Estimates of cancer risk from chloroform exposure during showering in Taiwan

Estimates of cancer risk from chloroform exposure during

showering in Taiwan

H.-W. Kuo

, T.-F. Chiang

, I.-I. Lo

, J.-S. Lai

, C.-C. Chan

, J.-D. Wang

Institute of En¨ironmental Health, China Medical College, No. 91, Hsueh-Shin Road, Taichung, Taiwan b

Institute of Occupational Medicine and Industrial Hygiene, College of Public Health, National Taiwan Uni¨ersity, Taipei, Taiwan

Received 14 December 1997; accepted 2 February 1998

Abstract

The purpose of this study was to compare the cancer risk with chloroform exposure during showering. The study concentrated on the three major metropolitan areas of Taiwan. Total exposure was measured based on a combination of ingestion, inhalation and skin absorption. A total of 137 tap water samples were taken from 26

Ž . Ž . Ž .

locations within the Taipei north , Taichung central and Kaohsiung south areas. Analysis of VOC compounds Ž was performed according to the US EPA Method 524. Chloroform concentrations were highest in Kaohsiung 60.19

. Ž . Ž .

mgrl , followed by Taipei 18.83 mgrl and Taichung 17.55 mgrl . Based on the two-resistance theory to volatilization in showers, when air flow rate is increased, chloroform concentrations in the air significantly decrease.

Ž .

A 10-min shower would result in chloroform exposure with a 3:4:3 ratio ingestion, inhalation, skin absorption . However, that changes to 1:7:2 for a 20-min shower under the same conditions. The cancer risk was highest in Kaohsiung at 17.59 per million for a 10-min shower and 64.77 per million for a 20-min shower. The lowest cancer risk was found in Taichung at 4.99 and 11.50 per million for a 10- and 20-min shower, respectively. Although ingestion is commonly considered to be the primary source of exposure to chloroform from tap water, inhalation and skin absorption exposure concentrations were found to be even higher. Q 1998 Elsevier Science B.V. All rights reserved.

Keywords: Chloroform; Cancer risk; Showering

1. Introduction

Tap water in Taiwan has been chlorinated for approx. 30 years. This has caused the production of relatively small amounts of chloroform and

U_{Corresponding author. Tel.:}

q886 4 054076; fax: q886 4 2019901; e-mail: wukuo@cmce.cmcedu.tw

Ž .

other disinfectant by-products DBP , such as

Ž .

haloacetic acids HAAs , haloacetonitriles

ŽHANs , chloropicrin CPK , chlorohydrate CH. Ž . Ž .

Ž .

and other disinfectant by-products Chiang, 1995 . Due to poor water quality from rivers and under-ground sources, the Taiwan Water Supply

Com-Ž .

pany Taiwan Water Supply Corp., 1995 has

added extra amounts of chlorine during treat-ment, with concentrations at their peak during

0048-9697r98r$19.00 Q 1998 Elsevier Science B.V. All rights reserved. Ž .

summer. Consequently, disinfectant by-products are being produced in dangerously high amounts.

Ž .

Morris et al. 1992 used meta-analysis to find a positive association between consumption of chlo-rination by-products in drinking water and blad-der and rectal cancer in humans. In a 1995 study on trihalomethanes in drinking water and tumors

Ž .

in animals, Attias et al. 1995 estimated a carci-nogenic risk for different THM combinations

which vary from 2.7=10y7 to 4.6=10y6 per

Ž

mrl. Results from studies Aschengram et al., .

1993; Bove et al., 1995 involving congenital

anomalies, stillbirths, and neonatal deaths in rela-tion to quality of drinking water were not statisti-cally stable. Although traditional risk assessments for water often only considered exposure from ingestion, in order to accurately measure total exposure to THMs inhalation and dermal absorp-tion need to be included as well. It has been shown that showering increases body burden due to inhalation, exposure and dermal absorption ŽWeisel and Jo, 1996 . Little 1992 applied the. Ž . two-resistance theory to contaminant volatiliza-tion in showers and found several factors can affect volatility including: the contaminant’s con-centration in water and its physical and chemical characteristics, volume of water and air exchange rate in the shower, water flow rate, water temper-ature, etc. Measuring these factors as well as others provides a more accurate assessment of variations in contaminant volatility. The two-resistance theory is applied in the current study to estimate total chloroform exposure and calcu-late the cancer risk for a comparison of the three major metropolitan areas in Taiwan. These re-sults can then be used by the government as a reference for regulation purposes.

2. Materials and methods 2.1. Study areas

The study areas which included the cities of Taipei, Taichung and Kaohsiung contain one-quarter of the total Taiwanese population. Due to the overall poor water quality throughout the island, tap water is not directly consumed so ‘drinking water’ is defined for purposes of this

study as tap water that has been boiled. A total of 137 samples of tap water and 68 samples of drinking water were taken from the three study sites.

2.2. Methods

Twenty-four purgable VOCs were analyzed us-ing modified US EPA Method 524. Details re-garding quality control and procedures can be

Ž .

found in published studies by Kuo et al. 1996

Ž .

and by Kuo et al. 1997 . Total exposure to

chloroform is divided into three major routes: ingestion, inhalation, and dermal absorption. Parameters are based on those reported by Little Ž1992 and by Beavers et al. 1996 . Ingestion was. Ž .

assumed to be 2 lrperson per day. One person

was estimated to have a body weight of 70 kg and an expected life span of 75 years. Inhalation concentrations were found to be based on the following estimates: volume of air in the shower ŽVs. was 1.2 m , volumetric water flow rate Q3 Ž L.

Ž .

was 5 lrmin, volumetric flow rate QG was 50

Ž .

lrmin, air exchange rate ACH was 2.5 timesrh,

Ž .

water temperature T_w was 448C, height of

shower head was 1.8 m and overall mass transfer

Ž .

coefficient with a liquid phase basis K_{OL A} was 7.4 lrtime. The temperature correction

coeffi-Ž .

cient for Henry’s law constant J was 0.1312

Ždimensionless . Estimated concentration. of

Ž .

chloroform in the air Y was:s

Ž . Žybt.

Y_ssarbq yarb exp

Ž ŽyN ._. as Q C 1yexp

_

₄

_

₄

. 4 Ž . Ž 3 .

qY r2 =T min =20 m rday =50%. The Ys si

Žinitial concentration in the shower air at ts0 of. chloroform was assumed to be 0. The volume of air inhalation was assumed to be 20 m3_{rday. The}

absorption efficiency in alveoli was assumed to be

59%. The shower timesT.

Dermal absorption was based on the assump-tion of the total surface area of skins18 000 cm2

total surface area of skin. The permeation

con-stant of the skin was assumed to be 0.2 cmrh.

The chloroform concentration in the water was

C . The skin absorption concentration was esti-_w

Ž . Ž . Ž 2.

mated to be T min =0.2 cmrh =18 000 cm

Ž

=80%=C . The equation risk estimation ofw

.

chloroformschloroform exposure=slope factor

was used to find the number of persons who may acquire cancer during one’s life span due to chloroform exposure from ingestion, inhalation, and dermal absorption based on a population of 1

Ž .

million Bull, 1990 .

3. Results and discussion

Drinking and tap water from the three major metropolitan areas of Taiwan were measured for THMs, 1,2-dichloroethane and benzene. Results are shown in Table 1. Based on measurements from tap water, the cancer risk was highest in

Ž . Ž .

Kaohsiung 376.70 , followed by Taipei 220.73

Ž .

and Taichung 161.77 . However, according to

drinking water measurements Taipei was highest Ž106.63 , then Kaohsiung 87.69 and Taichung. Ž . Ž63.10 . Although chloroform was found in high-. est concentrations in tap and drinking water among all THMs, its carcinogenic potential was lowest. The cancer risk was highest for bromod-ochloromethane and dibrombromod-ochloromethane. Be-cause benzene and bromoform were found in the lowest concentrations in tap and drinking water, their respective cancer risks were insignificant.

According to a Taiwan Department of Health

Ž .

report Taiwan Department of Health, 1996 , an island-wide comparison of mortality due to cancer was highest in the southern region including Kaohsiung city and county. The most common forms were lung, liver, bladder, and skin cancer. Two possible factors for the high cancer mortality in the southern region are due to its vicinity to Taiwan’s petrochemical industry and the poor quality of its underground and tap water. The current study found a significant correlation

Ž .

between standardized mortality ratio SMR of bladder cancer among males and THM

concen-Ž .

trations in tap water rs0.48, P-0.01 . Morris

Ž .

et al. 1992 used meta-analysis to find a signifi-cant relationship between bladder and rectal

can-Table 1

Estimates of cancer risk from tap and drinking water from the three metropolitan areas

y6 Ž . VOCS Area Cancer risk 10

Tap water Drinking water THMS Taichung 131.69 20.34 Taipei 197.73 83.41 Kaohsiung 301.32 49.32 CHCl₃ Taichung 5.97 1.310 Taipei 6.40 2.40 Kaohsiung 23.52 4.89 CH Cl Br_{2 2 2} Taichung 101.90 14.89 Taipei 144.08 66.08 Kaohsiung 219.78 36.19 CH ClBr_{2 2} Taichung 23.58 4.16 Taipei 47.45 14.93 Kaohsiung 58.02 8.24 CHBr₃ Taichung 0.24 NA Taipei NA NA Kaohsiung NA NA C H Cl2 2 2 Taichung 29.90 42.34 Taipei 22.74 22.96 Kaohsiung 75.38 37.97 C H6 6 Taichung 0.18 0.42 Taipei 0.06 0.26 Kaohsiung ND 0.40

Total risk Taichung 161.77 63.10 Taipei 220.73 106.63 Kaohsiung 376.70 87.69

cer and THM concentrations in drinking water. In studies concerning THM ingestion published

Ž .

in 1988 and 1981, Zierler and Danley 1988 and

Ž .

Wilkins and Comstock 1981 found that when chloroamide was added to the water supply, there was a significant increase in the incidence of and mortality due to bladder cancer. However, in a

Ž .

1995-study in Italy by Attias et al. 1995 no

correlation was found between THM concentra-tions in drinking water and the incidence of blad-der cancer. Because numerous factors lead to the incidence of cancer, it is difficult to find a clear correlation between ingestion of THMs and can-cer and to accurately measure the body’s total exposure to THMs. Mathematical models based on animal and epidemiological data analyses can be used to calculate cancer risk from THMs con-centration in drinking water which can then be used by the government for regulation purposes.

Ž3 . Using air flow rates of 0, 50 and 150 l rmin chloroform concentrations in the air were esti-mated during 10- and 20-min showers. Results are displayed in Table 2. Chloroform concentrations increased when shower time increased and de-creased when air flow rate inde-creased. Based on a 10-min shower in Kaohsiung with a 60.19 chloro-form concentration in water and 0 air flow, 1.34 mgrl of chloroform was produced in the air. When air flow was increased to 50 l3_{rmin, the}

chloroform produced decreased to 1.10 mgrl

Ž18.0% . At 150 l. 3rmin, chloroform decreased to

Ž .

0.78 mrl 42% from the original 0 air flow rate. Based on a 20-min shower with 0 air flow rate, the chloroform concentration in air was 2.45mgrl. With an air flow of 50 l3_{rmin, the chloroform}

concentration in air decreased to 1.71 mgrl

Ž30.2% . At 150 l. 3rmin, the concentration

de-Ž .

creased to 0.96 mgrl 60.8% from the original 0

air flow rate. In addition to air flow rate, many factors affect chloroform concentrations in the air which are difficult to control, such as water flow rate, volume of air in shower, water temperature, height and diameter of shower head, and chloro-form concentration in the water. Although in-creasing air flow rate is possibly the easiest and most effective way to decrease chloroform con-centrations in the air, the general population in Taiwan overlooks or is unaware of this fact. Due to the relatively small volume of shower space in Taiwan, chloroform concentrations in the air can

Ž .

be exceptionally high. Pellizzari et al. 1986 re-ported that concentrations of certain VOCs pre-sent in indoor air have been found to be more than 10 times higher than outdoors. One poten-tial source of VOCs in the air is the transfer from contaminated tap water during residential water use in showers, dishwashers, and washing ma-chines. According to a 1993-study by Wallace et

Ž .

al. 1993 , the possibility of chloroform inhalation increased significantly when dishwashers and washing machines were in operation.

The two-resistance theory was applied to calcu-late total exposure of chloroform due to skin absorption and inhalation during showering. Table 3 shows a comparison of the three study sites

Ž

based on total exposure ingestion, inhalation and .

skin absorption and the cancer risk after 10- and 20-min showers. Results showed total exposure and cancer risk increased with shower time. The exposure ratio was 3:4:3 for a 10-min shower and 1:7:2 for a 20-min shower. As shower time in-creased, exposure from inhalation and skin ab-sorption increased significantly when water con-tained high concentrations of chloroform. Be-cause chloroform concentrations in the water are highest in Kaohsiung, increasing shower time from 10 to 20 min, cancer risk can increase four times Žfrom 17.59 to 64.77 compared to only twice in.

Ž .

Taipei from 6.42 to 13.27 .

These results are consistent with those found

Ž .

by Andelman 1985a,b who used a

one-compart-Table 2

Estimate of chloroform concentrations in the air during showering based on different air flow rates

3

Ž .

Air flow rate Area Chloroform conc. Chloroform conc. in air mgrm 3

Žlrmin. in waterŽmgrl. _{10 min 20 min}

0 Taichung 17.55 3.9 7.2 Taipei 18.83 4.2 7.7 Kaohsiung 60.19 13.4 24.5 50 Taichung 17.55 3.2 6.9 Taipei 18.83 3.4 5.3 Kaohsiung 60.19 11.0 17.1 100 Taichung 17.55 2.3 2.8 Taipei 18.83 2.4 3.0 Kaohsiung 60.19 7.8 9.6

Table 3

y6

Ž . Ž .

Total exposure mg and cancer risk 10 from chloroform exposure during showering in the three metropolitan areas Taichung Taipei Kaohsiung % of total exposure

Chloroform conc. in 17.55 18.83 60.19

Ž .

water mgrl Shower time: 10-min

Ingestion 7.72 14.14 28.76 30

Inhalation 13.19 14.58 45.83 43

Skin absorption 8.42 9.04 28.89 27

Total dose 29.83 37.76 103.48

Cancer risk 4.99 6.42 17.59

Shower time: 20-min

Ingestion 7.72 14.14 28.76 12

Inhalation 43.06 45.83 294.44 67

Skin absorption 16.85 18.08 57.78 21

Total dose 67.63 78.05 380.98

Cancer risk 11.50 13.27 64.77

ment indoor air quality model to estimate that the uptake of VOCs inhaled was six times greater

Ž .

than that ingested. Beavers et al. 1996 estimated inhalation and ingestion of benzene and three other aromatic hydrocarbons typically found in gasoline-contaminated water. Results showed the

Ž

ratio was 3:4:3 ingestionrinhalationrskin

ab-.

sorption and shower-related and

non-shower-re-lated exposure to benzene was 55]45%.

Volatilization of chemicals from indoor water uses is of growing concern, particularly as water sup-plies become increasingly contaminated. Expo-sure ratios can vary due to differences in assump-tions and parameters used in mathematical

mod-Ž .

els. Maxwell et al. 1991 concluded that the ratio of the lifetime inhalation to ingestion doses is

Ž

probably in the range of 0.6]1.5 and may be as .

high as 5.7 and the ratio of the lifetime dermal to ingestion doses is probably 0.3 but may be as high

Ž .

as 1.8. Jo et al. 1990 quantified the relationship between chloroform concentrations in shower wa-ter, shower air, and exhaled breath of individuals exposed while showering. Data showed that the chloroform increase in shower air and breath concentrations increased with chloroform

concen-Ž .

trations in water. Chinery and Gleason 1993 Ž

used shower and PB-PK physiologically based

.

pharmacokinetic models to predict breath con-centration and the absorbed dose of chloroform after exposure while showering. They estimated that the ratio of dermally to inhaled absorbed doses ranged from 0.6 to 2.2 and the expected value was 0.75. They suggested a reasonable value of skin permeability coefficient for chloroform used in the simple steady-state model would be 0.2 cmrh.

Ž .

In a 1983-study, Mackay and Yeun 1983 noted that the rate of the volatilization of a chemical form is dependent on its molecular]diffusivity properties. A two-resistance model is used to describe the process in which the volatilizing chemical has to first diffuse across a liquid film at the air]water interface, followed by diffusion across the air film. The water]air interfacial areas and temperature of the water uses are critical determining factors in the rate of mass transfer. Other indoor water uses involve different quanti-ties and flows of water, residence times in the water appliances, degrees of mixing and turbu-lence and temperatures. So, the extents of volatilization among the water uses, even for a given chemical, should vary. Alveolar breathing sampling can be used to verify the post-shower breath data, help assess the importance of

dif-Ž

ferent routes of exposure e.g. proportion of der-.

mal vs. inhalation routes , establish uptake and

Ž .

eliminate kinetics. Lindstrom and Pleil 1996

used alveolar breathing measurements to calcu-late blood-contaminate concentrations and com-partment-specific half-lives, confirm minimum ab-sorbed dose and characterize significant short-term exposure phenomenon. Even though most parameters followed were established in previous studies, our findings are reliable and can be used by the government to regulate VOC concentra-tions in the water and to inform the public about certain safety measures that can minimize expo-sure to VOCs. Future studies using the two-resis-tance theory should establish parameters uniquely suited for Taiwan in order to make accurate cancer risk assessments due to VOC exposure.

4. Conclusion

In Taiwan, standards are set for ingestion of water but not for other routes of exposure, such as inhalation and dermal absorption. This is the first study to estimate the cancer risk from expo-sure to VOCs in Taiwan and the chloroform exposure during showering. Cancer risk results showed Kaohsiung was highest for tap water Ž376.70 and Taipei for drinking water 106.63 .. Ž . Based on the two-resistance theory, a 10-min shower resulted in chloroform exposure with a

Ž

3:4:3 ratio ingestion, inhalation and skin absorp-.

tion and 1:7:2 for a 20-min shower. As shower time increased, exposure due to inhalation creased significantly. Yet, as air flow rate in-creased concentrations of chloroform dein-creased. Because the government does not actively pro-mote efforts to decrease chloroform in tap water, the general population must make efforts to de-crease personal exposure, such as increasing air flow and decreasing the duration of shower time.

Acknowledgements

This study was supported by a special grant from National Science Council NSC 85-2621-p-039-001-Z.

References

Andelman JB. Human exposure to volatile halogenated or-ganic chemical in indoor and outdoor air. Environ Health Perspect 1985a;62:313]318.

Andelman JB. Inhalation exposure in the home to volatile organic contaminants of drinking water. Sci Total Environ 1985b;47:443_]460.

Aschengram A, Zierier S, Cohen A. Quality of community drinking water and outcome of late adverse pregnancy outcome. Arch Environ Health 1993;48:105]113.

Attias L, Contu A, Loizzo A, Massiglia M, Valente P, Zapponi GA. Trihalomethanes in drinking water and cancer: a risk assessment and integrated evaluation of available data, in animals and humans. Sci Total Environ 1995;171:61]68. Beavers JD, Himmelstein JS, Hammond SK, Smith TJ,

Ke-nyon EM, Sweat CP. Exposure in household using gasoline-contaminated water. J Occup Environ Med 1996; 38:35]38.

Bove F, Fulcomer MC, Sarrin JE. Public drinking water con-tamination and birth outcome. Am J Epidemiol 1995; 141:850]862.

Bull RJ. Evaluation of health risk association with disinfec-tion. Crit Rev Environ Control 1990;20:77_]113.

Chiang TG. By-product of drinking water and treatment tech-niques in Taiwan area. Taiwan EPA, EPA84-E3J1-0901, 1995.

Chinery RL, Gleason AK. A compartmental model for the prediction of breath concentration and absorbed dose of chloroform after exposure while showering. Risk Anal 1993;13:51]63.

Jo WK, Wiesel CP, Lioy PJ. Routes of chloroform exposure and body burden from showering with chlorinated tap water. Risk Anal 1990;10:575]580.

Kuo HW, Lo II, Chan CC, Lai JS, Wang JD. Volatile organic compounds in water near petrochemical factories in Tai-wan. Chemosphere 1996;33:913]920.

Kuo HN, Chiang TF, Lo II, Lai JS, Chan CC, Wang JD. VOC concentration in Taiwan’s household drinking water. Sci Total Environ 1997;208:41]47.

Lindstrom AB, Pleil JD. A methodological approach for expo-sure assessment studies in residences using volatile organic compound-contaminated water. J Air Waste Manage Assoc 1996;46:1058]1066.

Little JC. Applying the two-resistance to contaminant volatilization in shower. Environ Sci Technol 1992;26: 1341]1349.

Mackay D, Yeun ATK. Mass transfer coefficient correlations for volatilization of organic solutes from water. Environ Sci Technol 1983;17:211]217.

Maxwell NI, Barmaster DE, Ozonoff D. Trihalomethanes and maximum contaminant levels: the significance of inhalation and dermal exposure to chloroform in household water. Regul Toxicol Pharmacol 1991;14:297]312.

Morris RD, Audet AM, Angelillo IF, Chalmers TC, Mosteller F. Chlorination, chlorination by products and cancer. Am J Public Health 1992;82:955]963.

Pellizzari ED, Michael LC, Perritt RL, Smith D, Hartwell TD, Sebestik J. Comparison of indoor and outdoor toxic air pollutant levels in several Southern California communi-ties. Final Report on Contract EPA 68-02-4544, US Envi-ronmental Protection Agency, Washington, DC, 1986. Taiwan Department of Health. Cancer registry annual report.

Taiwan, ROC, 1996:12]15.

Taiwan Water Supply Corp. The statistical data and water quality. Taiwan, ROC, 1995.

Wallace L, Pellizzari ED, Goldon S. A liner model relating breath concentrations to environmental exposure:

applica-tion to a chamber study. J Expos Anal Environ Epidemiol 1993;3:75]102.

Weisel CP, Jo WK. Ingestion, inhalation, and dermal exposure to chloroform and trichloroethane from tap water. Environ Health Perspect 1996;104:48]51.

Wilkins JR, Jr, Comstock GW. Source of drinking water at home and site-specific cancer incidence in Washing county, Maryland. Am J Epidemiol 1981;114:178]190.

Zierler S, Danley RA. Bladder cancer in Massachusetts re-lated to chlorinated and chloroaminated drinking water. A case-control study. Arch Environ Health 1988;43:195]200.