Determination of secondhand smoke leakage from the smoking room of an Internet café

ABSTRACT

Although Internet cafes have been designated as nonsmoking areas in Korea, smoke-free legislation has allowed the installation of indoor smoking rooms. The purposes of this study were to determine secondhand smoke (SHS) leakage from an Internet café smoking room and to identify factors associated with SHS leakage. PM_2.5 (particulate matter with an aerodynamic diameter ≤2.5 μm) mass concentrations were measured simultaneously both inside and outside the door to the smoking room. During each measurement, a field technician observed how long the smoking room door was opened and closed, the direction of door opening, and the number of smokers. A multivariate linear regression model was used to identify the causality of SHS leakage from the smoking room. A time series of PM_2.5 concentrations both inside and outside the door to the smoking room showed a similar trend. SHS leakage was significantly increased because of factors associated with the direction of the smoking room door being opened, the duration of how long the smoking room door was opened until it was closed, and the average PM_2.5 concentration inside the smoking room when the door was opened. SHS leakage from inside the smoking room to outside the smoking room was evident especially when the smoking room door was opened. Since the smoking room is not effective in preventing SHS exposure, the smoking room should be removed from the facilities to protect citizens from SHS exposure through revision of the current legislation, which permits installation of a smoking room.

Implications: This paper concerns secondhand smoke (SHS) leakage from indoor smoking room. Unlike previous studies, the authors statistically analyzed the causality of PM_2.5 concentration leakage from a smoking room using time-series analysis. Since the authors selected the most common smoking room, the outcomes could be generalized. The study demonstrated that SHS leakage from smoking room and SHS leakage were clearly associated with door opening. The finding demonstrated ineffectiveness of smoking room to protect citizens and supports removal of indoor smoking room.

Introduction

On January 1, 2014, Korea implemented smoke-free legislation to protect nonsmokers and workers from secondhand smoke (SHS) by designating many publicly used facilities as a nonsmoking area (Ministry of Health and Welfare of Korea, 2015). Internet cafes were also designated as a nonsmoking area at this time. However, according to Article 9 of the Enforcement Regulations of the National Health Promotion Act 2015, the owners or managers of publicly used facilities could install smoking rooms for smokers. Although the recommendation was to install smoking rooms outdoors, where this was not possible, installation of indoor smoking rooms was permitted.

Smoking in indoor smoking rooms may cause considerable SHS leakage. Installation of a mechanical ventilation system in a designated smoking room is not sufficient to eliminate SHS (Carrington et al., 2003). Increasing the ventilation rate in smoking rooms does not reduce SHS exposure for the nonsmokers near the smoking room (Wagner et al., 2004). Noticeable leakage of SHS was observed when the smoking room door was opened and closed (Wan et al., 2010). Where indoor smoking rooms have been installed in certain cities, PM_2.5 (particulate matter with an aerodynamic diameter ≤2.5 μm) concentrations were reported to be significantly higher than in other cities that strictly adopted comprehensive smoke-free legislation (Schoj et al., 2010).

Smoking rooms in indoor facilities may cause health problems due to SHS leakage. However, the relationship between PM_2.5 concentrations inside and outside smoking rooms has only been shown either descriptively or by graphical demonstration (Cenko et al., 2004; Lee et al., 2010). Therefore, in this study, we statistically analyzed the causality of PM_2.5 concentration leakage from a smoking room using time-series analysis. The purposes of this study were to determine SHS leakage from a smoking room and to identify factors associated with SHS leakage.

Methods

SHS leakage from one indoor smoking room of an Internet café was assessed. This smoking room was selected on the basis of characteristics derived from 225 smoking rooms in 202 Internet cafes (Kim and Lee, 2016). This smoking room was the most common type identified in the previous study. Area and height of the smoking room were 5.6 m² and 2.4 m, respectively. The smoking room had one 0.85 m × 2 m hand-operated hinged door, two ventilation fans, and no window. Area and height of the Internet café were 257.9 m² and 2.4 m. Assessment was carried out from July 6 to 9, 2015, for more than 4 hr per day between 4:30 p.m. and 9:30 p.m.

PM_2.5 mass concentrations were measured and recorded using two personal aerosol monitors (SidePak model AM510; TSI, Shoreview, MN, USA). Monitors were placed both inside and outside the door to the smoking room. One monitor was placed at the nearest computer desk from the smoking room door. The distance from the door was 1.2 m and the height from the floor was 1.1 m, which was breathing zone when seated. The other monitor was placed at the center of the smoking room. The distance from the door was 1.2 m and the height from the floor was 1.1 m. To measure PM_2.5 concentrations, an aerodynamic diameter of 2.5 µm or less, the instrument was equipped with a 2.5 µm impactor, and the air flow rate was set to 1.7 liters per minute (LPM). Prior to each measurement, the impactor was cleaned and zero-calibrated with a high-efficiency particulate air (HEPA) filter. The conversion factor of 0.295 was applied to the measured data for calibration against a gravimetric measurement (Lee et al., 2008).

PM_2.5 mass concentrations were measured simultaneously both inside and outside the door to the smoking room. During each measurement, a field technician observed and recorded how long the door to the smoking room was opened and closed. If the door was reopened before it was completely closed, we considered the door to be still open. The direction of door opening was recorded as a dummy variable. The number of smokers in the smoking room was observed. All data were recorded continuously at 1-sec intervals during all measurement times. Data analysis was based on 1-sec concentrations. Log-transformed 1-min average PM_2.5 mass concentrations were used only when plotting the time series to show trend clearly, since a graph drawn at 1-sec intervals was difficult to comprehend.

Model development was based on a multivariate linear regression using R software, version 3.3.1 (R Core Development Team, 2016). In the preliminary analysis, there were high correlations between some variables. The average PM_2.5 concentration inside the smoking room when the door was opened was highly correlated with number of smokers (r² = 0.8532) and the 1-sec maximum PM_2.5 concentration inside the smoking room (r² = 0.9756) when the door was opened. Therefore, the average PM_2.5 concentration inside the smoking room when the door was opened was alternately included in the initial model to remove multicollinearity. Although the date of the measurement day did not show any significant difference, it was included in the initial model. Three interaction terms were included. Multivariate linear regression models for all reasonable combinations of the variables were tested. Akaike’s information criterion (AIC) was used for the model selection (Venables and Ripley, 1992; Chambers and Hastie, 2003). The initial multivariate linear regression model was as follows:

(1)

was the difference in the average PM_2.5 concentration outside the smoking room “for 5 sec before the smoking room door was opened” and “for 5 sec (approximately −1 to +3 sec based on the maximum concentration) after the smoking room door was opened.” was the direction of the smoking room door being opened and was a dummy variable (inward = 0, outward = 1). was the duration of door opening until the smoking room door was closed after the door was opened. was the average PM_2.5 concentration inside the smoking room when the door was opened. was the date of the measurement day. , , and were interaction terms. β₀–β₇ were estimates, and ε_i was the error term.

Results

PM_2.5 mass concentrations inside and outside the door to the smoking room were measured over four consecutive days. The maximum concentration the instrument can measure is 5900 µg/m³, so concentrations inside the smoking room that may have exceeded this level were entered into the data set as 5900 µg/m³. The average PM_2.5 concentrations inside the smoking room over 4 days were (geometric standard deviation [GSD] in parentheses) 147.6 (3.0), 146.1 (2.9), 143.8 (4.1), and 150.0 (3.2) µg/m³, respectively. The average PM_2.5 concentrations outside the smoking room over 4 days were 23.7 (1.7), 23.2 (1.7), 7.0 (2.0), and 8.0 (1.7) µg/m³, respectively. The 1-sec maximum PM_2.5 concentrations outside the smoking room over 4 days were 626.0, 641.0, 750.5, and 576.1 µg/m³, respectively.

Figure 1 shows the time series of 1-min average log-transformed PM_2.5 concentrations for both sides of the door to the smoking room on each measurement day. There were similar trends inside and outside the door to the smoking room. After most door openings, the graph indicated that a peak increase in PM_2.5 concentrations outside the smoking room might be associated with PM_2.5 concentrations inside the smoking room.

Figure 1. Time series of 1-min average log-transformed PM_2.5 concentrations measured inside and outside the smoking room.

The smoking room door was opened 777 times during the study period. Analysis was conducted for 427 door openings except where the time interval before and after door openings was less than 10 sec. The average number of smokers was about 0.8 during 16.7 hr, which was the total measurement time. The model without the date of the measurement day and without all the interaction terms showed the smallest AIC, and its explanatory power was the best. The final multivariate linear regression model was as follows:

(2)

was the difference in the average PM_2.5 concentration outside the smoking room “for 5 sec before the smoking room door was opened” and “for 5 sec (approximately −1 to +3 sec based on the maximum concentration) after the smoking room door was opened.” was the direction of the smoking room door being opened and was a dummy variable (inward = 0, outward = 1). was the duration of door opening until the smoking room door was closed after the door was opened. was the average PM_2.5 concentration inside the smoking room when the door was opened. β₀–β₃ were estimates, and ε_i was the error term. The residual was assumed to be normally distributed.

Table 1 shows a summary of the multivariate linear regression model used to determine causality of SHS leakage from the smoking room. The PM_2.5 concentration outside the smoking room was significantly increased when the smoking room door was opened in an outward direction compared with an inward direction (P < 0.0001). The longer the door was opened, the higher the increase in PM_2.5 concentration outside the smoking room, and this was a statistically significant increase (P < 0.0001). The average PM_2.5 concentration inside the smoking room also had a significant influence on the increase in PM_2.5 concentration outside the smoking room (P = 0.0005).

Table 1. Summary of the multivariate linear regression model.

Parameter	Estimate	Standard error	t value	P value
()	−28.0330	6.5577	−4.2748	<0.0001
()	27.0059	4.1252	6.5465	<0.0001
()	6.2422	1.3390	4.6620	<0.0001
()	0.0196	0.0056	3.4827	0.0005

Discussion

The SHS leakage was determined by simultaneous measurements inside and outside the door to the smoking room located in the Internet café. Unsurprisingly, the average PM_2.5 concentration inside the smoking room was very high. The 1-sec maximum PM_2.5 concentrations outside the smoking room were very high, whereas the average PM_2.5 concentrations were low. The average PM_2.5 concentrations outside the smoking room in the last 2 days were lower than the other 2 days. These lower average concentrations could be due to rainy weather at that time. However, the differences in the 1-sec maximum PM_2.5 concentrations over the 4 days were not statistically significant, and the weather did not appear to be a factor.

PM_2.5 concentration time series inside and outside the door to the smoking room were similar. This similar trend was consistent with other studies (Cenko et al., 2004; Lee et al., 2010). There was a high correlation and similar trend of the SHS indicator between the smoking room and adjacent nonsmoking areas (Cenko et al., 2004). The time series of the PM_2.5 concentrations outside the smoking room were consistent with those concentrations inside the smoking room in three smoking rooms at an airport (Lee et al., 2010). Peak PM_2.5 concentrations outside the smoking room were associated with opening of the door. These studies certainly demonstrated that SHS inside the smoking room was associated with air quality in adjacent indoor spaces.

Determinants of SHS leakage were identified by a multivariate linear regression model. The direction of the smoking room door being opened was the most important determinant in the increase in PM_2.5 concentrations outside the smoking room. SHS leakage was statistically significant when the smoking room door was opened outwards. The average PM_2.5 concentration inside the smoking room was a factor in SHS leakage. Although the coefficient of the average PM_2.5 concentration inside the smoking room was small, 0.0196, its contribution to SHS leakage was significant due to the extremely high PM_2.5 concentration. The duration of how long the smoking room door was opened until the smoking room door was closed was statistically significant. The longer that the smoking room door was open, the more SHS leakage accumulated and increased. Door pumping was the main reason for SHS leakage, and an open doorway configuration yielded SHS exposure (Wagner et al., 2004). The pumping action of the door as it is opened and closed can cause noticeable leakage of SHS by pushing SHS out of the smoking room (Wan et al., 2010).

There were some limitations to this study. Although there are various types of smoking rooms, assessment of SHS leakage was conducted in one indoor smoking room in an Internet café. However, the smoking room in this study was the most common type of smoking room among over 202 smoking rooms surveyed (Kim and Lee, 2016). Additionally, we did not measure the ventilation rate of the smoking room in this study. Although the ventilation rate could influence PM_2.5 concentrations inside and outside the room, it might not significantly influence the impact of other determinants.

Funding

The first author was partially supported by National Research Foundation of Korea (BK21 PLUS, No. 22A20130012682).

Additional information

Funding

The first author was partially supported by National Research Foundation of Korea (BK21 PLUS, No. 22A20130012682).

Notes on contributors

Hyejin Kim

Hyejin Kim and Jaehoon An are graduate students at the Graduate School of Public Health, Seoul National University.

Kiyoung Lee

Kiyoung Lee is a professor at the Graduate School of Public Health, Seoul National University, Seoul, Korea.

Jaehoon An

Hyejin Kim and Jaehoon An are graduate students at the Graduate School of Public Health, Seoul National University.

Sungho Won

Sungho Won is an associate professor at the Graduate School of Public Health, Seoul National University.