This article was downloaded by: [University of Newcastle (Australia)]

On: 27 February 2014, At: 20:51

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Archives of Environmental Health: An International

Journal
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/vzeh20

Mortality for Certain Diseases in Areas with High Levels

of Arsenic in Drinking Water
a a a
Shih-Meng Tsai , Tsu-Nai Wang & Ying-Chin Ko
a
Institute of Medicine, Kaohsiung Medical College , Kaohsiung, Taiwan
Published online: 05 Apr 2010.

To cite this article: Shih-Meng Tsai , Tsu-Nai Wang & Ying-Chin Ko (1999) Mortality for Certain Diseases in Areas with
High Levels of Arsenic in Drinking Water, Archives of Environmental Health: An International Journal, 54:3, 186-193, DOI:
10.1080/00039899909602258

To link to this article: http://dx.doi.org/10.1080/00039899909602258

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Mortality for Certain Diseases in Areas with High Levels
of Arsenic in Drinking Water

SHIH-MENC TSAl
Downloaded by [University of Newcastle (Australia)] at 20:51 27 February 2014

TSU-NAI W A N C
YINC-CHIN K O
Institute of Medicine
Kaohsiung Medical College
Kaohsiung, Taiwan

ABSTRACT. Blackfoot disease was prevalent in a limited area on the southwest coast of Tai-
wan, where artesian well water containing arsenic (median = 0.78 ppm arsenic) had been
used for many years. Previous studies of arsenic exposure in the blackfoot disease endemic
area have been focused on malignant tumors. We, therefore, conducted this study to ana-
lyze mortality of all death causes in blackfoot disease endemic areas and to determine other
neglected cancers or noncancer diseases related to artesian well water containing high lev-
els of arsenic. We calculated standardized mortality ratios for cancer and noncancer dis-
eases, by sex, during the period from 1971 to 1994 and compared them to the local refer-
ence group (i.e, Chiayi-Tainan County) and the national reference group (i.e., Taiwan
population). The results revealed marked standardized mortality ratio differences for the 2
reference groups. Greater mortality was found for males and females with bladder, kidney,
skin, lung, nasal-cavity, bone, liver, larynx, colon, and stomach cancers, as well as lymphoma
than in the local reference population. With respect to noncancer diseases, we found greater
mortality for males and females who had vascular disease, ischemic heart disease, diabetes
mellitus, and bronchitis than in the local reference group. Mortalities for other diseases-
including rectal cancer, cerebrovascular disease, and other diseases-were higher among
cases than the local reference group. Our results indicated that the hazardous effect of
arsenic is systemic. Diseases related to arsenic exposure included those reported previously
by other investigators, as well as diseases reported in the present study.

ARSENIC is an ubiquitous element present in various inhalation or Inhaled or ingested arsenic

compounds throughout the earth’s crust. In 1556, Agri-
cola reported that some health effects might be related
to arsenic exposure. This prescient conclusion was
made before arsenic compounds were used widely dur-
ing the 18th and 19th centuries.’ A carcinogenic effect
related to inorganic arsenic was proposed in 1879-a
time during which high rates of lung cancer in German
miners might have been caused by inhaled arsenic.2 In
fact, arsenic is the first chemical for which there was
evidence of a carcinogenic effect. A few epidemiologi-
cal studies revealed that skin, lung, liver, and bladder
cancers were associated with arsenic exposure via
causes cancer in humans but not in other species.’T2
Despite these results, a recent publication from the
International Agency for Research on Cancer (IARC)
listed arsenic as a “group 1” carcinogen to lung and
skin.8 In addition to the relationship between arsenic
exposure and internal cancers, recent studies-includ-
ing epidemiological or animal studies-contained
reports that many other health hazards or diseases were
related to arsenic exposure. These diseases included
sinonasal ~ a n c e r ,diabetes
~ mellitus,1° ischemic heart
’
disease,’ childhood leukemia,I2 hyperten~ion,’~ and
vascular disease.14

186 Archives of Environmental Health

 In addition to epidemiological studies, many labora-
tory studies have supported the concept that the health
effects of arsenic could be systemic. For example,
arsenic could act as a clastogen, thus increasing micro-
nuclei in exfoliated bladder c ~ I I ’ ~or, ’ increasing
~ the
frequency of sister-chromatid exchanges (SCEs) in lym-
phocyte~,’~,’~ and it may cause increased lipid peroxi-
dati~n’~t~~-w hich reportedly is related to many chron-
ic diseases. To determine certain diseases that are
associated with chronic arsenic exposure, we analyzed
the mortality of all diseases, including cancers and non-
cancers, for a blackfoot disease endemic community on
the southwest coast of Taiwan. In this community, arte-
sian well water-characterized by high concentrations
of arsenic, had been used since the first decade of this
century.21Although higher mortality of some cancers
and other diseases were reported for this community in
’
Downloaded by [University of Newcastle (Australia)] at 20:51 27 February 2014
previous ~tudies,~,’ no general health-hazard assess-
ment had been done in the community. The blackfoot
disease endemic community included the townships of
Peimen, Hsuechia, Putai, and Ichu-all of which are
located between Chiayi and Tainan counties. We chose
the populations of Chiayi and Tainan counties as our
local reference group when we calculated the standard-
ized mortality ratios (SMRs) for various health hazards,
by sex, in the blackfoot disease endemic area during the
period from 1971 to 1993. The entireTaiwan population
served as the national reference group.
Materials and Method
Study area. The areas included in this study were
limited to the four townships of Peimen and Hsuechia
in Tainan County and Putai and lchu in Chiayi County.
\
Local reference
Study
Fig. 1. Locations of study areas, local reference group, and national reference group.
May/June1999 [Vol. 54 (No. 3)]
According to the results of previous studies, blackfoot
disease was prevalent in all these areas.22These four
neighboring townships are located on the southwest
coast ofTaiwan (Fig. 1). The soil and water from shallow
wells in these areas have a high salt content. Some
inhabitants, therefore, have used water from artesian
wells since the 1900~,~’particularly those who have
lived in ,villages along the coast. Unfortunately, the
unique geological characteristics cause the water from
these wells to be very poor in quality and quantity.23
The arsenic content of the artesian wells ranged from
0.25 to 1.14 ppm (median = 0.78 ~ p m ~The ~ )soil
. in
these areas has a high arsenic content that ranges from
-
5.3 to 1 1.2 mg/kg (median 7.2 mg/kg).25During 1956
and thereafter, a tap-water system was gradually
installed in these areas. Nevertheless, some residents
continued to use artesian well water for drinking, wash-
ing, and agriculture. However, during the mid-19 7 0 ~ , ~ ’
artesian well water was no longer used. Agricultural
products and fish reportedly contain high concen-
trations of arsenic.25In 1959, Blackwel126estimated that
the daily arsenic ingested by local residents was as high
as 1 mg.
Study population. It is mandatory that all births,
deaths, marriages, divorces, and migrations into
households be registered at the registration office. These
statistics are verified annually via door-to-door surveys;
therefore, the population statistics in Taiwan are very
accurate and complete. According to the Taiwan-Fukien
Demographic Fact Book printed by the Ministry of
Interior27most of the residents in the blackfoot disease
endemic areas engaged in farming, fishery activities,
and salt production. Their education levels and socio-
economic status were below average, compared with
National reference
/?
187
 the general population of Taiwan. In addition, a survey
conducted in 1960 revealed that residents consumed
insufficient amounts of fresh vegetables and animal
protein.28
Mortality analyses. Physicians are required to submit
standardized certificates for each death to the Depart-
ment of Health; therefore, the vital statistics published
by the National Health Department of Taiwan are very
complete. In published official vital statistics, the under-
lying cause of death was diagnosed by physicians (99O/,
of cases). All cancers were also confirmed by path-
ological examination. All death information in our study
was obtained from vital statistics and from the computer
database of deaths, which was established by the
National Health Department in 1971. To minimize
misdiagnosis and/or misclassification, we coded all the
underlying causes of death according to the 8th (i.e.,
Downloaded by [University of Newcastle (Australia)] at 20:51 27 February 2014
prior to 1980) or 9th revision of the International
Statistical Classification of Diseases, Injuries, and
Causes of Death. Given that the study areas were lo-
cated between Chiayi and Tainan counties, we chose the
population of Chiayi-Tainan as the local reference group
to help adjust for geographic variations in mortality and
for the potential influence of confounding factors. We
chose the entire Taiwan population as the national
reference group. Person-years, by sex and age group
(i.e., 0 y, 1-4 y, 5-9 y . . ., 80-85 y, and 85+ y) of the
referencegroups, were obtained from the Taiwan-Fukien
Demographic Fact Book published by the Republic of
China Ministry of Interior.28The sex- and age-specific
mortalitities for each disease for the reference groups
were calculated by dividing the number of deaths by the
number of person-years, by sex and age group, from
1971 to 1994. The SMRs were the ratio of the total
observed deaths for each disease from 1971 to 1994 in
the blackfoot disease endemic areas relative to the
expected deaths, which we calculated on the basis of
age- and sex-specific mortality rates for the reference
groups. We used the method suggested by R ~ t h m a to n~~
estimate the formula for the 95% confidence interval
(CI) for each SMR. This calculation involved the setting
of limits for observed deaths, and we assumed the
expected deaths to be constant. We assessed SMRs for
all diseases, except when the cause of death was clearly
unrelated to arsenic exposure (e.g., accident).
Results
The total deaths and person-years from 1971 to 1994
for the blackfoot disease endemic area, local reference
Table 1.-Numbers of Deaths and Person-Years of Study Area and Local and National Reference Groups
Study area Local reference
Sex No. deaths Person-years No. deaths
Male 11,193 1,508,623 113,576
Female 8 874 1,404,759 80,350
188
group, and national reference group were 1 1,193 and
1,508,623; 113,576 and 18,046,123; and 12,290,606
and 227,273,814; respectively, for males. The corre-
sponding values for females were and 8,874 and
1,404,759; 80,350 and 16,434,421; and 836,203 and
209,548,487 respectively (Table 1). The exact deaths
and expected numbers for the study area, by sex and
causes of death, are listed in Tables 2 and 3 for males
and females, respectively.
For males, the SMRs and 95% CISfor various causes
of death related to the local and national reference
groups are shown inTable 2. The SMRs between the two
different reference groups were similar, except for can-
cers of the esophagus, stomach, and rectum, and for
diabetes mellitus. High mortality (i.e., SMR > 3) for both
groups was found for cancers of the bladder, kidney,
skin, lung, and nasal cavity, and for vascular disease.
The disease with high-but marginal (i.e., lower limit of
95% CI is close to 1)-mortality for both groups includ-
ed leukemia, cerebrovascular disease, liver cirrhosis,
and nephropathy.
For females, the SMRs and 95% CISfor various caus-
es of death are shown in Table 3. Discrepancies in SMRs
between the two different reference groups was found
for cancers of the stomach, colon, rectum, cervix, and
brain, and for hypertension and nephropathy. The dis-
eases for which there were high mortalities for both
groups (i.e., SMR > 3) included cancers of the bladder,
kidney, skin, nasal cavity, larynx, and lung. In neither
group were there diseases for which there was a ”high-
but-marginal” mortality.
The SMRs of the total-death causes and all malignant
tumors were also significantly higher than those of local
and national reference groups for both sexes.
Discussion
In our study, we used mortality-rather than the inci-
dence of health hazards related to arsenic exposure-
for assessment. It should be noted that mortality is a
function of incidence and fatality rates. If a disease is
not completely lethal, it follows that the fatality rate
will not be 100%, and the association between expo-
sure and disease will be underestimated. Only one
underlying cause of death, rather than multiple causes
of death, was cited on the death certificates; therefore,
the association between exposure and disease may
have been distorted, especially for mortality from dia-
betes mellitus.
Given that four townships in the blackfoot disease
National reference
Person-years No. deaths Person-years
18,046,123 1,290,606 227,273,814
16,434,421 836,203 209,548,487
Archives of Environmental Health
 Table 2.4tandardized Mortality Ratios for Certain Diseases in Blackfoot-DiseaseEndemic Area with High Arsenic Exposure,
1971-1 994 (Males Only)

Local reference National reference

Cause ICD8,9 Obs. Exp. SMR 95% CI Exp. SMR 95%CI

Total causes 000-799 11 193 8 465.758 1.32 1.29, 1.35* 8 657.21 1.29 1.27, 1.32*
All malignant tumors 140-208 2 774 1 263.95 2.19 2.11, 2.28. I 430.87 1.94 1.87, 2.01*
Oral cavity and pharynx cancers
Oral 140-1 45 23 20.00 1.15 0.73, 1.73 24.79 0.93 0.59, 1.39
Pharyngeal, except NPC 146, 148, 149 24 17.75 1.35 0.87, 2.01 21.17 1.13 0.73, 1.69
Nasopharyngeal 147 60 50.69 1.18 0.90, 1.52 54.68 1.10 0.84, 1.41
Digestive system cancers
Esophagus 150 69 41.20 1.67 1.30, 2.12* 70.84 0.97 0.76, 1.23
Stomach 151 195 143.84 1.36 1.17, 1.46* 202.86 0.96 0.83, 1.1 1
Intestine 152 15 7.1 5 2.10 1.20, 3.54* 7.1 6 2.09 1.17, 3.46*
Colon 153 91 61.05 1.49 1.20, 1.83* 67.21 1.35 1.09, 1.66*
Rectum 154 46 31.96 1.44 1.05, 1.92* 37.12 1.24 0.91, 1.65
Downloaded by [University of Newcastle (Australia)] at 20:51 27 February 2014

Liver 155 63 1 345.27 1.83 1.69, 1.98* 345.23 1.83 1.69, 1.98*
Gallbladder 156 13 1 1.68 1.1 1 0.59, 1.90 13.93 0.93 0.50, 1.60
Pancreas 157 30 24.57 1.22 0.82, 1.74 31.90 0.94 0.63, 1.34
Respiratory system cancers
Nasal 160 40 13.30 3.00 2.14, 4.09* 10.93 3.66 2.61, 4.98*
Laryngeal 161 30 16.81 1.78 1.20, 2.55* 17.01 1.76 1.1 9, 2.52*
Lung 162 699 225.39 3.10 2.88, 3.34. 264.68 2.64 2.45, 2.84'
Bone cancer 170 41 16.64 2.46 1.77, 3.34* 17.60 2.33 1.67, 3.16*
Skin cancer 171-172 66 13.65 4.83 3.74, 6.15' 11.05 5.97 4.62, 7.60*
Breast cancer 175 - - - - - - -
Genitourinary system cancers
Cervical 179-1 80 -
Ovary 183 - - - - - - -
Prostate 185 48 19.07 2.52 1.86, 3.34* 24.55 1.96 1.44, 2.59*
Bladder 188 312 34.99 8.92 7.96, 9.96* 29.72 10.50 9.37, 11.73*
Kidney 189 94 13.91 6.76 5.46, 8.27* 13.83 6.80 5.49, 8.32*
Brain cancer 191 19 15.03 1.26 0.76, 1.97 16.71 1.14 0.68, 1.78
Lymphatic system cancers
Lymphoma 2 00-2 08 56 34.40 1.63 1.23, 2.1 1* 39.44 1.42 1.07, 1.84*
Leukernia 204-208 67 50.07 1.34 1.04, 1.70* 50.04 1.34 1.04, 1.70*
Diabetes mellitus 250 188 139.69 1.35 1.16, 1.55* 165.37 1.14 0.98, 1.31
Hypertension 401-405 158 21 6.83 0.73 0.62, 0.85* 221.84 0.71 0.61, 0.83*
Ischemic heart disease 41 0-41 4 445 254.68 1.75 1.59, 1.92* 297.61 1.50 1.36, 1.64*
Pulmonary heart disease 41 5-41 7 33 65.39 0.50 0.35, 0.71 * 49.55 0.67 0.46, 0.94'
Heart disease 420-429 534 503.37 1.06 0.97, 1.15 455.00 1.1 7 1.08, 1.28*
Cerebrovascular disease 430-438 1286 1 123.26 1.14 1.08, 1.21* 1 184.60 1.09 1.03, 1.1 5'
Vascular disease 440-448 107 30.09 3.56 2.91, 4.30* 34.68 3.09 2.53, 3.73*
Chronic obstructive pulmonary
disease
Bronchitis 490-491 157 106.38 1.48 1.25, 1.73* 84.08 1.87 1.59, 2.1 8*
Emphysema 492 31 38.09 0.81 0.55, 1.1 5 40.76 0.76 0.52, 1.08
Asthma 493 147 166.13 0.88 0.75, 1.04 130.35 1.1 3 0.95, 1.33
Liver cirrhosis 571 428 360.05 1.18 1.08, 1.31' 332.89 1.29 1.17, 1.41*
Nephritis, nephrotic syndrome, 580-589 206 176.01 1.1 7 1.02, 1.34' 167.71 1.23 1.07, 1.41*
nephrosis
Congenital abnormalities 740-759 86 75.68 1.14 0.91, 1.40 123.69 0.70 0.56, 0.86'

Notes: ICD = International Classification of Diseases, Obs = observed, Exp = expected, SMR =standardized mortality ratio, CI = confidence
interval, and NPC = nasopharyngeal cancer.
*Statistically significant (p < .05).

endemic area are located between Chiayi and Tainan
counties, we used the entire populations of Chiayi and
Tainan counties as the local reference group and the
whole population of Taiwan as the national reference
group. The age- and sex-specific rates were more stable
for the national reference group than the local reference
group. The lifestyles and other related factors, except for
arsenic exposure level of the local reference group,
were more similar to those of the study group; therefore,
use of this group could reduce confounding or bias. The
effect of arsenic is more outstanding if the SMRs are sig-
nificant for comparisons in which both groups are used.
Conversely, if confounding or bias exist, discrepancies
between the local and national reference groups might
result. In our study, there were no notable differences
between the results when we used the local or national

MaylJune1999 [vol. 54 (No. 3)] 189

 Table 3.-Standardized Mortality Ratios for Certain Diseases in Blackfoot-Disease Endemic Area with High Arsenic Exposure,
1971-1994 (Females Only)

Local reference National reference

Cause ICD8,9 Obs. Exp. SMR 95%CI EXP. SMR 95%CI

Total causes 000-799 8 875 6 329.72 1.40 1.37, 1.43' 6 670.85 1.33 1.30, 1.36'
All malignant tumors 140-208 2 029 843.90 2.40 2.30, 2.51* 991.33 2.05 1.96, 2.14*
Oral cavity and pharynx cancers
Oral 140-1 45 12 7.46 1.61 0.83, 2.81 8.25 1.45 0.75, 2.54
Pharyngeal, except NPC 146, 148, 149 10 4.24 2.36 1.13, 4.34* 4.52 2.21 1.06, 4.07'
Nasopharyngeal 147 29 31.13 1.37 0.92, 1.97 22.34 1.30 0.87, 1.86
Digestive system cancers
Esophagus 150 12 7.59 1.58 0.82, 2.76 14.54 0.83 0.43, 1.44
Stomach 151 111 79.46 1.40 1.15, 1.68' 109.43 1.01 0.83, 1.22
Intestine 152 8 5.81 1.38 0.59, 2.72 6.38 1.25 0.54, 2.47
Colon 153 83 58.47 1.42 1.1 3, 1.76' 69.00 1.20 0.96, 1.49
Rectum 154 33 2 1.98 1.50 1.03, 2.1 1* 31.90 1.03 0.71, 1.45
Downloaded by [University of Newcastle (Australia)] at 20:51 27 February 2014

Liver 155 224 119.28 1.88 1.64, 2.14* 119.59 1.87 1.64, 2.14*
Gal Ibladder 156 11 12.18 0.90 0.45, 1.62 14.39 0.76 0.38, 1.37
Pancreas 157 19 19.75 0.96 0.58, 1.50 22.71 0.84 0.50, 1.31
Respiratory system cancers
Nasal 160 29 5.82 4.98 3.33, 7.1 5' 5.69 5.10 3.41, 7.32'
Laryngeal 161 13 2.73 4.76 2.53, 8.15' 3.46 3.76 2.00, 6.43'
Lung 162 471 114.02 4.1 3 3.77, 4.52* 134.42 3.50 3.19, 3.84*
Bone cancer 170 34 15.11 2.25 1.56, 3.14' 15.57 2.18 1.51, 3.05'
Skin cancer 171-172 68 1 1.96 5.68 4.41, 7.21' 9.99 6.81 5.29, 8.63'
Breast cancer 175 47 46.48 1.01 0.74, 1.34 70.17 0.67 0.49, 0.89
Genitourinary system cancers
Cervical 179-1 80 122 96.09 1.27 1.05, 1.52" 117.50 1.04 0.86, 1.24
Ovary 183 15 13.78 1.09 0.61, 1.80 19.53 0.77 0.43, 1.27
Prostate 185 - - - - - - -
Bladder 188 295 20.96 14.07 12.51, 15.78 16.71 17.65 5.70, 19.79
Kidney 189 128 14.40 8.89 7.42, 10.57* 12.20 10.49 8.75, 12.47'
Brain cancer 191 21 11.99 1.75 1.08, 2.68' 14.1 8 1.48 0.92, 2.26
Lymphatic system cancers
Lymphoma 200-208 35 20.57 1.70 1.1 8, 2.37* 24.44 1.43 1 .OO, 1.99'
Leukemia 2 04-2 08 40 37.36 1.07 0.76, 1.46 37.96 1.05 0.75, 1.43
Diabetes mellitus 250 343 221.72 1.55 1.39, 1.72' 277.78 1.23 1.11, 1.37*
Hypertension 401-405 239 198.69 1.20 1.06, 1.37' 234.97 1.02 0.89, 1.1 5
Ischemic heart disease 41 0-41 4 283 197.02 1.44 1.27, 1.61* 229.24 1.23 1.09, 1.39*
Pulmonary heart disease 41 5-41 7 27 51.1 8 0.53 0.35, 0.77* 39.00 0.69 0.46, 1.01
Heart disease 420-429 493 51 1.25 0.96 0.88, 1.05 473.21 1.04 0.95, 1.14
Cerebrovascular disease 43W38 1 352 1 089.41 1.24 1.18, 1.31' 1 153.98 1.17 1.11, 1.24*
Vascular disease 440448 68 29.51 2.30 1.78, 2.93* 3.39 2.04 1.58, 2.58*
Chronic obstructive pulmonary
disease
Bronchitis 490491 148 96.55 1.53 1.30, 1.80' 76.05 1.95 1.65, 2.29'
Emphysema 492 16 13.96 1.15 0.65, 1.86 17.71 0.90 0.52, 1.47
Asthma 493 103 123.14 0.84 0.68, 1.01 94.98 1.08 0.89, 1.32
Liver cirrhosis 571 164 157.71 1.04 0.89, 1.21 142.21 1.15 0.98, 1.34
Nephritis, nephrotic syndrome, 580-589 196 168.39 1.16 1.01, 1.39* 182.89 1.07 0.93, 1.23
nephrosis
Congenital abnormalities 740-759 70 59.96 1.16 0.91, 1.48 105.84 0.66 0.52, 0.84*

Notes: ICD = International Classificationof Diseases, Obs = observed, Exp = expected, SMR = standardized mortality ratio, CI = confidence
interval, and NPC = nasopharyngeal cancer.
*Statistically significant (p < .05).

reference groups, thus revealing the importance of
arsenic exposure.
Generally speaking, the outcomes of our study
accorded with the results of many previous studies
(Table 41, suggesting that the relationship between
arsenic exposure and internal cancer-including blad-
der, kidney, lung, liver, and skin cancers-was signifi-
cant for both sexes and for both reference groups. In
addition, we found that adverse health effects-includ-
ing bronchitis, liver cirrhosis, nephropathy, intestinal
cancer, rectal cancer, laryngeal cancer, and cerebrovas-
cular disease-might be related to chronic arsenic
exposure via drinking water, a conclusion unreported
heretofore.
With respect to digestive cancers, all mortality rates
for stomach, colon, rectal, and liver cancers were ele-

190 Archives of Environmental Health

 Table 4.-Studies about Cancer and Noncancer Disease Related to Arsenic Exposure

Health hazard Exposure route Author and reference no. Location

Stomach cancer Inhalation Bulbulyan et a1.6 Russia

Colon cancer Inhalation Enterline et aL3 United States
Ingestion Chen et aL4 Taiwan
Ingestion Tsuda et aL5 Japan
Liver cancer Ingestion Chen et al.4 Taiwan
Nasal cancer Inhalation Battista et aL9 Italy
Inhalation Chen et al.” Taiwan
Lung cancer Inhalation Enterline et aL3 United States
Ingestion Chen et aL4 Taiwan
Ingestion Tsuda et aL5 Japan
Inhalation Bulbulyan et aL6 Russia
Bone cancer Inhalation Enterline et aL3 United States
Skin cancer Ingestion Chen et al.4 Taiwan
Uterine cancer Ingestion Tsuda et aL5 Japan
Prostate cancer Ingestion Chen et al.” Taiwan
Downloaded by [University of Newcastle (Australia)] at 20:51 27 February 2014

Bladder cancer Ingestion Chen et aL4 Taiwan

Ingestion Tsuda et aL5 Japan
Ingestion Bates et al.’ United States
Ingestion Hopenhayn-Rich et al.36 Argentina
Kidney cancer Ingestion Chen et aL4 Taiwan
Leukemia Ingestion DurantI2 United States
Diabetes mellitus Ingestion Lai et al.1° Taiwan
Inhalation Rahman et Sweden
Ischemic heart disease Ingestion Chen et al.” Taiwan
Vascular disease Ingestion Tseng et al.37 Taiwan
Ingestion Engel et al.38 United States
Ingestion Wu et aLi4 Taiwan

Notes: The results of the present study indicate that the following diseases may be related to arsenic exposure:
intestinal cancer, rectum cancer, laryngeal cancer, lymphoma, bronchitis, nephropathy, and cerebrovascular disor-
ders.

vated for both sexes, compared with the local reference
group; this did not always hold true when we compared
mortality rates from digestive cancers to the national
reference group. The lifestyles of the local reference
group were very similar to those of the study group;
therefore, the SMRs for these cancers, based on the
comparison with the local reference group, were more
appropriate for us to judge the relationship between
cause and diseases than SMRs based on the national
reference group. Perhaps high mortality rates resulted
from the fact that these digestive organs are in direct
contact with arsenic via ingestion, and small amounts
of arsenic can be excreted in feces. This result was also
reported in Japanby Tsuda et aL5 In Taiwan, the mortal-
ity rate for liver cancer associated with hepatitis B viral
infection is high. However, liver cancer mortality in
these areas remains higher than in the local and nation-
al reference groups. The liver is the main organ for
detoxification of arsenic, and arsenic will accumulate
in this organ after exposure. It is possible that the high
liver cancer mortality is related to arsenic exposure.
With respect to cancers of the respiratory system, mor-
tality for cancers of the nasal cavities, larynx, and lung
were elevated significantly in both sexes and for both
reference groups. In many studies, investigators reported
that cancers of the upper-respiratory tract-including
sinonasal cancer and cancers of the larynx and phar-
ynx-were related to occupational arsenic exposure via
i n h a l a t i ~ n . ~All
, ~of
, ~these
~ cancers, except lung cancer,
were rare in Taiwan and could have been triggered by
several occupational carcinogens via inhalation. The res-
idents of the blackfoot disease endemic areas and the
local reference group engage in farming, fishery activi-
ties, and salt production. There are no industrial factories
in these areas. Exposure to arsenic via ingestion for the
study group must be responsible for the extraordinarily
high SMRs we observed for these cancers.
An important confounder for lung cancer is smoking
status. According to previous studies, smoking preva-
lence for males was slightly higher than in the general
population of Taiwan3’; however, the prevalence for
females did not differ. In our analyses, the SMRs for
chronic obstructive pulmonary disease (COPD) were
not significant, except for bronchitis in both sexes,
compared with the local and national reference groups.
Although we could attribute approximately 80% of
COPD mortality to cigarette smoking, the high S M R for
bronchitis could not be explained by smoking alone.
On the basis of this finding, we believe that the effect
of smoking is not significantly different among the
study area, local group, and national reference groups.
The high SMRs for smoking-related health effects, such
as lung cancer, cannot be explained only by smoking.
Perhaps lung cancer is caused by arsenic via inhala-

MaylJune 1999 Wol. 54 (No. 3)l 191

 tion, as has been supported by other ~tudies,~,~ as well
as by ingestion, as was suggested by the results of our
current study.
There are only a few studies in which the investiga-
tors have referenced a relationship between arsenic
exposure and bone cancer. For example, Enterline et
aL3 found a positive relationship between exposure to
arsenic in air at a copper smelter and bone cancer. Our
results also revealed higher SMRs for bone cancer and
lymphoma in both sexes than in both reference groups.
The mechanism is unclear, but it deserves further study.
Kadowski3’ reported that arsenic accumulated in bone
marrow in an animal Perhaps this finding
explains the higher SMRs for these diseases we ob-
served in the current study.
With respect to cancers of the uro-reproductive sys-
tem, the SMRs for bladder cancer, calculated on the
Downloaded by [University of Newcastle (Australia)] at 20:51 27 February 2014
basis of the local and national reference groups, were
highest in our study, especially in females. Bladder can-
cer occurs more frequently in males than females. The
results of our study revealed that the age-adjusted mor-
tality for females in these areas was equal to males and
suggested that arsenic exposure influenced bladder
cancer in these areas. In fact, there were greater dis-
crepancies in mortality rates for many health effects in
females between these areas and the local or national
reference groups than for males. Perhaps the fact that
males are exposed to more risk factors-in addition to
arsenic exposure-than females explains these results.
In addition to bladder cancer, kidney cancer mortality
was very high in the study area. Nephropathy-includ-
ing nephritis, nephrotic syndrome, and nephrosis-was
also a serious problem for the study group. Given that
arsenic is excreted mainly via urine, it would be an
important etiologic factor.
The SMRs were not significant in cancers of the
reproductive system, with the exception of prostate can-
cer for males and for both reference groups. A similar
finding was reported by Chen et al.,3 who noted that
Swedish glassworkers had a slightly elevated SMR for
prostate cancer, compared with a national reference
group. It would appear that arsenic exposure is not
related to cancers of the reproductive organs.
Recently, the relationship between many chronic
diseases and arsenic exposure has been investigated.
Such diseases have included diabetes rnellitus, hyper-
tension, ischemic heart disease, cerebrovascular dis-
ease, and vascular disease.10*’1,13,3’,33 The validation of
the analyses that involved mortality for chronic dis-
ease is less than that for cancer. As mentioned above,
the fatality rate for chronic diseases i s less than 100%.
If we used mortality, rather than incidence rate, to
assess associations between exposure and disease,
their relationship-if such a causal relationship, in
fact, exists-would be underestimated. Nevertheless,
our findings accord with the results of previous stud-
ies, with the exception of our findings regarding
hypertension of males.
Significantly high SMRs for ischemic heart disease,
cerebrovascular disease, and vascular disease indicated
that arsenic caused great damage to the vascular system
192
(i-e., blackfoot disease). Chronic arsenic exposure could
influence the peripheral circulation and nerve function
via va~odilation.~~
Chen et aI.l3 and Lai et al.1° used prevalence, rather
than mortality, to assess and confirm the relationship
between chronic arsenic exposure and hypertension
and diabetes mellitus, respectively. In our study, the
SMRs for diabetes mellitus were significantly-but
marginally-higher than for the local reference group,
and the SMRs for hypertension were significantly lower
for males than females. Perhaps these differences
resulted from misdiagnosis and/or misclassification of
cause of death.
In addition to arsenic, fluorescent hurnic substances
have been identified in the artesian well water in the
blackfoot disease endemic areas. Lu et al.35 demon-
strated that humic acid caused blackfoot disease in
mice. Vascular damage was also associated with hurnic
acid. Whether these effects resulted from arsenic alone
or in combination with humic acid was unclear.
Although humic acid is distributed widely, no correla-
tions have been observed between concentration of
humic acid and arsenic levels, blackfoot disease, or
cancers.l Therefore, we cannot attribute all of these
effects to humic acid alone.
In conclusion, the results of our study evidenced the
potential relationship between mortality from certain
diseases and chronic arsenic exposure. In addition to
diseases reportedly associated with arsenic, we suggest
that further studies be performed to investigate relation-
ships between arsenic and intestinal cancer, rectal can-
cer, laryngeal cancer, lymphoma, cerebrovascular dis-
ease, bronchitis, and nephropathy.
**********
Submitted for publication February 1, 1998; revised; accepted for
publication August 10, 1998.
Requests for reprints should be sent to Ying-Chin KO, Ph.D., M.D.,
Institute of Medicine, Kaohsiung Medical College, 100, Shih-Chuan
1st Road, Kaohsiung, Taiwan.
I
**********
References
1. Smith AH, Hopenhayn-Rich C, Bates MN, et al. Cancer risk from
arsenic in drinking water. Environ Health Perspect 1992; 97:
259-67.
2. Bates MN, Smith AH, Hopenhayn-Rich C. Arsenic ingestion and
internal cancers: a review. Am J Epidemiol 1992; 135:462-76.
3. Enterline PE, Day R, Marsh CM. Cancers related to exposure to
arsenic at copper smelters. Occup Environ Med 1995; 52:
28-32.
4. Chen CJ, ChuangYC, LinTM, et al. Malignant neoplasms among
residents of a blackfoot disease-endemic area in Taiwan: high-
arsenic artesian well waters and cancers. Cancer Res 1985;
45~5895-99.
5. Tsuda T, Babazono A, Yamamoto E, et al. Ingested arsenic and
internal cancer: a historical cohort study followed for 33 years.
Am J Epidemiol 1995 141:198-209.
6. Bulbulyan MA, Jurenkova NJ, Boffetta P, et al. Mortality in a
cohort of Russian fertilizer workers. Scand J Work Environ Health
1996; 22~27-33.
7. Bates MN, Smith AH, Cantor KP. Case-control study of bladder
Archives of Environmental Health
 cancer and arsenic in drinking water. Am J Epidemiol 1995;
141 :523-29.
8. Tomatis L, Aitio A, Wilbourn J, et al. Human carcinogens so far
identified. JapanJ Cancer Res 1989; 80:795 -807.
9. Battista G, Bartoli D, laia TE, et al. Art glassware and sinonasal
cancer: report of three cases. Am J Ind Med 1996; 30:31-35.
10. Lai MS, Hsueh YM, Chen CJ, et al. Ingested inorganic arsenic and
prevalence of diabetes mellitus. Am j Epidemiol 1994; 139:
484-92.
1 1 . Chen CJ, Chiou HY, Chiang MH, et al. Dose-response relation-
ship between ischemic heart disease mortality and long-term
arsenic exposure. Arterioscler Thromb Vasc Biol 1996; 16:
504-1 0.
12. Durant JL, Chen J, Hemond HF, et al. Elevated incidence of child-
hood leukemia in Woburn, Massachusetts: NlEHS superfund
basic research program searches for causes. Environ Health Per-
spect 1995; 103:93-98.
13. Chen CJ, Hsueh YM, Lai MS, et al. Increased prevalence of hyper-
tension and long-term arsenic exposure. Hypertension 1995; 25:
53-60.
14. Wu MM, Kuo TL, Hwang YH, et al. Dose-response relation
Downloaded by [University of Newcastle (Australia)] at 20:51 27 February 2014
between arsenic concentration in well water and mortality from
cancers and vascular diseases. Am J Epidemiol 1989; 130:
1123-32.
15. Warner ML, Moore LE, Smith MT, et al. Increased micronuclei in
exfoliated bladder cells of individuals who chronically ingest
arsenic-contaminated water in Nevada. Cancer Epidemiol Bio-
mark Prev 1994; 3:583-90.
16. Moore LE, Smith AH, Hopenhayn-Rich C, et al. Micronuclei in
exfoliated bladder cells among individuals chronically exposed
to arsenic in drinking water. Cancer Epidemiol Biomark Prevent
1997; 6:31-36.
17. Hantson P, Verellen-Dumoulin C, Libouton JM, et al. Sister chro-
matid exchanges in human peripheral blood lymphocytes after
ingestion of high doses of arsenicals. Int Arch Occup Environ
Health 1996; 68:342-44.
18. Lerda D. Sister-chromatid exchange (SCE) among individuals
chronically exposed to arsenic in drinking water. Mutat Res 1994;
312:lll-20.
19. Lee TC, Ho IC. Differential cytotoxic effects of arsenic on human
and animal cells. Environ Health Perspect 1994; 102:101-05.
20. Ramos 0, Carrizales L, Yanesz L, et al. Arsenic increased lipid
peroxidation in rat tissues by a mechanism independent of glu-
tathione levels. Environ Health Perspect 1995; 103(suppl 1):
85-88.
21. Tseng WP. Blackfoot disease in Taiwan: a 30-year follow-up
study. j Vasc Dis 1989; 7:547-58.
22. Wu HY, Chen KP, Tseng WP, et al. Epidemiologic studies on
blackfoot disease. 1 . Prevalence and incidence of the disease by
age, sex, year, occupation, and geographic distribution. Mem
MayIJune1999 [vol. 54 (No. 3)]
Coll Med NatlTaiwan Univ 1961; 7:33-50.
23. Tseng WP, Chu HM, How SW, et al. Prevalence of skin cancer in
an endemic area of chronic arsenicism in Taiwan. JNCl 1968;
40~453-63.
24. Chen KP, Wu HY, Wu TC. Epidemiologic studies on blackfoot dis-
ease in Taiwan. 3. Physicochemical characteristics of drinking
water in endemic blackfoot disease area. Mem Coll Med Natl
Taiwan Univ 1962; 8:115-19.
25. Lo MC. The arsenic content of farm products and fish in areas
where high arsenic well water was used for agriculture and pis-
ciculture. Report on Blackfoot Disease Research. Taiwan, ROC:
Taiwan Provincial Government, 1978; 6:28-46.
26. Blackwell RQ. Estimated total arsenic ingested by residents in the
endemic blackfoot area. J Formosan Med Assoc 1961; 60:
1 143-44.
27. Ministry of Interior, Republic of China. Demographic Facts,
1971-1994. Taipei, ROC: Ministry of Interior.
28. Chen CJ,k Wu MM, Lee SS, et al. Atherogenicity and carcino-
genicity of high-arsenic artesian well water: multiple risk factors
and related malignant neoplasm of blackfoot disease. Arte-
riosclerosis 1986; 8:452-60.
29. Rothman KJ. Stratified analysis. Modern Epidemiology. Boston,
Ma: Little, Brown; 1986.
30. Malker HSR, McLaughlin JK, Weiner JA, et al. Occupational risk
factors for nasopharyngeal cancer in Sweden. Br J Ind Med 1990;
47:213-14.
31. Chen CJ, Wang CJ. Ecological correlation between arsenic level
in well water and age-adjusted mortality from malignant neo-
plasms. Cancer Res 1990; 50:5470-74.
32. Kadowski K. Studies on the arsenic contents in organ tissues of
the normal Japanese.Osaka City Med J 1960; 9:2083-88.
33. Rathman M, Axelson 0. Diabetes mellitus and arsenic exposure:
a second look at case-control data from a Swedish copper
smelter. occup Environ Med 1995; 25:773-74.
34. Tseng CH, Chong CK, Chen CJ, et al. Abnormal peripheral
microcirculation in seemingly normal subjects living in black-
foot-disease-hyperendemic villages in Taiwan. Int J Cicrocirc
1995; 15:21-27.
35. Lu FJ. Blackfoot disease: arsenic or humic acid? Lancet 1990;
336:115-16.
36. Hopenhayn-Rich C, Biggs ML, Fuchs A, et al. Bladder cancer
mortality associated with arsenic drinking water in Argentina.
Epidemiology 1996; 7:117-24.
37. Tseng CH, Chong CK, Chen CJ, et al. Dose-response relationship
between peripheral vascular disease and ingested inorganic
arsenic among residents in blackfoot disease endemic villages in
Taiwan. Atherosclerosis 1996; 120:125-33.
38. Engel RR, Smith AH. Arsenic in drinking water and mortality from
vascular disease: an ecologic analysis in 30 counties in the Unit-
ed States. Arch Environ Health 1994; 49:418-27.
193