Outdoor tobacco smoke exposure at the perimeter of a tobacco-free university

Abstract

There are few studies measuring exposure to outdoor tobacco smoke (OTS). Tobacco users often gather at the boundaries of tobacco-free campuses, resulting in unintended consequences. The objective of this study was to measure exposure levels from OTS on sidewalks bordering a tobacco-free university campus. Data were collected while walking along a sidewalk adjacent to a medium traffic road between May and August 2011. Monitoring occurred during “background,” “stop,” and “walk-through” conditions at and near hot spot area to measure fine particulate matter (<2.5 μm; PM_2.5) from OTS using a portable aerosol monitor. The average PM_2.5 levels during stop and walk-through conditions were significantly higher than during background conditions. PM_2.5 peak occurrence rate and magnitude of peak concentration were significantly different depending on smoking occurrence. The peak occurrence rate during the stop condition was 10.4 times higher than during the background condition, and 3.1 times higher than during the walk-through condition. Average peak PM_2.5 concentrations during the stop condition were 48.7% higher than during the background condition. In conclusion, individuals could be exposed to high levels of PM_2.5 when stopping or even passing by smokers outdoors at the perimeter of tobacco-free campuses. The design and implementation of tobacco-free campus policies need to take into account the unintended consequences of OTS exposure at the boundaries.

Implications:

In this study, outdoor tobacco smoke (OTS) exposure was measured at the perimeter of tobacco-free campus. OTS exposure could be determined by peak analysis. Peak occurrence rate and peak concentration for OTS exposure were identified by using peak analysis. People could be exposed to high levels of PM_2.5 when standing or even passing by smokers at the perimeter of tobacco-free campus. OTS exposure measurement in other outdoor locations with smokers is needed to support outdoor smoking regulation.

Introduction

When secondhand smoke (SHS) occurs outdoors, it is called outdoor tobacco smoke (OTS). When indoor smoking restrictions are implemented, many smokers move outdoors to smoke, especially near buildings entrances where people entering or exiting the building may be exposed (Kaufman et al., 2011). OTS is just as dangerous as indoor SHS, depending on the number of and proximity to smokers and the weather (e.g., wind speed, wind direction, temperature). The transient peak PM_2.5 (particulate matter with an aerodynamic diameter <2.5 μm) concentration has been reported to be as high as 1000 μg/m³ within 0.5 m of a smoker (Klepeis et al., 2007). The average concentration over the duration of a burning cigarette was higher than 200 μg/m³. The level exceeded 500 μg/m³ in the downwind direction, whereas upwind concentrations were negligible.

In the United States, many colleges, universities, and health care campuses have addressed the health threat of OTS by implementing tobacco-free campus policies (Lee et al., 2010). There are at least 1182 colleges and universities that had adopted 100% smoke- or tobacco-free policies without exemptions as of January 2014 (American Nonsmokers’ Rights Foundation, 2014). These policies typically cover only areas owned and controlled by the institution, and they may not have the authority to restrict smoking on city-, county-, or state-owned streets and sidewalks at campus perimeters. An unintended consequence of tobacco-free policy is that smokers may move to the campus perimeter.

There are few studies measuring OTS on sidewalks where smokers congregate. The purpose of this study was to measure exposure levels from OTS on sidewalks bordering a tobacco-free campus. The selected campus for the study is a flagship university in the U.S. tobacco-belt. The tobacco-free campus policy had been in effect for about 18 months at the time of data collection.

Experimental Methods

The study was conducted along a sidewalk adjacent to a medium traffic road (urban principal arterial with average annual daily traffic of 27,000 vehicles; Kentucky Transportation Cabinet, 2014) on the perimeter of a tobacco-free university campus between May 9 and August 16, 2011. The concentration of PM_2.5 was measured using an aerosol monitor called SidePak (model AM510; TSI Inc., MN, USA) with a 2.5-μm impactor. Prior to each measurement, the SidePak monitor was zero-calibrated with a high-efficiency particulate absorption (HEPA) filter according to the manufacturer’s specifications. The air flow rate was set to 1.7 L/min, and the recording of PM_2.5 concentrations was done every second. The measured value from the SidePak monitor was multiplied by a calibration factor of 0.295 obtained from an earlier comparison against a gravimetric measurement method (Lee et al., 2007).

The SidePak monitor was placed in a backpack with the sampling inlet placed at approximately mouth level. A team of three individuals collected the data on nine separate days; one carried the backpack and monitored time on the atomic clock while the other two used a field log sheet to document time, number of smokers, wind gust, smokers’ distance from the sidewalk, and location. On each day, the temperature, humidity, wind speed and direction, and atmospheric conditions were recorded using a weather forecast Web site (http://www.weather.com).

The daily monitoring occurred in three separate “conditions”: (1) background, (2) stop, and (3) walk-through. Data for the background condition were collected by walking along the sidewalk in areas with no smoking nearby for 10 min before and after each “stop” and “walk-through” condition. The stop condition was defined as standing in areas with high levels of smoking (i.e., “hot spots”) for 10 min. The walk-through condition involved walking back and forth through the hot spots at a pace of 1 step per second. The order of each day’s observations was (1) background, (2) stop, (3) background, (4) stop, (5) background, (6) walk-through, and (7) background conditions.

Because concentrations measured outdoors can fluctuate due to various environmental factors, a “peak analysis” approach was used by defining a “peak” as an increase of 35 μg/m³ in 1 sec (C_i − C_i−1), where i = 1, 2, … n, and n is the total number of seconds. Because of the emphasis on the concentration increase, only positive differences in reading pairs were included in the count. The selection of 35 μg/m³ for defining the peak was determined by the distribution of the difference between C_i and C_i−1 where no smoking occurred (Cho and Lee, 2014). The 99.5 percentile of the distribution of observed PM_2.5 concentrations was 33 μg/m³ based on about 100 hr of data spent in various road site locations. However, 35 μg/m³ was selected because this value is the 24-hr U.S. National Ambient Quality Standards for PM_2.5. When there was more than one “peak” within each successive 9-sec interval, the maximum concentration within that interval was selected as the “peak.” For example, one peak was selected in each of the first 9-sec intervals, C₁–C₉, C₁₀–C₁₈, C₁₉–C₂₇, …, and so on. One peak per 9 sec was selected based on the assumption that multiple peaks within a short interval may be caused by the same source, such as the person’s exhalation. Peak occurrence rate was defined as the number of peaks per hour.

Average PM_2.5 levels were calculated for each condition after excluding 20 sec at the beginning of the condition. PM_2.5 levels of background conditions were compared with those of stop and walk-through conditions by the geometric linear model (GLM) procedure. The number of peaks, the average peak levels, and the peak occurrence rates were calculated for each condition. Average PM_2.5 peak concentrations and peak occurrence rates were analyzed using the Kruskal-Wallis test to compare OTS exposure levels in the three conditions. The SAS 9.3 (SAS Institute, Cary, NC) statistical package was used for statistical analysis, and the graphs were drawn in SigmaPlot 10.0 (Systat Software Inc., San Jose, CA).

Results

The experiments were conducted on nine different days. A total of 17 stop, 8 walk-through, and 9 background conditions were available for the analysis. Average daily temperature was 24.6 (SD 3.9) °C; average daily relative humidity was 63.0% (SD 12.4%); and average wind speed was 3.3 (SD 1.5) m/sec. The average PM_2.5 level when smoking was observed was significantly higher than background levels (P = 0.0037). The geometric mean PM_2.5 level in the background condition was 8.7 (GSD = 1.8) μg/m³. Geometric mean PM_2.5 levels in the stop and in the walk-through conditions were 15.9 (GSD = 1.8) μg/m³, and 10.6 (GSD = 1.8) μg/m³, respectively (see Table 1).

Figure 1. Average PM_2.5 levels by condition compared with background (units: μg/m³)

Condition	n	GM (GSD)	Min.	Max.	P Value
Background	9	8.7 (1.8)	2.5	22.4	0.0037
Stop	17	15.9 (1.8)	5.4	33.2
Walk-through	8	10.6 (1.8)	5.8	29.6

Since outdoor concentrations fluctuated within a few seconds, the peak occurrence rate and magnitude of peak concentrations were compared across conditions. The distributions of peak occurrence rates and average peak concentrations in each condition are shown in Figure 1. The background condition had a peak occurrence rate of 3.8 peaks per hour. Peak occurrence rates in stop and walk-through conditions were 39.6 and 11.6 peaks per hour, respectively. The average magnitude of peak concentration in the background condition was 87.7 (GSD = 1.8) μg/m³, compared with 130.4 (GSD = 2.1) μg/m³ in the stop condition and 227.8 (GSD = 3.2) μg/m³ in the walk-through condition. The peak occurrence rates and the average magnitude of peak concentrations were significantly different in the three conditions (P < 0.0001 for peak occurrence rate and P < 0.05 for magnitude of peak concentration).

Figure 1. Distributions of peak occurrence rate and peak PM_2.5 concentration for three conditions. Box plots show the median as a center bar, the 25th and 75th percentiles as a box, the 5th and 95th percentile values at the whiskers, and the mean as a dotted line. There were only four cases of walk-through condition.

Discussion

Exposure to PM_2.5 was measured near smokers at the perimeter of a tobacco-free university campus. When PM_2.5 levels were compared across the three conditions, PM_2.5 concentrations in the stop condition were the highest, followed by the walk-through condition and followed by the background condition. The finding is consistent with high PM_2.5 levels reported when smokers are within a 5-m radius from monitors in outdoor bars and restaurants (St.Helen et al., 2011). Since PM_2.5 can be generated by several other sources including automobile exhaust (Gertler et al., 2000), measuring the background level on the same street is critical when monitoring the effects of OTS.

The average PM_2.5 concentrations when a person stopped near a smoker were 83% higher than background levels. PM_2.5 level can be high near smoker. Acevedo-Bolton et al. (2013) reported in a study of indoor measurements that a nonsmoker sitting within 1 m of a burning cigarette was exposed to over 160 μg/m³ during the smoking period. In outdoor environments, a strong “proximity effect” of the smoker was observed at bus stops (Ott et al., 2014). The mean PM_2.5 personal exposures of nonsmokers at distances 0.5, 1.0, and 1.5 m from the smoker were 59, 40, and 28 μg/m³, respectively, compared with a background level of 1.7 μg/m³. In another study, the average PM_2.5 level measured in 69 outdoor dining areas of cafes and restaurants was 325% higher when cigarettes were actively burned within 1 m from the monitor (Cameron et al., 2010). Similarly, the average PM_2.5 level was 358% higher in outdoor seating areas of 28 cafes and pubs with at least one smoker (Stafford et al., 2010). In this study, monitoring was conducted in open spaces without any walls or coverings and with smokers located up to 6 m from the monitor.

Unlike indoor environments, outdoor PM_2.5 concentrations along a street can change instantly due to the wind and the motion of nearby motor vehicles. Because fine particle air pollution can be affected by wind speed or wind direction, the average PM_2.5 concentration may not be the best way to analyze the data. Therefore, peak analysis was used to account for temporal fluctuation. In this study, increased peak occurrence rate and average peak concentrations confirmed the presence of OTS when smoking was observed.

There were several limitations to this study. First, although detectable OTS level was measured up to 9 m from a smoking source (Hwang and Lee, 2013), more site-specific information such as distance between each smoker and the monitor and time-specific wind speed and direction were not collected. Second, although monitoring occurred without prior notice and the monitor was concealed, nonsmoking technicians walking back and forth in a location of high concentration may have altered smokers’ behaviors. Third, use of a single calibration factor was a limitation, since the calibration factor is expected to be different for the many combined sources that make up ambient air and for emission due to traffic (Jiang et al., 2011). Ambient measurements using a SidePak monitor were 2.6–3.1 times higher than those using a gravimetric method (Jenkins et al., 2004), which would be equivalent to a calibration factor of 0.32–0.38, since the calibration factor is the reciprocal of this ratio. Although the use of the 0.295 calibration factor might cause error, such variation is not likely to be significant. Lastly, this study did not measure nearby vehicle traffic. However, since we collected data on the same street, contribution from traffic emissions would be minimal. Future field studies of OTS need to monitor vehicular traffic density simultaneously with PM_2.5 monitoring to characterize the actual contributions from smokers.

In conclusion, individuals may be exposed to high levels of fine particles when stopping near smokers or passing by them at the perimeter of tobacco-free campuses. The design and implementation of tobacco-free campus policies need to take into account the unintended consequences of OTS exposure at the boundaries. Further research is needed to measure exposure to fine particle in outdoor locations with smokers.

Acknowledgment

The authors are grateful for field staff, the undergraduate nursing research intern program, and technical support from University of Kentucky College of Nursing Tobacco Policy Research Program.

Funding

This work was supported by the National Research Foundation of Korea (NRF), grant funded by the Korean Government (no. 2011-0016022) and BK21 Plus project (22A20130012682).

Additional information

Notes on contributors

Hyeri Cho

Hyeri Cho is a graduate student, Kiyoung Lee is an associate professor, and Yunhyung Hwang is a doctoral candidate at Seoul National University.

Patrick Richardson

Patrick Richardson, Elizabeth Teeters, and Rachael Record are graduate students, Carol Riker is an associate professor, and Ellen J. Hahn is a professor at University of Kentucky.

Hilarie Bratset

Hilarie Bratset is an air quality coordinator, Kentucky Center for Smoke-Free Policy.