DSpace at VNU: Contamination by arsenic and other trace elements in tube-well water and its risk assessment to humans in Hanoi, Vietnam

Contamination by arsenic and other trace elements intube-well water and its risk assessment to humans in Hanoi, Vietnam Tetsuro Agusaa, Takashi Kunitob, Junko Fujiharaa,1, Reiji Kubotaa,

Contamination by arsenic and other trace elements in

tube-well water and its risk assessment to

humans in Hanoi, Vietnam

Tetsuro Agusaa, Takashi Kunitob, Junko Fujiharaa,1, Reiji Kubotaa,

a Center for Marine Environmental Studies (CMES), Ehime University, Bunkyo-cho 2-5, Matsuyama 790-8577, Japan

b Department of Environmental Sciences, Faculty of Science, Shinshu University, 3-1-1 Asahi,

Matsumoto, Nagano 390-8621, Japan

c Research Center for Environmental Technology and Sustainable Development, Hanoi University of Science,

VNU Hanoi, 334 Nguyen Trai, Thanh Xuan, Hanoi, Vietnam Received 4 November 2004; accepted 23 April 2005

High concentrations of arsenic, manganese and barium were found in tube-well water and human

hair in suburban areas of Hanoi, Vietnam

Abstract

Concentrations of As and other trace elements and their association were examined in groundwater (nZ25) and human hair (nZ59) collected at Gia Lam District and Thanh Tri District, suburban areas of Hanoi, Vietnam, in September 2001 Concentrations of As in the groundwater ranged from !0.10 to 330 mg/l, with about 40% of these exceeding WHO drinking water guideline of 10 mg/l Also, 76% and 12% of groundwater samples had higher concentrations of Mn and Ba than WHO drinking water guidelines, respectively Arsenic concentrations in hair of residents in Gia Lam and Thanh Tri Districts (range 0.088e2.77 mg/g dry wt.) were lower than those in other As-contaminated areas of the world, but were higher than those of people in non-contaminated areas Concentrations of As and Mn in hair of some individuals from the Gia Lam and Thanh Tri Districts exceeded the level associated with their toxicity and, therefore, a potential health risk of As and Mn is a concern for the people consuming the contaminated water in this area Cumulative As exposure was estimated to be lower than the threshold levels at the present, which might explain the absence of manifestations of chronic As poisoning and arsenicosis in the residents of Gia Lam and Thanh Tri Districts To our knowledge, this study revealed for the first time that the residents are exposed not only to

As but also Mn and Ba from groundwater in the Red River Delta, Vietnam

Ó 2005 Elsevier Ltd All rights reserved

Keywords: Arsenic; Manganese; Barium; Groundwater; Human hair; Hanoi; Vietnam

1 Introduction Arsenic is not contained so much in the Earth’s crust but greatly concentrated in pyrite, hydrous Fe oxides

* Corresponding author Tel./fax: C81 89 927 8171.

E-mail address: shinsuke@agr.ehime-u.ac.jp (S Tanabe).

1

Present address: Department of Legal Medicine, Shimane

Univer-sity School of Medicine, 89-1 Enya, Izumo, Shimane 693-8501, Japan.

0269-7491/$ - see front matter Ó 2005 Elsevier Ltd All rights reserved.

doi:10.1016/j.envpol.2005.04.033

www.elsevier.com/locate/envpol

2002) It is known that As would be easily solubilized

in water from these minerals depending on pH, redox

conditions and temperature (Smedley and Kinniburgh,

2002) Nickson et al (1998) suggested that As may be

released from As-rich Fe oxyhydroxides into

ground-water under reductive conditions in alluvial sediments

in Bangladesh and West Bengal Increased usage of

groundwater for drinking in these areas has caused

serious health problems (Bagla and Kaiser, 1996;

Nickson et al., 1998), because inorganic As is

carcino-genic and causes skin and various internal cancers

(IARC, 1980; WHO, 2001) It is reported that about 36

million of the population in the Bengal Delta is at risk

from drinking As-contaminated groundwater (Smith

et al., 2000).Tondel et al (1999)reported a relationship

between As concentration in drinking water and

prevalence of skin lesions in Bangladesh A significant

doseeresponse relationship was also observed between

As level in well water and mortality from cancers of

bladder, kidney, skin, and lung in residents of the

southwestern coast of Taiwan (Wu et al., 1989)

Furthermore, recent studies suggest that As acts as an

endocrine disruptor at very low concentrations Stoica

et al (2000) reported that As activates estrogen

receptor-a through an interaction with the

hormone-binding domain at concentrations as low as 1 nM

(0.075 mg/l) Kaltreider et al (2001)demonstrated that

As blocked glucocorticoid receptor-dependent

induc-tion of gene expression when administered at 0.3 mM

(22.5 mg/L)

et al (2001) found elevated As concentrations (up to

3050 mg/l) in the groundwater, which is over the WHO

drinking water guideline (10 mg/l) and the levels are

comparable to those in Bangladesh and West Bengal,

India The area of the Red River Delta is much

smaller than that of the Bengal Delta, but there are 11

million people Although the health risk to the

population consuming As-contaminated groundwater

is of great concern, there is no information on As

contamination in residents around the Red River Delta,

Vietnam

In the present study, the contamination status of As

was evaluated in groundwater pumped from the tube

wells in the households at the suburban areas of Hanoi

Also, As level in human hair from the corresponding

household with the tube well was determined Recently,

Frisbie et al (2002) revealed that concentrations of

Mn, Pb, Ni, and Cr as well as As were above the

WHO drinking water guidelines in some groundwater

of Bangladesh, suggesting the need for research on

contamination of multiple elements in groundwater and

their mixture toxicity Hence, concentrations of other

trace elements in groundwater and human hair were also

determined to assess the health effect of multielements in

the present study

2 Materials and methods 2.1 Samples

Groundwater samples were randomly collected from

25 tube wells in Gia Lam District and Thanh Tri District

in the suburban area of Hanoi, Vietnam (Fig 1), during September (dry season) 2001 These sampling locations correspond to the As-contaminated area reported by Berg et al (2001) Pond water (nZ1) and rainwater stored for drinking (nZ2) in Thanh Tri District were also sampled Polypropylene bottles (100 ml), which were soaked in a nitric acid bath and then washed with Milli-Q water, were used for water sampling Collected groundwater, pond water, and rainwater samples were stored at
20C as immediately as possible

Human hair samples (nZ59; 24 families) were also collected from each house equipped with a tube well at Gia Lam District (nZ20; 8 families) and Thanh Tri District (nZ39; 13 families) We obtained the informed consent from all the donors in an ethical manner Hair samples were taken into the polyethylene bags and were kept at
20C until chemical analysis The details of the hair samples are shown inTable 1

2.2 Water analysis After transporting to the laboratory, the water samples were acidified with H2SO4 for As and HNO3 for other elements Milli-Q water acidified with H2SO4

or HNO3was used as control Arsenic concentration of water was measured by hydride generation atomic absorption spectrometry (AAS) using a Shimadzu HVG-1 hydride system coupled to a Shimadzu-AA680

Concentrations of V, Cr, Mn, Co, Cu, Zn, Ga, Rb, Sr,

Mo, Ag, Cd, In, Sn, Sb, Cs, Ba, Tl, Pb and Bi were determined by inductively coupled plasma mass spec-trometry (ICP-MS; Hewlett-Packard, HP-4500) (Anan

et al., 2001; Agusa et al., 2003) Yttrium was used as an internal standard for ICP-MS measurements Concen-tration of Hg was determined with a cold vapor AAS (CV-AAS) (Sanso, Model HG-3000) Iron was mea-sured by AAS Water pH was meamea-sured by a glass electrode pH meter (Asahi Techno Glass)

2.3 Hair analysis Human hair samples were washed by sonication with 0.3% polyoxyethylene lauryl ether (Okamoto et al.,

1985), and subsequently dried for 12 h at 80 C For As analysis, dried hair sample was accurately weighted directly into a Kjeldahl flask and 8 ml of purified HNO3 was added After pre-digestion at room temperature overnight, the sample was treated with 16 ml of acid mixture (HNO3:HClO4:H2SO4Z1:2:1) and digested by

heating to over 300C until the perchloric acid was

removed (Kubota et al., 2001) Arsenic concentration was

measured by HG-AAS For analysis of other trace

elements, about 0.1 g of hair sample was digested in

1.5 ml of concentrated HNO3in a Teflon PTFE tube in

a microwave oven (Agusa et al., 2003; Ikemoto et al.,

2004) Concentrations of 20 elements (V, Cr, Mn, Co, Cu,

Zn, Ga, Rb, Sr, Mo, Ag, Cd, In, Sn, Sb, Cs, Ba, Tl, Pb and

Bi) and Hg were determined by ICP-MS and CV-AAS,

respectively The accuracy of the method was assessed

using a certified reference material NIES No.5 Human

Hair (Okamoto et al., 1985) provided by the National

Institute for Environmental Studies (NIES), Japan

Recoveries of the elements ranged from 79.2 to 108%

2.4 Statistical analysis

One half of the value of the respective limit of

detection was substituted for those values below the

limit of detection and used in statistical analysis All

data were tested for goodness of fit to a normal

distribution with KolmogoroveSmirnov’s one sample test Because some variables were not normally distrib-uted, nonparametric tests were used to compare different groups and correlation for analysis The ManneWhitney U-test was employed to detect gender and regional (Gia Lam and Thanh Tri) differences in trace element concentrations KruskaleWallis test was also used for comparison of more than two groups Spearman’s rank correlation coefficient was used to measure the strength of the associations between As concentration and other variables A p value of less than 0.05 was considered to indicate statistical significance All statistical analyses were performed with StatView (version 5.0, SASÒ Institute, Cary, NC, USA) and SYSTAT (version 9, SPSS, Chicago, IL, USA)

3 Results and discussion 3.1 Contamination status of As in groundwater Concentrations of As in the groundwater ranged from !0.1 to 330 mg/l (Table 2) Median As concen-tration in groundwater in Gia Lam (5.0 mg/l) was comparable to that in Thanh Tri (1.5 mg/l) ( pO0.05; Table 2) About 40% of these samples contained As concentrations exceeding WHO drinking water guide-line of 10 mg/l (WHO, 1996) (Fig 2), which confirmed the findings of Berg et al (2001) in which an elevated level of As was observed in the groundwater of the Red River Delta In contrast, As concentrations were low in pond water and rainwater (Table 2andFig 2)

To understand the magnitude of As contamination, concentrations of As in groundwater in the present study were compared with those in other As-contaminated

Table 1

Details of human hair samples from suburban areas of Hanoi

Gia Lam

Thanh Tri

a Mean and range.

Gia Lam

Thanh Tri Vietnam

China

Laos

Thailand

Cambodia

Fig 1 Map showing sampling location Groundwater, rainwater, pond water, and human hair were collected from Gia Lam District and Thanh Tri District in the suburban areas of Hanoi during September 2001.

Table 2

Concentrations of arsenic and other trace elements (mg/l) in groundwater, rainwater, and pond water in suburban areas of Hanoi

Gia Lam

Thanh Tri

Gia Lam

Thanh Tri

areas (Fig 3a) Median values in groundwater in Gia

Lam and Thanh Tri were lower than those in other

As-contaminated areas However, the maximum value

(330 mg/l) was comparable to the levels in other

concentrations of As were observed in groundwater

of Gia Lam (2e3050 mg/l) and Thanh Tri (9e3010 mg/l)

(Berg et al., 2001) In the present study, however,

a lower concentration (!0.1e330 mg/l) of As was

observed in groundwater though these samples were

collected from the same area (Fig 3a) Because the As

level in groundwater varied considerably between tube

wells (2e3050 mg/l) (Berg et al., 2001), the difference in

As concentration between the present and previous

studies might be related to the sampling points even

within the same area Also, annual and seasonal

variation of the As level might be partly responsible

for the inconsistency between the studies

3.2 Contamination status of other trace elements

in groundwater

Various trace elements in the groundwater, pond

water and rainwater were also measured in the present

study (Table 2) Concentrations of Fe and Mn in

groundwater were higher than those of other elements,

followed by alkaline earth metals such as Sr and Ba On

the other hand, concentrations of Ag, In, Sn, Sb, Cs,

Hg, Tl and Bi were very low Concentrations in pond

water and rainwater were lower than those in

ground-water for most of the elements

Concentrations over the WHO drinking water guideline were also found for Mn (500 mg/l) and Ba (700 mg/l) (WHO, 1996) in some groundwater samples from Gia Lam and Thanh Tri (Fig 4a,b) Particularly, the median value of Mn concentration in groundwater

in both Gia Lam and Thanh Tri was higher than

1000 mg/l and about 76% of the samples exceeded the WHO guideline of 500 mg/l (Fig 4a) Our study also indicates that Ba concentrations in three samples of groundwater were higher than WHO drinking water guidelines (700 mg/l) (Fig 4b) These findings indicate that people in the Red River Delta may be exposed not only to As but also to Mn and Ba from groundwater, and the possible adverse effects are of concern as described later

3.3 Influence of various factors on As concentration

in groundwater

A significant positive correlation was found between

( p!0.001; Fig 5) Hence, the As level in groundwater may be the result of reductive dissolution of Fe oxyhydroxides adsorbed with As in the Red River alluvial tract as suggested by Berg et al (2001) Significant positive correlations were also found between

As, and Mo, Ga and Ba in groundwater ( p!0.01), while the As concentration was negatively correlated with Pb and V ( p!0.01) The reasons for these relationships are not known, but these might be related

to geological and geochemical properties in the sub-surface of Red River alluvial tract Concentrations of

100 50

As concentration (µg/l) WHO guideline (10 µg/l)

Gia Lam district

Thanh Tri district

Pond water Rainwater Groundwater Groundwater

F12 F09 F07 F14 F13 F11 F10 F09 F08 F07 F05 F04 F03 F02 F31 F30 F29 F28 F26 F25 F24 F23 F22

ND

ND

ND ND

ND ND

ND

ND ND

Fig 2 Arsenic concentrations in groundwater, rainwater and pond water in the suburban areas of Hanoi Sample number of groundwater (F01-14, F21-31) indicates well in each home.

As and Mn were not significantly correlated in

groundwater ( p>0.05), suggesting that the

contamina-tion source of As and Mn was different in the wells

Depth of well (12e45 m) and pH in groundwater

(7.6e9.0) were not related to the As concentration in

the groundwater (data not shown)

3.4 Contamination status of As in human hair

Arsenic concentrations in hair of residents from Gia

Lam and Thanh Tri ranged from 0.088 to 2.77 mg/g dry

wt (Table 3) Arsenic levels in hair of residents at both

sites were comparable, which was consistent with the

results of groundwater (Table 2) Variation in As levels

in hair resembled the geographical pattern of As level in

groundwater, showing higher concentrations of As both

in hair and water of households F05, F07, F08 and F11

(Figs 2 and 6a) Also, a significant positive correlation between As concentrations in groundwater and human hair was observed (rZ0.57, p!0.001;Fig 7a) A similar doseeresponse relationship between As concentrations

in groundwater and human hair was reported in southwest Finland (Kurttio et al., 1998) This result suggests that groundwater is probably the main source

of As exposure for these people

Arsenic levels in human hair in the present study were lower than those from other contaminated areas such as Bangladesh and West Bengal, India (Das et al., 1995; Chowdhury et al., 2000; Karim, 2000), but higher than those of people in non-contaminated areas (Arnold

et al., 1990) (Fig 3b) Also, As levels in hair of some individuals exceeded the level that may be related to skin pathology (Arnold et al., 1990) (Fig 6a), suggesting the potential health effects for populations in the area investigated

Gia Lam, Vietnam (This study) Thanh Tri, Vietnam (This study)

Chile (Das et al., 1995) West-Bengal-1 (Das et al., 1995) Bangladesh-2 (Chowdhury et al., 2000) West-Bengal-2 (Chowdhury et al., 2000) Bangladesh-1 (Karim et al., 2000) Finland-1 (Kurttio et al., 1998) Kratie, Cambodia (Agusa et al., 2002) Utah, USA-2 (Das et al., 1995) Kandal, Cambodia (Agusa et al., 2002) Alaska, USA (Das et al., 1995) Normal people (Arnold et al., 1990)

As concentration (µg/g dry wt.)

(b)

Gia Lam, Vietnam (This study) Thanh Tri, Vietnam (This study) Bangladesh-1 (Karim, 2000) West-Bengal, India-1 (Das et al., 1995)

Vietnam (Berg et al., 2001) Thailand (Foy et al., 1992) Inner Mongol, China (Yoshida and Yamauchi, 2000)

Taiwan (Wu et al., 1989) Bangladesh-2 (Chowdhury et al., 2000) West-Bengal, India-2 (Chowdhury et al., 2000)

Finland-1 (Kurttio et al., 1998) Kratie, Cambodia (Agusa et al., 2002) Kandal, Cambodia (Agusa et al., 2002) Mexico (Cebrian et al., 1983) Utah, USA-1 (Lewis et al., 1999) Finland-2 (Kurttio et al., 1999)

10 -1 10 0 10 1 10 2 10 3 10 4 10 5

As concentration (µg/l)

(a)

Fig 3 Arsenic concentrations in groundwater (a) and hair (b) in arsenic-contaminated areas Bar and circle indicate range and median (or arithmetic mean), respectively.

3.5 Contamination status of other trace elements

in human hair

Concentrations of trace elements other than As were

also determined in human hair from Gia Lam and Thanh

Tri (Table 3) To understand the accumulation status of

trace elements in this population, trace element levels of

hair were compared with those in other studies (

Chittle-borough, 1980; Takeuchi et al., 1982) Concentrations of

‘‘Rb, Cd and Hg’’ in hair of the residents from Gia Lam

and Thanh Tri (Rb: !0.001e0.124 mg/g dry wt., Cd:

!0.001e0.965 mg/g dry wt., Hg: 0.09e0.98 mg/g dry wt.;

Table 3) were lower than those reported by

Chittle-borough (1980) (Rb: 0.2e0.23 mg/g dry wt.; Cd: 0.34e

1.6 mg/g dry wt.; Hg: 1.08e1.55 mg/g dry wt.) and

Takeuchi et al (1982) (Cd: 1.18 mg/g dry wt.; Hg:

3.9 mg/g dry wt.) In contrast, concentrations of V (median, 0.119 mg/g dry wt.) and Mn (median, 16.7 mg/

g dry wt.) in residents of Gia Lam and Thanh Tri (Table

3) were high, compared to the median values of V (0.029 mg/g dry wt.) and Mn (0.42 mg/g dry wt.) for Japanese (Takeuchi et al., 1982) Also, the hair concentrations of Ga and Ba in residents of Gia Lam and Thanh Tri (Ga: 0.019e3.02 mg/g dry wt., Ba: 0.3e34 mg/

g dry wt.; Table 3) were higher than the levels in the population in non-contaminated areas (Ga: 0.02e 0.14 mg/g dry wt., Ba: 0.6e5.6 mg/g dry wt.; Chittle-borough, 1980) The high hair concentrations of Mn and

Ba in the residents were consistent with the results obtained for the drinking water (Fig 4) Also, concen-trations of Mn ( p!0.001), Co ( p!0.001), Ga ( p!0.01),

Ba ( p!0.05), Mo ( p!0.05) and Sr ( p!0.05) in human

0

1000

WHO guideline (500 µg/l)

Mn concentration (µg/l)

WHO guideline (700 µg/l)

Ba concentration (µg/l)

F12 F09 F07

F12 F09

F14 F12 F10 F08 F06 F04 F02

F14 F12 F10 F08 F06 F04 F02

F31 F29 F27 F25 F23 F21

F31 F29 F27 F25 F23 F21

Gia Lam district

Thanh Tri district

Pond water Rainwater Groundwater Groundwater

Gia Lam district

Thanh Tri district

Pond water Rainwater Groundwater

Groundwater

(b) (a)

Fig 4 Concentrations of manganese (a) and barium (b) in groundwater, rainwater and pond water in the suburban areas of Hanoi Sample number

of groundwater (F01-14, F21-31) indicates well in each home.

hair were positively correlated with those in groundwa-ter Thus, consumption of the groundwater may be the main source to the residents for these elements

Although Mn is known as an essential element, teratogenicity and neurotoxic effects are caused at high doses (Barceloux, 1999) Abnormal neurological scores were observed in older persons (nZ77) who had consumed Mn-contaminated drinking water (range 1800e2300 mg/l) in northwest Peloponnesos, Greece; their mean hair Mn concentration was 10.99 mg/g dry

wt (Kondakis et al., 1989) Recently,Woolf et al (2002) also reported striking difficulties in both visual and verbal memory, which are known toxic effects of Mn,

in a child who received Mn loading for 5 years from drinking water The Mn concentrations of water and hair of the child in the study ofWoolf et al (2002)were

1210 mg/l and 3.09 mg/g dry wt., respectively The mean

Mn values of both Gia Lam (1520 mg/l for groundwater and 15.5 mg/g dry wt for hair) and Thanh Tri (1260 mg/l for groundwater and 38.9 mg/g dry wt for hair) in the present study were comparable to or higher than these

Mn levels in drinking water and hair associated with chronic Mn poisoning (Tables 2 and 3) Although neurological tests were not conducted in the present study, possible adverse effects of Mn through consump-tion of Mn-contaminated groundwater is of great

10 -1

10 0

10 1

10 2

10 3

Fe concentration (µg/l)

p <0.001

Gia Lam Thanh Tri

Fig 5 Relationship between arsenic and iron concentrations in

groundwater in the suburban areas of Hanoi Samples with arsenic

concentrations below the detection limit are plotted as the value of

detection limit.

Table 3

Concentrations of arsenic and other trace elements (mg/g dry wt.) in hair of residents in suburban areas of Hanoi

a N, number of samples with detectable concentration.

concern for the residents in Gia Lam and Thanh Tri.

Barium exerts toxic effects associated with hypokalemia

and electrocardiographic changes (Anonymous, 1988),

but there is very little information on the relationship

between Ba concentration in hair and its health effects

3.6 Factors influencing As, Mn and

Ba concentrations in human hair

A significant influence of age on As accumulation was

not observed in hair of residents of Gia Lam and Thanh

Tri, which agreed with the results of Paschal et al

(1989) Also, sexual difference in As level was

not significant in the present study In addition,

a conspicuous trend in hair As level within the family

(father, mother, son, and daughter) was not found

(Fig 6).Chung et al (2002)also reported that total As

levels in urine samples did not differ between father, mother, sons, and daughters Similar to the results of As

in hair, Mn concentration in hair was not related to age and sex in this population In marked contrast, a weak but significant positive correlation was found between hair Ba concentration and age of the residents ( p!0.05) and Ba concentration in hair of female was significantly higher than those of male ( p!0.01), leading to the highest level of Ba in the mother among family members ( p!0.01) (Fig 6b)

3.7 Risk assessment of cumulative exposure to As Although unsafe levels of As, Mn and Ba were observed in groundwater as described above, available data are insufficient to discuss the relationship between the chronic exposure and toxicity of Mn and Ba Thus,

10 5

Father Mother Son Daughter Unknown

Father Mother Son Daughter Unknown

Ba concentration (µg/g dry wt.)

(b)

F21 F24 F25 F27 F28 F29 F30 F31 F01 F02 F03 F04 F05 F06 F07 F08 F09 F10 F11 F12 F13

F21 F24 F25 F27 F28 F29 F30 F31 F01 F02 F03 F04 F05 F06 F07 F08 F09 F10 F11 F12 F13

As concentration (µg/g dry wt.)

Gia Lam district

Thanh Tri district

Gia Lam district

Thanh Tri district

Skin pathology (1 µ g /g dry wt.) (a)

Fig 6 Arsenic (a) and barium (b) concentrations in hair of residents in the suburban areas of Hanoi Sample number of family (F01-13, F21, F24, F25, F27-31) indicates the well used in the household, as shown in the legend of Fig 4 Unknown indicates that information on family members was not available.

we will particularly focus on the risk assessment of

chronic As exposure hereafter

Skin pathology is an initial symptom of chronic As

exposure (Abernathy et al., 1999; Tsunetoshi, 2000), but

manifestations of chronic As poisoning and arsenicosis

have not been observed in the residents in spite of the

high level of As in groundwater in the suburb of Hanoi

as observed in the present study andBerg et al (2001)

The absence of As poisoning and arsenicosis in the area

might be due to the fact that the private tube wells in the

area were installed only 1e10 years ago and that chronic

As poisoning usually occurs only after 5e10 years of

exposure through consumption of As-contaminated

water, as suggested byBerg et al (2001) To assess the

risk of chronic As exposure from groundwater in this area, cumulative As exposure was estimated from the As level in groundwater, year of tube-well usage, annual ingestion rate of groundwater, and daily water con-sumption, using the following equation:

½Cumulative As intake ðmgÞZ½As level in groundwater ðmg=lÞ!½Age of well ðyearÞ!½Ingestion rate of groundwaterð182:5 days=yearÞ!½Water consumption ð2l=dayÞ:

A value of annual ingestion rate of groundwater was set at 182.5 days/year (6 months) in the present study,

Cumulative As intake (mg)

p <0.001

(b)

Skin pathology (1 µg/g dry wt.)

As concentration in ground water (µg/l)

p <0.001

Gia Lam Thanh Tri

Gia Lam Thanh Tri

(a)

Skin pathology (1 µg/g dry wt.)

10 -1

10 0

10 1

10 -2

10 -1

10 0

10 1

10 -2

10 -2 10 -1 10 0 10 1 10 2 10 3

Fig 7 Relationships between arsenic concentrations in hair and groundwater (a) and between arsenic concentrations in hair and cumulative arsenic intake (b) in residents in the suburban areas of Hanoi Samples with arsenic concentrations below the detection limit are plotted as the value of detection limit.