Journal Archive

We investigated a total of 573 facilities, composed of 10 types of public institutions that are not currently subject to legal management by the “Indoor Air Quality Control in Public Use Facilities, etc. Act.” The first survey was conducted from March to October of 2005 and involved 338 facilities including movie theaters, restaurants, academies, performance halls, internet cafés, karaoke bars, and pubs. The second survey was conducted from June 2006 to August 2007 and involved 235 facilities including wedding halls, gymnasiums, exhibition halls and social welfare facilities such as handicapped care facilities, child welfare facilities, elderly welfare facilities and female care facilities (Table 1).

This study examined HCHO and five representative VOCs: benzene, toluene, ethyl benzene, xylenes and styrene. Target pollutants were classified into carcinogen and non-carcinogenic chemicals according to classification of carcinogenicity by the US EPA(1997). In this study, the carcinogens include formaldehyde (B1: Probable human carcinogen) and benzene (A: Human carcinogen) and the non-carcinogens include toluene, ethyl benzene, xylenes and styrene. All of the non-carcinogens are classified as D (not considered a human carcinogen) by US EPA.

Indoor air samples of target pollutants were collected during business hours at each facility, and the collection location within each facility was chosen based on ventilation, interior structures, and emission sources in each facility.

Samples of 5 VOCs were collected by personal air sampler (Sibata, Japan) with a Tenax sorbent tube (1/4′′×20 cm stainless steel, Supelco, USA). Samples were collected over a period of 30 minutes and sampler flow was 0.2 L/minute. After the sample was collected, the sorbent tube was put into Swagelok, sealed with Teflon caps on both ends and refrigerated at less than -75°C until analysis. VOCs were measured within a week of sampling, using GC/MSD (gas chromatography/mass selected detector, USA) connected to a thermal desorber within.

Indoor samples of HCHO were collected with a personal air sampler by installing an ozone scrubber cartridge to remove the interference of ozone in front of DNPH-silica cartridge (Waters Corp, USA) charging 350 mg DNPH (2,4-dinitrophenylhydrazone)-silica (1.0 mg DNPH) or 1.0cm(i.d.)×2.0cm(o.d.)×4.3cm (total length) cartridge. HCHO samples were collected at the same sampling location as VOCs, from a height of 1.0-1.5 m at a flow rate of 0.5 L/min for 30 minutes. After sampling, the DNPH cartridge was sealed with plastic caps on the both ends and stored at -70°C until analysis. The absorbed formaldehyde was derived and analyzed with HPLC (high performance liquid chromatography, USA).

This study involved a nationwide survey with three different institutions in charge of sampling and analysis in different regions of Korea. In order to QA/QC the results from several different institutions, the following measures were each performed twice: adjustment of the personal air pump, cleaning check of TVOC and HCHO sorbent tube and DNPH cartridge, tube-use history management of TVOC sorbent tube, and a RRT (Round Robin Test). An RRT is an inter-laboratory test and a reciprocal comparative test of a sample at a certain level. Samples for the first and the second RRT were collected from the same spot at the same time and were quantitatively analyzed for QA/QC of measurement and analysis at each of the institutions. Calibration curves were drawn of each substance, by institution, and the coefficient of determination of all substances was over 0.997 (r²). The first RRT showed that deviation of the range of TVOC and HCHO levels according to institution was less than 10% (Table 2).

In uncontrolled public facilities there was no statistically significant difference in variables such as region, allowance of smoking, location above-ground or underground, business hours, occurrence of construction, or use of air freshener and ventilation. Smoking did not significantly increase the IAQ because only 30% of the facilities allow smoking inside. Therefore the risk assessments did not take into account whether or not a facility allowed smoking indoors.

Human exposure from polluted indoor air should be calculated by considering contamination levels, daily inhalation rates, body weight, exposure frequency, exposure duration and life expectancy (Table 3). Daily inhalation rate was applied according to the characteristics of workers and users of each facility. Daily inhalation rates of users are cited in Korea Exposure Factors Handbook (Jang et al., 2007). The average daily inhalation rate for adults (13.4m³/day) was based on a lifetime chronic exposure assessment of adults from Korea Exposure Factors Handbook (Jang et al., 2007). For short-term exposure, the total volume of rest, sedentary exercise, light exercise, middle-intensity activity and vigorous exercise for men and women were estimated at 0.45m³/hr, 0.5m³/hr, 1.1m³/hr, 1.4 m³/hr and 3.2m³/hr, respectively (Jang et al., 2007). In addition, the characteristics of each facility were considered when calculating inhalation rates. There are no data on inhalation rates of workers in South Korea, so the ‘Exposure Factors Handbook’ of US EPA was used to estimate occupational inhalation rates (US EPA, 1997). Workers risks were calculated based on the inhalation rate of a chronic exposure via single route, 20m³/day (90^th upper confidence limit value) by the precautionary principle (US EPA, 1997).

Body weight was estimated using the average body weight by age as provided by the Korean Agency for Technology and Standards (KATS, 2004). Age distribution at each facility was determined by questionnaire. Average life span was estimated to be 70 years which is the average life expectancy in Korea. Exposure time and duration were based on survey data collected from a questionnaire completed by facility users (customers) and workers during the sampling periods.

Exposure frequency was divided into three groups according to exposure scenario. Group 1 was composed of workers with the highest exposure frequency (Worker Exposure Scenario; WES), group 2 was composed of frequent users (User Worst Exposure Scenario; UWES) and group 3 was composed of average users whose exposure was expected to be low (User Average Exposure Scenario; UAES).

Human exposure based on exposure scenario in each facility was calculated by modifying the formula slightly to reflect exposure conditions according to exposure time (1).

Where LADD: Lifetime average daily dose (mg/kg/day) 　 　　C_IA : Concentration of chemicals in indoor air at facility (mg/m³) 　 　　IR_kj : Inhalation rate for exposure scenario, k and facility, j (m³/hr) 　 　　ET_kj : Exposure time for exposure scenario, k and facility, j (hrs/day) 　 　　EF_kj : Exposure frequency for exposure scenario, k and facility, j (days/yr) 　 　　ED_kj : Exposure durations for exposure scenario, k and facility, j (yrs) 　 　　BW_i : Body weight at age, I (kg) 　 　　AT : Average time for lifetime (days)

Cancer potency (q₁*) is the probability of cancer calculated by unit body weight and per dose as a 95% upper limit of linear coefficient in a dose-response curve. Unit risk is an excess cancer risk which can be estimated when a healthy adult is exposed to air contaminated by a unit concentration (1 μg/m³) of a pollutant.

Formaldehyde was analyzed using a cancer potency of 4.6×10^-2 (mg/kg/day)^-1 in a linearized multistage model based on dose-response data from Kerns et al. (1983). The cancer potency of benzene (WHO, 2000) was 3.6×10^-2 (mg/kg/day)^-1 (Table 4).

The risk of non-cancer pollutants was assessed with reference concentration (RfC), which is defined as an exposure reference level that does not result in toxic effects after lifetime exposure to the substance. RfC of toluene was determined to be 46 mg/m³ as No Observed Adverse Effect Level (NOAEL), which means that 46 mg/m³ is the lowest concentration that does have a toxic effect on humans following inhalation exposure to toluene (Neubert et al., 2001; Cavalleri et al., 2000). The inhalation exposure reference level or RfC of toluene was calculated to be 5 mg/m³ when taking into account uncertainties associated with toxicity levels. Inhalation exposure references of non-cancer pollutants were determined by following this same procedure. RfC values for (Korsak et al., 1994; Mutti et al., 1984; Hardin et al., 1981) each substance are presented in Table 5.

Assumption of risk is a process that combines the results of dose-response assessment and exposure assessment in risk assessment and then determines lifetime Excess Cancer Risk (ECR) and carcinogens hazard quotient (HQ) of non-cancer pollutants after an actual exposure to a pollutant (US EPA, 1987). Lifetime ECR and HQ for various exposure scenarios in the facilities were calculated using the following US EPA formula (US EPA, 1997):

When calculating the HQ of non-cancer pollutants, the reference concentration (RfC, mg/m³) should be converted to reference dose (mg/kg/day) to reflect human exposure units. RfC is calculated based on an average adult in the U.S.; therefore, RfC was converted based on average body weight (70 kg) and daily inhalation rate (20m³/day) for the U.S. (US EPA, 1987).

Health protection criteria for uncontrolled public facilities were determined in three steps. In the first step, the reasonable maximum toxicity reference dose that does not result in harmful effects on humans was determined. After reviewing several toxicity information databases such as US EPA IRIS and WHO, the highest reference or guidance value was selected. The purpose of this first step is to establish the upper limit of toxicity for each substance for health protection.

In the second step, the indoor air target level was calculated as part of establishing the target risk goal. This step is similar tithe concept of preliminary remediation goals as described in the US EPA risk assessment guidelines for Superfund, human health evaluation manual (RAGS/HHEM), part B (US EPA, 1991). The target level is derived from risk-based calculations that consider exposure patterns for users and workers of subject facilities. The target level was calculated using the following US EPA formula (US EPA, 1991):

Target risk is the acceptable or achievable risk goal, described as a function of the acceptable individual lifetime excess cancer risk for carcinogens and acceptable hazard quotient for non-carcinogens. Target excess cancer risks for workers and users are 1×10^-4 and 1×10^-5 respectively, and target HQs (Hazard Quotients) for non-carcinogens are 1.0 and 0.1 for workers and users, respectively. The US EPA suggests that is the target HQ be used to calculate a risk-based goal at a pre-specified HI (Hazard Index) of 1.0 for multimedia and multiroutes. If the HI for multimedia and multiroutes is not available, target HQ can be calculated using the relative risk contribution (RRC) by single medium and route to total HI. The default value of target HQ is 0.1 (considering 10% RRC to total HI) for a single exposure medium and pathway (US EPA, 2001). We used 0.1 as the default value of single exposure media target HQ for a facility’s user.

In the third and final step, the lowest levels between the reasonable maximum toxicity reference dose (calculated in the first step) and the risk-based target level (calculated in the second step) are selected according to substance and use patterns in the facility. Finally, health protection criteria are determined by rounding down to the lowest level of each substance.

Table 6 shows the concentration of target pollutants in the facilities. The average level of formaldehyde in each facility ranged from 23.4 μg/m³ to 141 μg/m³. The highest average value of formaldehyde was in performance halls (141 μg/m³), with levels in exhibition halls (112 μg/m³) and pubs (84.5 μg/m³) slightly lower. The high levels of formaldehyde in indoor air are probably caused by indoor materials and human activities (Weng et al., 2009). The high levels in performance and exhibition halls is likely caused by stage and display exhibit remodeling, which involves the use of wood, glue, paint and new products. High levels of carbonyls shortly after refurbishment might be due to emissions from décor and construction materials(Weng et al., 2009). Cooking food for customers (Zhang and Smith, 1999) and smoking in bars also increased the levels of HCHO (Jones, 1999; Godish, 1991). In academies, tables, chairs, school supplies, electronics and books contributed to higher levels of formaldehyde (Fantuzzi et al., 1996).

When indoor HCHO levels were examined according to country, the levels of HCHO in Brazil (at a workplace in a university) and China (in a hotel) were 22.5-161.5 μg/m³ and 26.3-63.0 μg/m³ respectively (Cavalcante et al., 2005; Feng et al., 2004). Weng et al. (2009) found that HCHO levels in theaters (65.2-114.6 μg/m³) tend to be slightly higher than levels observed in this study (54 μg/m³). Conversely, indoor HCHO concentrations in cinemas were much higher in this study than compared with Weng et al.’s work in China (2009) (Table 7).

The occurrence and concentrations of HCHO and VOCs in homes can be affected by indoor sources, human activities, ventilation rates, and seasonal factors such as temperature changes and humidity (Van der wall et al., 1997). HCHO and VOC levels in this study were higher than those reported in previous studies, although the facilities, seasons and places of the studies were not same.

For VOCs, benzene and toluene levels were highest in internet cafés (29.9 μg/m³ and 186.8 μg/m³, respectively) and pubs (29.4 μg/m³ and 177.9 μg/m³, respectively). Benzene levels ranged from 1-30 μg/m³ and toluene ranged from 40-180 μg/m³. The internet cafés and pubs that were sampled are located underground, allowed smoking, and had a lack of ventilation. Furniture, electronics, carpeting, air fresheners and cooking were additional sources of VOCs (Chao and Chan, 2001; Wolkoff and Nielsen, 2001). Many other toxic VOCs such as xylenes have been identified in ETS (Environmental Toxic Substance) (Xie et al., 2003). It is also regarded as an important source of indoor pollution: smoking has been identified as the major indoor source of benzene, contributing an average of 2-3 μg/m³ to total indoor air concentrations (Ilgen et al., 2001a). The ranges of benzene (1-30 μg/m³) reported in this study were similar to those reported by Bruinen de Bruin et al. (2008), but higher than those measured by Guo et al. (2003) (Table 7).

Tables 8 and 9 show the lifetime health risks of carcinogens and non-cancer pollutants.

Lifetime ECRs of HCHO (Table 8) and benzene (Table 9) according to exposure scenario in each facility were higher in workers (who spend more time in a facility) than in users (who are typically exposed for less time) except in the case of social welfare facilities. In a social welfare facility, the risks of carcinogens were higher for users than for workers because the users are living in the facility 24 hours per day, compared to workers who only spend 8 hours/day in the facility.

When looking at lifetime ECR of HCHO according to exposure group, the risk to workers was higher in performance halls (1×10^-3) than in movie theaters, restaurants, academies, internet café, karaoke bars and pubs (1×10^-4) and in wedding halls and gymnasium (1×10^-5). In restaurants, academies, performance halls, internet café, karaoke bars, pubs, gymnasiums, exhibition halls and social welfare facilities, the risks of frequent users and average users were calculated to be 1×10^-4 to 1×10^-5 respectively (Table 8). Only users of movie theaters and workers in social welfare facilities were below government standards, with a risk of one per million (1×10^-6). Weng et al. (2009) reported risk of 1×10^-3 to both workers and users, illustrating that the risk to low-frequency users was as high as the risk to workers. In movie theaters, the risk to workers and frequent users was 1.20×10^-4 and 9.34×10^-6, respectively. The risk of HCHO exposure for users in movie theaters reported by Weng et al. (2009) (average risk 4.4×10^-4) was higher than that of this study (average risk 3.58×10^-6). This difference was likely caused by frequency of use patterns and background levels in the facilities.

The lifetime ECRs of HCHO of workers and frequent users in restaurants, academies, performance halls, karaoke bars and pubs was found to be 1×10^-4 and 1×10^-5, respectively. The highest HQ of toluene for frequent users was found in restaurants, which was likely caused by cooking and smoking in these facilities. In addition, the lifetime ECR may have been influenced by an increase in the number of individuals eating in restaurants.

The high risk of HCHO exposure in academies and performance halls was likely caused by various sources of pollution such as tables, chairs and construction materials in academies and equipment in performance halls. In addition, repeated exposure for 20-25 days per month led to higher risk in academies. In karaoke bars and pubs, the underground location, lack of ventilation, and presence of smoking in a small space contributed to the high risk, rather than exposure time or frequency. In exhibition halls, both workers and frequent users had high risks, averaging 1.17×10^-4 and 1.21×10^-4, respectively. This was caused by various new products, stage refurbishment and exhibition equipment. Lifetime ECRs of formaldehyde for frequent and average users in the social welfare facilities were 1×10^-5 and 1×10^-6, respectively, and were higher than the values reported for workers in the same facilities.

The risk of benzene exposure for workers in the restaurants, academies, performance halls, internet café and pubs were estimated at 1×10^-4 and the risk to workers in theaters and karaoke bars were 1×10^-5 (Table 9). Risk to users in wedding halls, gymnasiums, exhibition halls and social welfare facilities were within safe limits (1×10^-6) due to low frequency of use. However, the risks of formaldehyde exposure to frequent and average users in the social welfare facilities were higher (1×10^-5) than that of workers. As said before, this is because users spend more time in the facility than workers. Previous risk assessments in the same facilities included in this study are lacking, so the results of this study were compared with risks assessments of similar public facilities. According to Weng et al. (2009), the risk of formaldehyde exposure was high to workers and users of shopping centers (1×10^-3) and workers and users of supermarkets (1× 10^-4). These high exposure risks in public places may contribute to increased risk of cancer in the general population (Weng et al., 2009).

Facilities were divided according to the age of their users (adults, adolescents, all age groups and sensitive groups) based on information collected from a two-year survey of user ages. VSDs (Virtually Safety Doses) of each substance were then determined (Table 10). By comparing the levels with health protection upper limits considering toxicity, stricter levels were selected as health protection criteria. Additionally, feasibility of the Act including the criteria was also considered. In the results of each facility according to exposure scenario, a level of over 1,000 μg/m³ followed low average concentration, low exposure frequency and short exposure period and difference in exposed subjects by facility in making exposure scenario for risk assessment, so it poorly reflected the reality. To supplement these, health protection guidelines are suggested by considering health protection upper limits considering toxicity, health protection upper limits considering exposure and feasibility. VSDs of offices and bars (the only facilities used by adults only) are presented in the following table according to facility and substance. Facilities were classified by user age as follows: those used only by adults (offices and bars), those used by adolescents (academies, internet café and karaoke bars) and those used by sensitive groups (social welfare facilities). The health management criteria of the facilities for sensitive groups (social welfare facilities) were the most conservatively.

The most conservative value for each pollutant is selected for the health protection criteria as follows. The guidelines for health protection in indoor air are suggested to be 60 (max 100) μg/m³ for formaldehyde, 400 (max 500) μg/m³ for TVOCs, 10 (max 20) μg/m³ for benzene, 150 (max 170) μg/m³ for toluene, 400 (max 500) μg/m³ for ethyl benzene, 100 μg/m³ for xylenes, and 130 (max 170) μg/m³ for styrene. Indoor air concentrations in excess of the health protection criteria are shown in Table 11.

Formaldehyde is classified as a legally controlled substance with a standard of 100 μg/m³. There are no guidelines for each VOC in Korea, and the recommended guidelines for TVOCs are less than 400 μg/m³ in sensitive facilities such as medical institutions, 500 μg/m³ in subway stations and 1,000 μg/m³ in indoor parking lots. As the risk assessment conducted for this study shows that the risk of formaldehyde exposure is high, guidelines for formaldehyde exposure are suggested to be 60 (for facilities used by sensitive groups and adolescents)-100 (for facilities done only by adults) μg/m³. For VOCs, the current criteria for TVOCs (400 μg/m³) can be used for all facilities, including sensitive facilities such as medical institutions. In addition, this study shows that risk of exposure to benzene is also high and needs to be managed. The suggested exposure limit is less than 10 μg/m³.

In Germany the guidelines are presented as ranges rather than specific values. The German Ministry of Environment, Nature Conservation and Nuclear Safety suggests a range of 200-300 μg/m³, assuming long-term exposure. Guidelines are not provided for formaldehyde and benzene. Poland’s Ministry of Health and Social Welfare has no guidelines for TVOC and suggests less than 10 μg/m³ for benzene category A (24 h exposure), 20 μg/m³ for benzene category A (8 h exposure), 50 μg/m³ for formaldehyde category A (24 h exposure), and 100 μg/m³ for formaldehyde category A (8 h exposure). In Japan, there are no guidelines for benzene and the guidelines for formaldehyde and TVOC are less than 100 μg/m³ and 400 μg/m³, respectively (KEI, 2004). Finland’s Ministry of Environment does not provide guidelines for TVOCs. For VOCs, benzene, and formaldehyde the guidelines are set at less than 50 μg/m³ (Bruinen de Bruin et al., 2008).

However, unlike other countries, Japan presented references of each VOC as follows: less than 260 μg/m³ for toulene (compared to 150 μg/m³ in this study), 870 μg/m³ for xylenes (compared to 100 μg/m³ in this study), 3,800 μg/m³ for ethyl benzene (compared to 500 μg/m³ in this study) and 220 μg/m³ for styrene (compared to 150 μg/m³ in this study) (WHO, 2005). When these levels were compared with those suggested by this study, the levels of TVOC, toluene and styrene were similar while the levels of xylenes and ethyl benzene were different. This difference was caused by a difference in VOC concentrations in facilities, as well as a different approach to management and HRAs. However, in most countries the suggested guidelines for formaldehyde are 50-100 μg/m³, and guidelines for TVOCs either are not recommended or are 200-400 μg/m³. As a result, the levels suggested as guidelines are similar to reference levels suggested by this study, which were calculated from the risk assessment and based on survey results describing the actual conditions of public facilities.

One of the limitations of this study was that although indoor smoking was an important variable when considering exposure to benzene and other substances, it was not measured during business hours due to opposition from some business owners who believed that the measurements and results could hinder business. This means that the risk of exposure to pollutants in smoking facilities was almost certainly assessed at a lower level than actual conditions, which explains why the risks were lower than expected in some smoking facilities. In addition, human risk was assessed based on the assumption that an individual would be exposed to a level for all one’s life (70 years) (worst condition), and there are some uncertainties about this assumption (Guo et al., 2003). One of the limitations when deriving health protection criteria for the general population is that the relative risk contribution (RRC) is assumed to be 0.1, because low frequency and exposure time were conservatively estimated to be approximately 2 hours for the general user.

Indoor pollutants vary from chemical substances to microorganisms and mineral fibers. The current ‘Indoor Air Quality Control in Public Use Facilities, etc. Act’ describes legal regulations and recommended guidelines for indoor air quality and contaminants. The five pollutants that are the subject of legal regulations include PM₁₀, CO₂, HCHO, TBC and CO. The five pollutants for which there are recommended guidelines include NO₂, Rn, TVOC, asbestos and ozone. Guidelines and management differ according to country (WHO, 2000) (Table 12). While formaldehyde, radon, PM₁₀ and VOCs are commonly controlled as major pollutants, the control of other substances varies greatly according to country (WHO, 2005). Guidelines are provided for biological pollutants such as mold and mites, which are known to provoke asthma and allergy in children, although actual references are presented in only a few cases. In addition, cigarette smoke is a pollutant that requires management, but most countries do not suggest guidelines.

In addition to the indoor pollutants and facilities described in this study, health protection criteria for various indoor pollutants related to environmental disease should be established through health risk assessment methodology for various indoor facilities.