,62–65 It does not come as a surprise that such a policy does not work for hotels either. Findings from this study suggest that it is time to abandon smoke-free exemptions for hotels and offer all non-smokers a consistent, complete and well-implemented 100% smoke-free policy throughout the hospitality industry. New hotels should adopt complete smoking bans to protect all employees and guests from exposure to tobacco smoke toxicants and to protect the hotel from cleaning expenses.

What this paper adds

  • This study is the first to evaluate the effectiveness of partial smoking bans in hotels (ie, smoking is banned in designated non-smoking rooms and allowed in designated smoking rooms) in protecting non-smokers from tobacco smoke exposure.

  • A partial smoking ban protects neither nonsmoking rooms from tobacco smoke pollution nor guests staying in these rooms from exposure to tobacco smoke.

  • Hotel guests interested in protecting themselves from exposure to tobacco smoke should avoid hotels that permit smoking and instead stay in completely smoke-free hotels.

References

Supplementary materials

  • Press release

    Files in this Data Supplement:

Footnotes

  • Correction notice This article has been corrected since it was published Online First. It was brought to the authors' attention that Vermont allows smoking in designated hotel/motel rooms. As such, in the Introduction the sentence ‘As of 2 January 2013,4 only Indiana, Michigan, North Dakota, Vermont and Wisconsin prohibited smoking in hotel and motel rooms’ has been amended to ‘As of 2 January 2013,4 only Indiana, Michigan, North Dakota and Wisconsin prohibited smoking in hotel and motel rooms’.

  • Contributors All authors made substantial contributions to the conception and design, acquisition of data or analysis and interpretation of data; drafting the article or revising it critically for important intellectual content; and final approval of the version published. GEM is responsible for the overall content as guarantor.

  • Funding This research was supported by funds from the California Tobacco-Related Disease Research Grants Program Office of the University of California, Grant number 17RT-0162H.

  • Competing interests None.

  • Ethics approval San Diego State University.

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

  • Data sharing statement We will make the published data and associated documentation available to users only under a data-sharing agreement that provides for: (1) a commitment to using the data only for research purposes and not to identify any individual participant; (2) a commitment to securing the data using appropriate computer technology; (3) a commitment to destroying or returning the data after analyses are completed and (4) a commitment to sharing expenses associated with the preparation of data files for sharing. Additional unpublished data from this study are not available for data sharing.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.