Provided for non-commercial research and educational use only.
Not for reproduction or distribution or commercial use.
This article was originally published by IWA Publishing. IWA Publishing recognizes
the retention of the right by the author(s) to photocopy or make single electronic
copies of the paper for their own personal use, including for their own classroom use,
or the personal use of colleagues, provided the copies are not offered for sale and
are not distributed in a systematic way outside of their employing institution.
Please note that you are not permitted to post the IWA Publishing PDF version of
your paper on your own website or your institution’s website or repository.
Please direct any queries regarding use or permissions to firstname.lastname@example.org
Health impacts of long-term exposure to disinfection
by-products in drinking water in Europe: HIWATE
Mark J. Nieuwenhuijsen, Rachel Smith, Spyros Golﬁnopoulos, Nicky Best,
James Bennett, Gabriella Aggazzotti, Elena Righi, Guglielmina Fantuzzi,
Luca Bucchini, Sylvaine Cordier, Cristina M. Villanueva,
Victor Moreno, Carlo La Vecchia, Cristina Bosetti,
Terttu Vartiainen, Radu Rautiu, Mireille Toledano, Nina Iszatt,
Regina Grazuleviciene and Manolis Kogevinas
Mark J. Nieuwenhuijsen (corresponding author)
Cristina M. Villanueva
Centre for Research in Environmental
Epidemiology (CREAL), Parc de Recerca
`dica de Barcelona-PRBB (ofﬁce 183.05),
C. Doctor Aiguader, 88, 08003 Barcelona,
Tel.: (+34) 93 316 0646
Fax: (++34) 93 316 05 75
Municipal Institute of Medical Research
(IMIM-Hospital del Mar) and CIBER
Epidemiologia y Salud Pu
Mark J. Nieuwenhuijsen
Imperial College London, UK
University of the Aegean, Greece
National Public Health Institute, Finland
University of Crete, Greece
´de Rennes, France
University of Modena and Reggio, Italy
ICON Ltd, UK
Vytautas Magnus University, Lithuania
Carlo La Vecchia
Istituto di Ricerche Farmacologiche ‘Mario Negri’,
Hylobates Consulting Srl, Italy
Catalan Cancer Institute (ICO), Spain
There appears to be very good epidemiological evidence for a relationship between chlorination
by-products, as measured by trihalomethanes (THMs), in drinking water and bladder cancer,
but the evidence for other cancers, including colorectal cancer appears to be inconclusive
and inconsistent. There appears to be some evidence for a relationship between chlorination
by-products, as measured by THMs, and small for gestational age (SGA)/intrauterine growth
retardation (IUGR) and preterm delivery, but evidence for other outcomes such as low birth
weight (LBW), stillbirth, congenital anomalies and semen quality appears to be inconclusive
The overall aim of the HIWATE study is to investigate potential human health risks (e.g. bladder
and colorectal cancer, premature births, SGA, semen quality, stillbirth, congenital anomalies)
associated with long-term exposure to low levels of disinfectants (such as chlorine) and DBPs
occurring in water for human consumption and use in the food industry. The study will comprise
risk– beneﬁt analyses including quantitative assessments of risk associated with microbial
contamination of drinking water versus chemical risk and will compare alternative treatment
options. The outcome will be improved risk assessment and better information for risk
management. The work is divided into different topics (exposure assessment, epidemiology, risk
assessment and management) and studies.
cancer, chlorination, disinfection by-products, epidemiology, reproductive health, risk
185 QIWA Publishing 2009 Journal of Water and Health
It has been more than 30 years since trihalomethanes
(THMs) were ﬁrst discovered in the Netherlands (Rooks
1974). Chlorination disinfection by-products (DBPs) are
formed when water is chlorinated and the organic matter
in the water reacts with chlorine to form these by-products.
The formation and occurrence depends on many factors,
including the chlorine dose, type of treatment, pH, tempera-
ture, residence time and bromine levels (Nieuwenhuijsen
et al. 2000a;IPCS 2000). Up to 600 DBPs have been
identiﬁed (Richardson 1998;Richardson et al. 2008).
Different mixtures of by-product may exist in different
locations depending on the various factors mentioned
above, making it more difﬁcult to assess any health effects
of DBPs, particularly in epidemiological studies. In Europe
there is relatively little known about the occurrence of DBPs
other than THMs and their levels (Palacios et al. 2000), with
some exceptions in a few places such as Poland (Dojlido et al.
1999), Finland (Nissinen et al. 2002), Spain (Villanueva et al.
2003), the UK (Malliarou et al. 2005), Greece (Golﬁnopoulos
& Nikolaou 2005) and Italy (Fantuzzi et al. 2007), while in
the US extensive surveys have been conducted to assess DBP
occurrence under different water treatment methods (e.g.
Weinberg et al. 2002;Krasner et al. 2006).
In the USA and Canada there has been considerable
progress in the assessment of health risks and policy
development in relation to DBPs, including a research
programme on the occurrence and health risks relation to
the by-products by the USEPA’s ofﬁce on Water and National
Health and Environmental Effects Research Laboratory
A considerable amount of work has been carried out
measuring a range of different DBPs, animal testing of a list
of high priority by-products and epidemiological studies.
However, results may not be extended to Europe because
mixtures of DBPs may be different as a result of different
determinants such as treatment, total organic content (TOC),
pH etc. In Europe there has been a much slower response to
the recent ﬁndings. Disinfection is used in many countries in
Europe and is therefore of European concern and requests a
European approach and solution. Relatively little research has
been carried out on DBPs in relation to adverse birth
outcomes and cancer in Europe. Where work has been
carried out, this has been carried out in isolation.
Safe drinking water has a high priority in Europe in
accordance with the Water Framework Directive and the
Directive on Quality of Water intended for Human
Consumption. Water treatment safety has become particu-
larly acute since the quality of water resources may be
declining because of water scarcity in some regions,
increasing the cost of drinking water production and the
likelihood of chemical interactions during the treatment
process. Water is an important part of the food chain.
Consumer health and well-being, quality, safety and con-
sumer concern, are highly important and should be
addressed where possible, particularly where environmental
health risks are concerned. Recently there has been
consumer concern about the quality of drinking water
from the tap and this may have led to an increase in the
consumption of expensive bottled water in developed
nations (Doria et al. 2005;Doria 2006), reducing the
money that can be spent on more beneﬁcial items.
Ingestion of water may not be the only concern since an
individual can also be exposed to volatile DBPs (e.g. THMs)
through inhalation and absorption, during activities such as
showering, bathing and swimming (Nieuwenhuijsen et al.
2000a). Recent modelling has suggested that this route may
lead to the highest levels in the blood (Whitaker et al. 2003).
Uptake through showering, bathing and swimming showed
considerable increased risk in a recent bladder cancer study
(Villanueva et al. 2007). For non-volatile DBPs, such as
haloacetic acids (HAAs), ingestion is thought to be the main
route of exposure.
In this paper we brieﬂy summarise the epidemiological
evidence and limitations regarding the two main areas of
health effects from exposure to DBPs, cancer and repro-
ductive effects. We then present the background and design
of a major research initiative in the EU (Health Impacts of
long-term exposure to disinfection by-products in drinking
WATEr, HIWATE project) that will provide an extensive
evaluation of exposure, hazard identiﬁcation, risk assess-
ment and risk beneﬁt analysis for these compounds in
186 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
EPIDEMIOLOGICAL STUDIES EXAMINING HEALTH
EFFECTS RELATED TO EXPOSURE TO
The health effects of DBPs in drinking water have been a
concern since DBPs were ﬁrst reported in the 1970s. Early
studies focused on cancer outcomes, while the more recent
studies have focused on reproductive outcomes (IPCS
2000). According to the recent review by IPCS (2000):
more studies have considered bladder cancer than any
other cancer. The authors of the most recently reported
results for bladder cancer risks caution against a simple
interpretation of the observed associations. The epide-
miological evidence for an increased relative risk for
bladder cancer is not consistent—different risks are
reported for smokers and non-smokers, for men and
women, and for low and high water consumption. Risk
may differ among various geographic areas because the
DBP mix may be different or because other water
contaminants are also present. More comprehensive
water quality data must be collected or simulated to
improve exposure assessments for epidemiological
The document also mentioned the difﬁculties in
exposure assessment for epidemiological studies of cancer
and DBPs, due to the long lag periods and the general lack
of detailed historical data.
A very important recent pooled analysis by Villanueva
et al. (2004), which provided quantitative information,
conﬁrmed this. For men there was an exposure response
relationship between DBP intake and bladder cancer, but
there was no relationship in women (Table 1). Furthermore,
the latest Spanish study suggested that not only is exposure
through ingestion an important risk factor but also exposure
through swimming, showering and bathing (Villanueva et al.
2007). Furthermore in this study the authors identiﬁed
genetically susceptible groups such as those with gluta-
thione S-transferase theta 1 (GSTT1) and glutathione
S-transferase zeta 1 (GSTZ1) polymorphisms (Cantor et al.
2006). Some studies have suggested an association between
DBPs and colorectal cancers, while others have not (Young
et al. 1981,1987;Wilkins & Comstock 1981;Doyle et al. 1997;
Koivusalo & Vartiainen 1997;Hildesheim et al. 1998;King
et al. 2000a;Bove et al. 2007). Studies on colorectal cancer
have relatively limited sample size and have used relatively
crude measures of exposure assessment focusing principally
on THMs levels in the water, without examining different
exposures or gene–environment interactions. There is little
evidence for an association between exposure to DBPs
and other cancers such as liver, kidney, brain, lung and
breast cancer, lymphomas or cancer of the pancreas, but
the number of studies is small (IPCS 2000) and very few of
these have involved populations in Europe. A recent
report suggested an association between THMs and skin
cancer, but further work needed to be conducted (Karagas
et al. 2008).
Reproductive health outcomes should be easier to study
from an exposure point of view, because of the shorter
relevant exposure period. Among others, birth weight,
prematurity, spontaneous abortion, congenital anomalies
and stillbirth have been the focus of these studies. Overall
there appears to be some evidence for a relationship
between chlorination by-products, as measured by THMs,
and small for gestational age (SGA)/intrauterine growth
retardation (IUGR) and preterm delivery, but evidence for
other outcomes such as low birth weight (LBW), stillbirth,
congenital anomalies and semen quality appears to
be inconclusive and inconsistent (Kramer et al. 1992;
Aschengrau et al. 1993; Bove et al. 1995,2002;Savitz et al.
1995,2006;Kanitz et al. 1996;Reif et al. 1996;Gallagher et al.
1998;Waller et al. 1998,2001;Dodds et al. 1999,2004;
Pooled analysis of bladder cancer and THM (after Villanueva et al. 2004)
THM exposure level (mg) Male ORs (95%CI) Female ORs (95%CI)
0–15 1.00 1.00
.15–50 1.22 (1.01–1.48) 0.92 (0.65 – 1.32)
.50–400 1.28 (1.08–1.51) 0.94 (0.70 – 1.27)
.400–1000 1.31 (1.09–1.58) 1.02 (0.74 – 1.41)
.1000 1.50 (1.22–1.85) 0.92 (0.65 – 1.30)
OR (95%CI) ¼odds ratio (95% conﬁdence interval).
187 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
Klotz & Pyrch 1999;Magnus et al. 1999;King et al. 2000b,
2005;Kallen & Robert 2000;Yang et al. 2000,2007;
Nieuwenhuijsen et al. 2000b,2008;IPCS 2000;Gevecker
Graves et al. 2001;Jaakkola et al. 2001;Dodds & King 2001;
Cedergren et al. 2002;Hwang et al. 2002;Fenster et al. 2003;
Hwang & Jaakkola 2003;Shaw et al. 2003;Wright et al.
2003,2004;Aggazzotti et al. 2004;Yang 2004;Hinckley
et al. 2005;Porter et al. 2005;Toledano et al. 2005;
Lewis et al. 2006,2007;Tardiff et al. 2006;Luben et al. 2007).
Infante-Rivard (2004) found that the association
between THMs and intrauterine growth retardation was
modiﬁed by a metabolic polymorphism, with newborns
without the CYP2E1 (G1259C) variant at high risk, but
found no indication that MTHFR C677T modiﬁed the effect
of exposure to chloroform and risk to foetal growth in
humans. Neither did Shaw et al. (2003) for neural tube
defects (NTDs). This sheds some light on the possible
mechanism of action. However, the mechanisms through
which DBPs may cause adverse health effects, including
cancer and adverse reproductive effects are not well
investigated. Several mechanisms have been suggested
that involve genotoxicity (DeMarini et al. 1997;Pegram
et al. 1997;Ross & Pegram 2004;Richardson et al. 2008),
oxidative stress (Tomasi et al. 1985;Larson & Bull 1992;
Ni et al. 1996;Scholl & Stein 2001;Meek et al. 2002;Gemma
et al. 2003;Sciuto et al. 2003;Weber et al. 2003;Myatt & Cui
2004;Crider et al. 2005;Engel et al. 2005a,b;Min et al.
2006), disruption of folate metabolism (Kamen 1997;Ray &
Laskin 1999;Dow & Green 2000;Geter et al. 2005),
disruption of the synthesis and/or secretion of placental
syncytiotrophoblast-derived chorionic gonadotropin (Chen
et al. 2003,2004) and lowering of testosterone levels (Potter
et al. 1996).
The major limiting factor in these studies has generally been
crude exposure assessment. Use of ecologic water supply
zone estimates as an exposure index may result in exposure
misclassiﬁcation. Furthermore, ingestion has generally been
the primary interest, while uptake through showering,
bathing and swimming is considerable (Whitaker et al.
2003). Combining information on individual water use with
water zone estimates would provide more detailed exposure
assessment, if done appropriately, taking into account
classical and Berkson error models (Nieuwenhuijsen et al.
2000b). Exposure estimates have been based primarily on
residence. This ignores any exposure which occurs outside
the home (e.g. in the workplace) and also ignores the
possibility that a mother has moved house during her
pregnancy. Exposure assessment based on residence there-
fore, results in exposure misclassiﬁcation.
Most of the epidemiological studies have used THMs as
a proxy for total DBP load, but THMs are not necessarily a
good proxy measure. The metabolism of different DBP
species varies (IPCC 2000), so it is insufﬁcient to analyse
DBPs as a whole, or to use TTHM (total THM) as a proxy.
Investigation of the relation between non-THM by-products
and reproductive outcomes is required in order to help
elucidate the speciﬁc DBPs driving the associations
In addition, when chlorine dioxide is used as disin-
fectant agent, chlorite and chlorate are the main DBPs;
the toxicological action due to chlorite and chlorate has
not yet been fully investigated. Only one study has been
carried out in Europe on the association between personal
exposure to these DBPs and pregnancy outcomes. This
study was carried out in nine Italian provinces and
evidenced a small increase in the risk of SGA at term
(term SGA) and high levels of chlorite in drinking water
(Aggazzotti et al. 2004).
Also, for reproductive epidemiological studies, in-depth
analyses comparing exposure metrics for the different
trimesters of pregnancy are required to discover the critical
window in which DBP exposure affects foetal growth.
The retrospective and registry based nature of many of
the reproductive epidemiological studies has meant that
information on potential confounders, and other risk
factors for foetal growth restriction, such as maternal
smoking and alcohol consumption have often been lacking.
Furthermore, for reproductive epidemiological studies,
there is also a need for better case identiﬁcation for
outcomes such as foetal growth restriction and congenital
anomalies. Previous epidemiological studies have used a
variety of outcomes as proxies for foetal growth restriction:
term low birth weight (LBW), intrauterine growth retar-
dation (IUGR) and small for gestational age (SGA). There
are some limitations to these measures. LBW is rather
188 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
crudely deﬁned; the ﬁxed criterion of birth weight below
2,500 g takes no account of population-speciﬁc birth weight
distributions (Wilcox 2001). Somewhat confusingly, the
terms IUGR and SGA have been used interchangeably in
the literature and criteria for IUGR/SGA diagnosis have
varied, some studies using the 5th and some the 10th
percentile of gestational speciﬁc weight according to a
standard population growth chart as a cut-off point. These
measures fail to distinguish between those babies which are
constitutionally small and those which are pathologically
small (i.e. growth restricted). Some small but normally
grown babies will fall below the cut-off point, and some
growth restricted babies will reach a weight above the cut-
off point. A proportion of infants therefore are misclassiﬁed,
and in epidemiological studies this may bias any association
towards the null. There is evidence to show that the use of
customised foetal growth charts, which take into account
factors such as maternal height and ethnicity, signiﬁcantly
reduces the proportion of false-positive and false-negative
diagnoses of foetal growth restriction, compared with the
use of a standard population growth chart (Gelbaya &
Nardo 2005;Gardosi 2006).
Congenital malformations have often been classed into
main categories (e.g. neural tube, major heart and abdomi-
nal defects) as a result of the small number of cases in the
studies. These malformations, however, are generally
heterogeneous with respect to both phenotype and pre-
sumed aetiology. Nieuwenhuijsen et al. (2008) showed that
focusing on isolated subcategories may result in different
Even though there has been some good animal work
that suggests strong effects on semen quality (Smith et al.
1989;Toth et al. 1992;Linder et al. 1994a,b,1995,1997a,b),
only two small US epidemiological studies have been
conducted (Fenster et al. 2003;Luben et al. 2007).
The pooled bladder cancer analysis results have
provided good evidence for some risk related to DBPs;
however, the authors did not examine whether the results in
North America and Europe were consistent (Villanueva
et al. 2004). There may be differences because of different
water treatment practices. Further, the results for colon
cancer have been inconsistent and inconclusive and have
not examined the role of different exposure pathways and
routes, which may be important.
Very few studies have examined gene–environment
interaction and/or the presence of susceptible groups
(Shaw et al. 2003;Infante-Rivard 2004), which is important
for guideline setting and can elucidate potential mechanisms.
THE HIWATE PROJECT
HIWATE is a major research initiative that has started in
Europe (www.hiwate.org) to address the shortcomings of
previous research on DBPs. The overall aim is to investigate
potential human health risks (e.g. cancer, premature births,
SGA, semen quality, stillbirth, congenital anomalies)
associated with long-term exposure to low levels of
disinfectants (such as chlorine) and DBPs occurring in
water for human consumption and use in the food industry.
The study will comprise risk – beneﬁt analyses including
quantitative assessments of risk associated with microbial
contamination of drinking water versus chemical risk
and will compare alternative treatment options. The out-
come will be improved risk assessment/management. The
study will make use of existing studies/databases and newly
collected information. The project involves 16 teams in
eight European countries (Table 2). The work is divided
into different topics (exposure assessment, epidemiology,
risk assessment and management) and studies (Table 3).
The speciﬁc objectives of the proposed work are
(I) To determine the DBP composition and levels in
drinking water in various regions in Europe (see Table 4).
Representative water samples will be primarily collected
in the regions where the epidemiological studies are carried
out, to give a wider picture on their presence and levels.
Samples will be analysed for THMs (chloroform, bromodi-
chloromethane, chlorodibromomethane and bromoform),
haloacetonitriles (HANs) (including chloroacetonitrile, dich-
loroacetonitrile, trichloroacetonitrile, bromoacetonitrile,
dibromoacetonitrile, tribromoacetonitrile, chlorobromoace-
tonitrile, dichlorobromoacetonitrile and dibromochloroace-
tonitrile), HAAs (including chloroacetic acid, dichloroacetic
acid, trichloroacetic acid, dibromoacetic acid, tribromoacetic
189 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
acid, dibromochloroacetic acid, chlorobromoacetic acid,
bromoacetic acid, dichlorobromoacetic acid), haloketones
(HAKs) (including 1,1-dichloropropanone, 1,3-dichloropro-
panone, 1,1,1-trichloropropanone), 3-chloro-4-(dichloro-
methyl)-5-hydroxy-2(5H)-furanone (MX), chlorate hydrate
(CH), chloropicrin (CP), bromate, chlorite and chlorate,
depending on the type of water disinfectant treatment used.
The study will produce a database containing the levels of
these DBPs in the various regions in the UK, France, Spain,
Greece, Italy and Lithuania. The number of samples collected
in each region is given in Table 4. The samples will be
collected over a two year sampling period to provide
information on the temporal variability of the DBPs over
different seasons. Some work will be done to examine the
effects of ﬁlters and boiling water. Detailed sampling and
analyses protocols have been developed (see www.hiwate.
org). Furthermore, an interlaboratory comparison pro-
gramme will be set up to compare the performance of the
various laboratories within the HIWATE consortium with
laboratories outside the HIWATE consortium.
(II) To identify the determinants of DBPs and develop
In addition to the DBP analysis for a range of DBPs (see
objective 1), the study will obtain information regarding the
possible determinants of the DBPs including organic matter
content, water source, temperature, pH and (residual) disin-
fectant levels (e.g. chlorine and bromide level) (Table 5).
Statistical techniques will be employed to quantify the effect
of these determinants on the formation of DBPs and
use this to build a predictive model of DBP formation
(Golﬁnopoulos & Arhonditsis 2002;Nieuwenhuijsen 2003;
Nikolaou et al. 2004;Whitaker et al. 2005). Furthermore,
the correlation between THMs and other DBPs will be
Initially a separate hierarchical model will be built to
describe the data originating from each of the region-
s/countries. These region-speciﬁc models will be of a similar
structure but the determinants included in the ﬁnal models
of each region may differ. We will explore ways in which
these region-speciﬁc models might be combined, for
instance by including an extra level in the hierarchy of the
model structure. This extra level will allow us to explore the
variability between the regions.
Basic mixture model
For each DBP and each region we will build a separate
hierarchical model in order to describe the temporal and
Participants in HIWATE
Participant organisation name Short name organisation Country
Centre for Environmental Epidemiology CREAL Spain
Imperial College London Imperial United Kingdom
University of the Aegean UA Greece
National Public Health Institute KTL Finland
Vytautas Magnus University VMU Lithuania
University of Crete UC Greece
´de Rennes INSERM France
Municipal Institute of Medical Research Foundation FIMIM Spain
Centre for Genomic Regulation CRG Spain
University of Modena and Reggio UMR Italy
Istituto di Ricerche Farmacologiche ‘Mario Negri’ MN Italy
Swedish Institute for Infectious Disease Control SIIDC Sweden
Hylobates Consulting Srl HCS Italy
ICON Ltd ICON United Kingdom
Scarab Scarab Sweden
Catalan Institute of Cancer ICO Spain
190 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
Studies by topic area in the HIWATE study
Exposure assessment studies Epidemiological studies Health risk assessment and management
DBP measurements in UK, France, Spain,
Greece, Italy and Lithuania (WP1)
A nation-wide study of congenital anomalies in
the UK (approx 20,000 cases and 2.5 million
A risk–beneﬁt analysis study including
quantitative assessments of risk associated with
microbial contamination of drinking water
versus chemical risk, compare alternative
treatment options, and produce burden of
disease estimates in Barcelona, Bradford,
Rennes, Heraklion, Kaunas and Modena (WP8)
Modelling of various DBPs using data from
WP1 and information determinants of the
A study of congenital anomalies in the Emilia
Romagna region in Italy (approx 150,000 births)
A water treatment intervention study of stillbirth
and low birth weight in the UK (approx 360,000
A study of small for gestational age and
premature birth in ﬁve pregnancy cohorts in
the UK, Spain, Greece, France and Lithuania
(23,000 births) (WP4)
A review of the water and health policies in
Europe, USA and worldwide in relation to water
A case-control study of semen quality in the UK
(1,700 cases and controls (WP5)
A guide to best practice in terms of water
disinfection and a brief assessment of
disinfection alternatives (WP9)
A pooled analysis of European bladder cancer
studies with almost 6,000 cases and controls
Organise a workshop to bring together scientists
working on environmental, toxicological,
epidemiological and policy aspects of
chlorination DBPs, microbiologists, policy-
makers, and representatives from the water
industry and consumer organisations in Europe
to develop guidelines for policy across Europe
and the future research agenda (WP9)
A case-control study of colon cancer in Italy and
Spain (2,000 cases and controls) (WP7)
191 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
Proposed number of samples to be collected in various regions in Europe
For which workpackage/study
Number of DBP samples
analysed by University
of Aegean, Greece
Number of samples
MX analysed by KTL,
Finland (ﬁrst year)
Number of bromate
by KTL, Finland
analysed by University of
Modena and Reggio, Italy
WP3 congenital anomalies, Italy 20 (10 Emilia Romagna,
5 Milan, 5 Friuli)
20 (7) (10 Emilia
Romagna, 5 Milan,
20 (10 Emilia Romagna,
5 Milan, 5 Friuli)
100 chlorate Emilia
Romagna, 50 Milan,
WP3 þWP5 congenital anomalies,
low birth weight, stillbirth and semen
quality, United Kingdom
100 10 (3) 4
WP4 Birth weight and prematurity,
Bradford, United Kingdom
150 10 (1)
WP4 Birth weight and prematurity,
INMA study areas, Spain
200 10 (3) 4 Some chlorite/chlorate
WP4 Birth weight and prematurity,
Pelagie study area, Rennes, France
60 smaller network
40–80 bigger network
WP4 Birth weight and prematurity,
RHEA study, Crete, Greece, Evripidis,
150 3 (3)
WP4 Birth weight and prematurity,
144 8 (4) 4
WP6 bladder cancer areas No additional
(to be re-evaluated
for speciﬁc areas Spain)
WP7 Colon cancer, Barcelona, Spain 114 10 (3)
WP1 other, e.g. Athens 48
WP1 boiling/ﬁlter experiment To be deﬁned
Total 1,066 81 (27) 32
192 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
193 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
spatial variability of the data in that region. We will follow
the basic hierarchical structure as used by Whitaker et al.
(2005) in modelling THMs. The data will be transformed so
that the values of each DBP that we model are approxi-
mately normally distributed.
In previous work modelling THMs it has been found
that a mixture model is necessary. This form of model is
particularly suited to non-normal distributions where the
underlying data may have arisen from a number of distinct
sources or populations. In Whitaker et al. 2005, the model
assigns a water source type, or some mixture of types (for
instance ground, lowland surface or upland surface), to each
water supply zone. For each DBP model we will explore the
different number of components needed in the mixture
model. It may be desirable to use jump Markov Chain
Monte Carlo techniques to allow the number of com-
ponents to be estimated in the model. For some DBPs a
mixture model may not be required at all. Seasonal variation
is then taken into account by adding a quarterly effect
common to all zones supplied by the same source type.
Furthermore, measurements under the detection limit
will be modelled to obtain an estimate between 0 and the
detection limit, rather than arbitrarily assigning a value of,
for example, half the detection limit. In this approach a
zone mean depends on measurements taken within that
particular zone and the DBP levels for water of the same
source type in other zones, taking into account seasonal
variability across the region. This model can then be used,
after it is back-transformed onto the original scale, to
predict quarterly zone speciﬁc estimates of the DBP of
interest. The model produces robust estimates for each zone
together with an estimate of the degree of uncertainty
around the estimates.
One of the main advantages of the model is that it
provides good estimates for zones where few or no measure-
ments are available. This hierarchical modelling approach
ﬁts well into a Bayesian framework and the software
WinBUGS (Bayesian analysis using Gibbs sampling) will
be used for the estimation (Spiegelhalter 2003). These
techniques have already been successfully applied else-
where and provide a cost efﬁcient way to provide exposure
estimates for current and past exposures for epidemio-
logical and risk assessment studies in areas where infor-
mation on potential determinants is available, but where
there is no or little information on DBP levels. In order to
visualise the modelled DBP estimates and check for any
unusual estimates or potential errors in the modelling, the
modelled quarterly zone mean DBP estimates will be
classiﬁed into exposure categories. The categorised DBP
estimate for each zone and quarter will then be mapped for
each water region, together with the raw annual mean DBP.
These will then be sent to the local water utilities for
The basic model described above aims to explain geo-
graphical and seasonal variability. This model will then be
extended by incorporating factors affecting the creation of
the relevant DBP. There is a wide literature on the
determinants of DBP concentrations in water (see Sadiq
& Rodriguez 2004 for an overview).
By adding additional regression terms to the model
described above we will seek to further explain the
variability in the relevant modelled DBP. In determining
the form of these regression models we will follow the
regression mapping methodology used in this type of study
(Sadiq & Rodriguez 2004). Here, the ﬁtted DBP levels at
each sampling point, as produced in the above hierarchical
model, are treated as ﬁxed and become the dependent
variable in a regression analysis against possible DBP
A simple statistical regression model can be expressed
in the form:
) denotes the log transformed exposure con-
the background level, var
determinant of exposure,
the regression coefﬁcient of
providing the magnitude of the effect, and Ea random
variable with mean 0, often called the error term.
Further, levels can be added to take account of multi-
(III) To assess the risk of reproductive effects in relation to
disinfection practices and levels of disinfection by-products,
194 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
epidemiological studies will be conducted to examine the
relationship between DBP exposure estimates and a
number of outcomes.
(a) Congenital anomalies, including neural tube, major
heart, major stomach wall, and urinary tract defects, cleft
palate/lip will be studied in a large, nation-wide, cross-
sectional study, using registry data in the UK, where mainly
chlorination is used as a disinfectant. The study included
over 2.5 million births and approximately 20,000 cases with
congenital anomalies. The study uses novel Bayesian
statistics for the exposure assessment modelling (Whitaker
et al. 2005). Initial results have been published and showed
no association between THMs and cleft palate/lip, abdomi-
nal wall, major cardiac, neural tube, urinary and respiratory
defects, except for a restricted set of anomalies with isolated
defects. There were excess risks in the highest exposure
categories of total THMs for ventricular septal defects, OR
(odds ratio) ¼1.43 (95% conﬁdence interval (CI) 1.00–
2.04) and of bromoform for major cardiovascular defects
and gastroschisis, OR ¼1.18 (95% CI 1.00 – 1.39) and
OR ¼1.38 (95% CI 1.00–1.92), respectively (Nieuwen-
huijsen et al. 2008).
Congenital anomalies, including neural tube, major
heart and urinary tract defects will also be studied using
registry data in Italy in the Emilia Romagna region, where
mainly chlorine dioxide is used as a disinfectant. The study
includes around 150,000 births and will be analysed as a
case-control study (1:2 ratio). The main exposure variables
are the concentrations of some DBPs (THMs, chlorite and
bromate) in drinking water networks supplying the homes
of the mothers of cases and controls during the ﬁrst
trimester of pregnancy. Moreover, other information on
determinants of DBPs will be collected. On the basis of each
subject’s home address the local water network supplying
drinking water during the period of interest will be
identiﬁed. The following data on waterworks will be
collected: type of water source, disinfection treatment and
supplied population The following analytical data will be
collected: total and individual THMs (chloroform, bromo-
form, bromodichloromethane (BDCM), chlorodibromo-
methane (CDBM)), chlorite, bromate, nitrate, residual
disinfectant, total organic carbon, oxidant power, pH and
hardness. Information on potential confounders will be
collected, such as: social and demographic variables of
mother and father (residence and address, age in years,
nationality, education, occupation, blood relationship);
reproductive history of the mother (parity, number of
previous live births, stillbirths, previous terminations);
present pregnancy (date of last menstrual period, obstetric
history, hospital admissions); termination (date …); delivery
(date, single/plural births, live birth, stillbirth, gender,
weight, length, cranial circumference); and stillbirth
(causes according to international classiﬁcation of diseases
(b/c) Stillbirth and LBW will be studied in an interven-
tion study in the North East of England in the UK, in areas
where enhanced coagulation in the water treatment plant
was installed in 2003. The rates of stillbirth and low birth
weight will be examined 3 years before and after the
intervention. In each year there are around 60,000 births.
Primary analysis will aim to determine whether there is
evidence for a reduction in the rates of stillbirth after
introduction of the new water treatment practices com-
pared with before the changes took place. Secondary
analysis will focus on speciﬁc THM species. Differences in
small-area rates of stillbirth before and after treatment
changes will be modelled against change in total and
individual mean annual THM concentrations using Poisson
regression. For each mgl
decrease in THM concentration,
the increase/decrease in rates will be determined. In
addition, areas will be categorised into low, medium and
high change in THM concentrations and the change in rates
of stillbirth will be estimated in each category using Poisson
regression. Models will be adjusted for potential confoun-
ders such as maternal age and social deprivation (Carstairs’
deprivation score and/or index of multiple deprivation).
(d) SGA, FGR and premature birth will be studied
in ﬁve pregnancy cohorts in the UK, Spain, Greece, France
and Lithuania (Table 6), where a range of treatments
Cohort in France (Pelagie). The ongoing French epide-
miological PELAGIE study (Perturbateurs endocriniens:
´tude Longitudinale sur les Anomalies de la Grossesse,
´et l’Enfance) is a follow up study conducted from
2002 onwards in three departments of Brittany (France):
ˆtes d’Armor and Finiste
`re. Recruitment of
study subjects ﬁnished at the end of 2005, with a population
195 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
of 3,500 study subjects (80% participation rate). The general
objective of the study is the assessment of exposure during
pregnancy to environmental and occupational pollutants,
and evaluate the association with reproductive adverse
effects such as intrauterine growth retardation/small for
gestational age, low birth weight, prematurity and con-
genital malformations. Mothers are recruited in the ﬁrst
trimester of pregnancy through a gynaecologist/obstetri-
cian, general practitioner or echographist. Then, they are
administered a questionnaire and urine samples are
collected. At birth, samples of placenta, mother hair and
cord blood are obtained. Six months after birth, a
neurological test is administered to the child. The existing
exposure assessment: 1) compares regulatory with ad hoc
measurements; 2) evaluates seasonal and geographical
variability of THM levels; and 3) evaluates the relevance
of personal habits (drinking water, showering, bathing, etc.)
in the assessment of THM exposure. In the study area, 150
tap water samples were collected: 100 during October –
November 2004 and 50 during April–May 2005. Individual
questionnaires to collect data on, for example, water
consumption, frequency and duration of showers, baths
and swimming pool attendance have been distributed.
Cohort in Spain (INMA study). The Spanish birth cohort,
called INMA—INfancia y Medio Ambiente (Environment
and Childhood), is a network of research groups in Spain
that have built up a project aiming to study the role of the
most important environmental pollutants in air, water and
diet, life-style and socioeconomic conditions during preg-
nancy and early in life and their effects on child growth and
development (Ramon et al. 2005). It is a prospective
population-based cohort study. Pregnant women are
assessed at 12, 20 and 32 weeks of gestation to collect
information about environmental exposures and foetal
growth, and to obtain maternal venous blood samples
(20 ml). For the DBP analysis, the INMA project will follow
up a population sample of 2,500 pregnant mothers and
newborns recruited in four study areas: Basque country
(N ¼500, enrolment 2005– 2007), Valencia (N ¼800,
2003–2005), Asturias (N ¼500, 2005–2007) and Saba-
dell/Barcelona (N ¼800, 2005 – 2007) (participation rate
approximately 80%). The main exposures of interest in the
study are environmental contaminants in air and water
(trihalomethanes), persistent and semi-persistent pollutants
in different biological samples, maternal occupation, diet
and dietary determinants such as antioxidants, folate and
fatty acids, genetic determinants, social determinants
including parental education, marital status, employment
status and paternal-to-child attachment and paternal
mental status. For the exposure assessment to disinfection
by-products, tap water samples will be taken from the study
areas to measure trihalomethanes. They will also collect
available data from water companies and local authorities.
The study population are administered a questionnaire
including data on water consumption and water-related
habits (e.g. showering, bathing, swimming pool use).
Cohort in Greece (Rhea study). The Greek birth cohort
was initiated on the island of Crete and will enrol all births
in one year within the prefecture of Heraklion, which
includes urban and rural areas with different water supply
sources (N ¼1,700). About one-third of the subjects live in
rural or semi-urban areas and about one-ﬁfth of the
pregnant mother are recent immigrants. The existence of a
well-developed health care system in Crete provides an
Pregnancy cohorts included in the HIWATE study
Total subjects with
Subjects with questionnaire,
and cord and mother’s blood
Subjects with questionnaire
and only cord blood
Subjects with questionnaire
and only mother’s blood
France, Pelagie 3,500 1,500
Greece, Rhea 1,700 1,700
Lithuania, Kaunas, 4,000 4,000
Spain, INMA 2,500 2,500
UK, BiB 10,000 10,000
Total 23,000 15,500 1,500 4,000
196 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
advantage for the identiﬁcation and close follow-up of a
cohort in a relatively closed population. On the basis of
pilot studies, an 80% participation rate is expected for both
questionnaire and biological sample collection. The
majority of study subjects will be identiﬁed through four
main hospitals in Heraklion. Information will be collected
on life-style factors, occupational and environmental
exposures and nutrition, which are predominantly based
on a Mediterranean diet. Follow-up will combine compu-
terised archives with active contacts following procedures
applied in previous children cohorts in Crete. Sources of
water differ substantially in the areas of the study and will
provide a population with contrasted exposures. Analyses
of DBP levels in the past have indicated the presence of
DBPs at levels below those in other Mediterranean coastal
areas while relatively high levels of brominated compounds
have been identiﬁed. Subjects will be personally interviewed
with a computerised interview regarding sources of water,
and other habits related to use of water such as showers,
swimming pools or contact at work. Biological samples will
include blood samples from the mother, cord blood, and
child at age four, urine samples form the mother at
pregnancy, hair of the child, and toenail (mother).
Cohort in Lithuania. Kaunas is a second city of Lithuania
with 400,000 inhabitants and 4,000 births per year. The
Lithuanian epidemiological study is a population-based
cohort study that includes all pregnant Kaunas city women
in 2007–2008 (n¼4,000). The main objective of the study
in Kaunas is to identify the environmental factors that are
associated with newborns’ development and early child-
hood allergy. Pregnant women will be recruited through
antenatal clinics in the city. Mothers will be identiﬁed in the
ﬁrst trimester of pregnancy, mainly though a general
practitioner or gynaecologist and will be interviewed. The
health institutions in Kaunas that register mothers include
four clinical hospitals, 19 outpatient departments, nine
private treatment centres and 15 family health centres.
A second interview will take place in the four main Kaunas
hospitals’ maternity departments. Exposure assessment for
the critical trimester of pregnancy will be based on personal
information on water consumption and other THMs-related
activities obtained through questionnaire, and water work-
level information on water quality—both routinely collected
information and based on water quality analyses of THMs
and exposure modelling. There are four water utility
networks that supply underground water treated by sodium
hypochlorite. Blood samples of the mother will be collected
for genotyping. Information on potential confounders and
modiﬁers (health behaviour, job exposures, sociodemo-
graphic data) will be collected prospectively, during inter-
view by standardised questionnaire.
Cohort in the UK (Born in Bradford). The study popu-
lation is to be drawn from the metropolitan district of
Bradford in the UK. Bradford is the eighth most deprived
health community in the UK with an infant mortality rate
which is signiﬁcantly higher than the UK average. A greater
proportion of babies born in Bradford are of low birth
weight (9.7%) compared with England and Wales as a
whole (7.5%). Nearly 50% of the 5,500 babies born each
year in Bradford are to parents of South Asian origin. The
high prevalence of low birth weight and ethnicity in the
Bradford community provides a unique setting in which to
investigate causes of foetal growth restriction and low birth
weight. Study families (mother, father and index child) will
be recruited by the Born in Bradford (BiB) prospective
cohort study. The study aims to investigate risk factors for
abnormal foetal growth and birth outcomes. Recruitment is
began in February 2007, and it is aimed to recruit 10,000
families over a 2-year period. Participants will be enrolled at
the antenatal glucose tolerance test (26 – 28 weeks ges-
Pooled analyses. We expect to be able to extract from the
existing cohort studies (France, Spain) and obtain from the
new cohort studies (Lithuania, Greece, UK) around 23,000
births for pooled analysis (Table 6). All subjects will have
complete questionnaire data and in most cases both cord
blood and mother’s blood will be available.
Exposure assessment. The ﬁve studies include question-
naires that cover several different areas such as socio-
demographic, life-style, nutrition, occupation, medical and
reproductive history, family history, environmental
exposures and other. The questionnaires used in the studies
in Crete and Spain are fairly similar. All studies have
information on water intake and sources of drinking water.
197 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
In addition, all studies have information on showers, baths
and swimming pool use during pregnancy. The degree of
detail, however, of this information differs considerably
between studies, and an effort will have to be made to adapt
some of the questionnaires. One set of analyses will be
based on the average level of THMs during pregnancy based
on routinely collected THMs for regulatory purposes, and
indices based on the combination of THM measurements
and personal activities such as ingestion, showering, bath-
ing and swimming as an estimate of total dose. This analysis
will be completed with information from measured DBPs
under Work Package 1 that will include several other
compounds apart from THMs. This information will be
modelled (WP2) based on available water quality para-
meters, treatment and water source for the study regions.
Exposure categories will be formed (e.g. none, low, medium
and high) for initial analysis, followed by continuous
indices, if appropriate. The cohorts in Crete, Bradford and
Kaunas will measure many DBPs (through WP1) during the
subjects’ pregnancy, while INMA and PELAGIE have
measurements on THMs and will collect information on
many DBPs (through WP1) after the subjects’ pregnancy.
Modelling techniques will be used (through WP2) to obtain
estimates on various DBPs for all the subjects during the
whole length of pregnancy.
Various exposure indices will be used including average
exposure over the whole pregnancy and also average
exposure during the ﬁrst, second and third trimesters. Use
of trimester-speciﬁc exposure estimates will allow evalu-
ation of the critical exposure window. The questionnaires of
all cohorts include information on the main confounders of
interest such as maternal age and education, socioeconomic
status, parity, smoking and alcohol consumption.
The outcomes that will be measured are:
†low birth weight (LBW)
†small for gestational age (SGA) including symmetrical
and asymmetrical SGA
†foetal growth restriction (FGR) !preferential measure
†parameters derived from the ultrasounds
In addition to DBP metabolising genes, a series of other
genes will be selected that may inﬂuence reproductive
outcomes through other mechanisms such as genes on
oxidative stress and related to the folate-mechanism (e.g
MTHFR). To interpret the function of some of these genes
information should be available through the nutritional
questionnaire on folate and multivitamin use during the
pregnancy since that may have a modifying effect. Candi-
date genes will be identiﬁed on the basis of their reported
involvement in the metabolism of DBPs (i.e. their potential
interaction with environmental exposures). The criteria
used for the selection of candidate genes will be based on
reported biological and genetic relevance (e.g. http://www.
cdc.gov/genomics): (i) evidence from epidemiologic studies
on disease association and gene–environment interaction;
and (ii) evidence of the involvement of the genes in any
reproductive outcome pathobiological pathway. Selection
of speciﬁc SNPs (single nucleotide polymorphisms) in the
genes or regions of interest will be based on established
criteria, including ethnicity, population frequency (e.g.
MAF—minor allele frequency—above 10% for most SNPs),
validation status, location and type of sequence (e.g. coding
sequences, promoter regions, 50UTR and 30UTR, splicing
regions, etc.) and reported or predicted function (e.g. SNPs
evolutionary conserved, SNPs in well-deﬁned domains,
etc.). The ﬁnal selection of genes and SNPs to be analysed
will be decided at a later stage. The genes to be analysed will
include CYPE1, GSTT1, GSTZ and others.
(e) Semen quality will be studied using an existing case-
control study (CHAPS-UK) (Clyma et al. 2008) in the UK,
where mainly chlorination is used for water disinfection.
Subjects were drawn from new patients attending fertility
clinics for investigation: sperm donors were speciﬁcally
excluded. Cases were new male patients seen at any of the
clinics over a 25 month period who had ,12 £10
progressively motile sperm in their initial semen sample.
Around 1,700 cases and controls have been recruited. The
study uses novel Bayesian statistics for the exposure
assessment modelling (Whitaker et al. 2005) and the
exposure and health data will be linked in GIS (geographic
information system). Information on potential confounders
has been collected and the analyses of semen quality and
DBP levels will be adjusted for potential confounders in
logistic regression models. Unfortunately no information is
available on the various exposure pathways and routes and
only DBP concentrations in the water will be used as an
exposure index for the critical exposure windows.
198 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
(IV) To assess the risk of cancer, particularly bladder
cancer and colon cancer, in relation to disinfection by-
product practices and DBP levels, including the examin-
ation of any gene–environment interactions (e.g. CYP2E1,
The study will obtain risk estimates from existing case-
control bladder cancer studies in Spain, France (includes
ozonation as treatment) and Finland, and produce speciﬁc
risk estimates for Europe. The work will build on a pooled
analysis that has been conducted examining long-term
exposure to chlorination by-products by combining resi-
dential information from questionnaires with information
from water utilities gathered in six case-control studies from
the US, Canada and Europe (Villanueva et al. 2004).
It included 2,806 cases of bladder cancer and 5,254
controls. The Finnish case-control study contained 732
bladder cancer cases and 914 controls (Koivusalo et al.
1998). The French study was a hospital-based case-control
study of bladder cancer conducted between 1985 and 1987,
including 765 cases and 765 controls (Chevrier et al. 2004).
The Spanish study is the most recent and included 1,226
cases and 1,271 controls (Villanueva et al. 2007). The cases
and controls have been genotyped (e.g. CYP2E1, NAT2,
GSTM1 and GSTT1), funded by the National Cancer
Institute, and the results will be included in the current
study (Garcia-Closas et al. 2005). The current study will
compare and contract risk estimates from the above studies
and the recently conducted pooled analysis to obtain the
best or a range of risk estimates for various disinfectant
practices and DBPs for the risk – beneﬁt analysis, including
genetically susceptible populations.
A case-control study will be conducted to examine the
relationship between DBPs and colorectal cancer in Spain
and Italy. The main aims are the evaluation of the long-term
exposure to various DBPs in the study subjects through
ingestion, inhalation and dermal absorption and the risk
of colorectal and rectal cancers associated, including
the examination of any gene– environment interactions.
There will be 500 cases and 500 controls in Italy (areas
of great Milan and the provinces of Pordenone and
Udine) and in Spain (Barcelone
`s, Baix Llobregat, Valle
Occidental, Maresme, and Valle
`s Oriental, in Barcelona
province), bringing the total study population to 1,000 cases
and 1,000 controls. Study subjects will be interviewed
face-to-face using a structured questionnaire administered
by trained interviewers. The questionnaire includes infor-
mation on socio-demographics, smoking habit, coffee and
alcohol consumption, diet, physical activity, occupational
exposures, medical history and drug use, family history of
cancer, and detailed information on water use and water-
related habits: drinking water source at each residence from
birth (municipal/private well/other); quantity and type
(bottled/tap) of water consumed, including water based
ﬂuids (coffee, tea and herbal infusions); average frequency
and duration of showering and bathing; lifetime swimming
pool attendance; and dish washing habits. Main potential
confounders and covariates are included in both question-
naires. Each centre has included questions on other
potential risk factors that are not the main focus of this
proposal (e.g. drugs, medical history, etc.). A blood sample
will be collected from each subject. Retrospective infor-
mation on water source, treatment and quality in the study
municipalities will be obtained through a questionnaire
aimed at water companies and local authorities. Tap water
samples will be collected in the study areas to measure a
range of DBPs (as part of WP1). Retrospective DBP levels in
the study areas will be modelled on the basis of historical
data on water source and treatment (see Villanueva et al.
2006). Data on DBP levels will be combined with personal
information on water-related habits. Personal indices of
exposure to DBP through different routes (ingestion,
inhalation and dermal exposure) will be calculated. An
overall index combining different exposure routes will be
also calculated applying weighting factors obtained from
the literature (see Villanueva et al. 2006).
Polymorphisms analysed include several types of mar-
ker: SNPs (single nucleotide polymorphism), In/Del poly-
morphisms (polymorphisms of short deletions or insertions)
and large deletions (e.g. null GST alleles). The selected
study design is the ‘candidate gene approach’ based on the
analysis of those genes potentially involved in a functional
way; for the ﬁrst phase of the study we will focus on those
involved in the DBP metabolism (e.g. GSTT1, CYP2E1,
GSTZ1) and in folate metabolism (e.g. MFTHR). A compre-
hensive review will be conducted to identify key genes that
may be involved in the interplay between DBP exposure
and colorectal cancer risk. Candidate genes will be selected
after discussion between partners.
199 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
Risk assessment and management
(V) To conduct risk–beneﬁt assessment including quanti-
tative assessments of risk associated with microbial contami-
nation of drinking water versus chemical risk, compare
alternative treatment options, and produce burden of
disease estimates (e.g. DALYs, disability-adjusted life years).
The study will build on and make use of expertise and
experience of EC-funded projects such as MICRORISK
(www.microrisk.com) and INTARESE (www.intarese.org).
The purpose of the assessment is deﬁned as the following
research question: ‘What is the net human health impact
of microbial and disinfectant by-product contamination of
drinking water?’ The pyrkilo methodology, an open risk
assessment method, will be used to create an integrated
risk–beneﬁt model (Tuomisto & Pohjola 2007).
We will develop an overall framework for the risk –
beneﬁt analyses of microbial and chemical risks, speciﬁcally
for DBPs in drinking water. The framework will integrate
long-term chemical effects versus the short-term microbial
effects to make realistic comparisons. We will conduct risk–
beneﬁt analyses, including quantitative assessments of risk
associated with microbial contamination of drinking water
versus chemical risk, compare treatment options (e.g.
chlorination, chlorine dioxide and ozonation), and produce
burden of disease estimates. The risk – beneﬁt analyses will
be the result of integrated DPBs and microbial risk
assessments, from modelling of alternative treatment
options and from different risk– beneﬁt metrics, including
burden of disease (e.g. DALYs). As far as we are aware only
one such study has been reported in the literature,
describing a risk–beneﬁt model for Cryptosporidium par-
vum and bromate exposure and comparing the risks and
beneﬁts of ozonation using disability-adjusted life years
(Havelaar et al. 2000).
The work will start with a review to identify the relevant
microbial and DBP exposures and related diseases (e.g.
infectious diarrhoea, gastrointestinal illness and reproduc-
tive and cancer outcomes, respectively). This will be
followed by an exact framing of the risk assessment (more
details in Merila¨ inen et al. 2008). All DBPs from the
exposure assessment part of the HIWATE study (WP1
and 2) will be considered. Information on personal
habits including, for example, ingested amounts of water,
showering, use of ﬁlters and boiling water will be obtained
from the epidemiological studies and the EC-funded
projects MICRORISK and INTARESE, for which this
information was also collected, and from other available
studies (Barbone et al. 2002;Kaur et al. 2004;Westrell et al.
2006) and will be organised in a meaningful and coherent
DBP exposure and risk estimates, including exposure –
response relationships of DBPs will be provided by the
exposure assessment (WP1 and 2) and epidemiological
research areas (WP3, WP4, WP5, WP6, and WP7) of the
current proposal and from the literature, particularly where
pooled or meta-analyses are available (e.g. for bladder
cancer, Villanueva et al. 2004), or we have to rely on
toxicological data. The outcome data for the risk – beneﬁt
analyses will be prioritised using set criteria. For outcomes
such as cancer, long-term exposure will be taken into
account. Where necessary, novel dose–response relation-
ships for DBPs will be derived combining data from the
epidemiological studies, from published toxicological and
other relevant studies (see e.g. Peters et al. 2005).
Exposure estimates for microbiological load will come
from routinely collected data (heterotrophic plate counts
and indicator bacteria including coliforms, Escherichia coli
and Clostridium perfringens as set out in Council Directive
98/83/EC) provided by water companies in the area, litera-
ture, or newly collected data, where necessary. The data will
be linked with failure frequency distributions by converting
indicator values into hypothetical input incidences in
the distribution network (see e.g. Westrell et al. 2003).
We will take into account the relationship with the raw
water quality and its potential content and variability of
microbial load (see e.g. Westrell et al. 2004). Dose – response
relationships will come from MICRORISK and the
The dose–response of and the barrier efﬁciency of
other treatment steps for several other speciﬁc microbes
causing waterborne diseases worldwide but not routinely
measured (including Campylobacteraceae, Mycobacteria,
Giardia,Cryptosporidium, protozoa and enteric viruses)
for disinfection methods will be assessed (see e.g. Persson
et al. 2005). Also, the indicator value of heterotrophic plate
counts for pathogenic waterborne bacteria will be
200 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
Risk estimates for related infectious diseases will come
from the EC-funded MICRORISK, other sources or will be
derived where not available. For the risk assessment, a
combined or best dose–response will be selected based on
the ability to predict cases in an independent data set
(validation with one of the studies, see above). Before
entering the risk–beneﬁt (or risk–risk) analysis, areas of
non-independence of the microbiological and chemical
risks will be examined (e.g. same susceptible or highly
exposed populations, correlation between high DBPs
exposure and higher microbial load). In all the above
work, variability and uncertainty will be incorporated in the
models and sensitivity analyses will be conducted on the
As part of the study, the water consumption, water treat-
ment techniques, treatment performance in water works
and raw water quality will be evaluated in the case study
areas: Barcelona, Bradford, Rennes, Heraklion, Kaunas and
Modena. A few scenarios will be constructed for interven-
tion, and the difference in the outcome measures estimated:
a) change in treatment by water company; b) changes in
behaviour (e.g. change from tap to bottled water); and c) use
of point-of-use measures (e.g. ﬁlters) where they are needed
(speciﬁcally for food industry).
(VI) To review the water and health policies in Europe,
USA and worldwide in relation to water disinfection.
Best practice in terms of water disinfection and a brief
assessment of disinfection alternatives will complete the
study. A ﬁnal workshop will be organised in 2010 as an
open conference that will aim to bring together scientists
working on environmental, toxicological, epidemiological
and policy aspects of chlorination DBPs, microbiologists,
policy-makers, and representatives from the water industry
and consumer organisations in Europe to provide infor-
mation for the development of guidelines for policy across
Europe and the future research agenda. Speciﬁc objectives
include: comparison of policies related to DBPs in drinking
water in Europe, North America and worldwide; review the
current literature on toxicological and epidemiological
ﬁndings of DBPs and adverse health outcomes, including
ﬁndings from the HIWATE epidemiological studies; assess-
ment of the ﬁndings of the HIWATE study in terms of
* DBP measurements
* DBP modelling
Reproductive epidemiological studies
* Birth defects and stillbirth
* Prematurity and SGA
* Semen quality
Microbial vs. chemical risk
*Review of water and health
Cancer epidemiological studies
* Bladder cancer
* Colon cancer
Data, models and expertise from
MICRORISK and INTARESE Literature data on toxicology
Concentration data from
Linkage of work and the application of the risk assessment work into policy.
201 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
current guideline values and treatment practices in
Europe; and recommendations concerning EU legislation
regarding the Water Framework Directive, Directive
98/83/EC, and other related directives. A conceptual
model for application of the risk assessment work into
policy is given in Figure 1.
There appears to be very good epidemiological evidence for
a relationship between chlorination by-products, as
measured by THMs, in drinking water and bladder cancer,
but the evidence for other cancers, including colorectal
cancer appears to be inconclusive and inconsistent. There
appears to be some evidence for a relationship between
chlorination by-products, as measured by THMs, and small
for gestational age (SGA)/intrauterine growth retardation
(IUGR) and preterm delivery, but evidence for other
outcomes such as low birth weight (LBW), stillbirth,
congenital anomalies and semen quality appears to be
inconclusive and inconsistent. Major limitations in exposure
assessment may account for the inconclusive and incon-
sistent results in epidemiological studies, but there are
other issues such as outcome deﬁnition and bias and
The HIWATE study brings together a number of leading
researchers in Europe to carry out the research. A concerted
European research effort has so far been lacking in this
area, resulting in a widening gap of knowledge compared
with North America. A larger number of people including
scientists, policy-makers, industry and consumer represen-
tatives will meet during the proposed open workshop
at the end of the project to produce European guidelines
and recommendations and set a research agenda for
Innovative aspects of the work include detailed
exposure assessment methods in many of the studies
taking into account not only the measurement of water
levels of many by-products but also water-related activi-
ties/pathways such as ingestion, showering, bathing and
swimming and routes (oral, skin absorption and inhala-
tion) producing integrated exposure indices, particularly
for THMs, but also other DBPs where relevant;
examination of gene–environment interaction and identi-
ﬁcation of genetically susceptible groups both in the
epidemiological and risk–beneﬁt studies, and pooling of
studies across countries to increase the power of the
studies. It will provide new risk estimates for various
health outcomes in Europe, including not only cancer
(speciﬁcally colon cancer) but also reproductive outcomes
(speciﬁcally semen quality, foetal growth restriction,
IUGR) and improved risk estimates for various other
outcomes (congenital malformations, stillbirth, low birth
weight) in relation to DBPs in the risk–beneﬁt study.
It will provide a framework and methodology to compare
the microbial and chemical risks, particularly DBPs, and
the risk–beneﬁt study will include a range of DBPs rather
than using, for example, only ‘chlorinated water’ or THMs.
The gene-interaction studies may provide further insight
into the mechanisms of action. For the ﬁrst time there
will be comparable data for a range of DBPs throughout
various regions in Europe. The work is expected to ﬁnish
in April 2010 and an international workshop is planned
in London a few months before the end of the project
(see www.hiwate.org for news).
The work is (partly) funded by HIWATE (www.hiwate.org).
HIWATE is a three-and-a-half year Speciﬁc Targeted
Research Project, funded under the EU Sixth Framework
Programme for Research and Technological Development
by the Research Directorate-Biotechnology, Agriculture
and Food Research Unit (Contract no Food-CT-2006-
036224). The HIWATE consortium consists of more
participants than the current author list and we would
like to thank them for their input.
Aggazzotti, G., Righi, E., Fantuzzi, G., Biasotti, B., Ravera, G.,
Kanitz, S., Barbone, F., Sansebastiano, E., Battaglia, M. A.,
Leoni, V., Fabiani, L., Triassi, M. & Sciacca, S. 2004
Collaborative group for the study of chlorinated drinking
water and pregnancy chlorination by-products (CBPs) in
drinking water and adverse pregnancy outcomes in Italy.
J. Water Health 2, 233–247.
202 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
Aschengrau, A., Zierler, S. & Cohen, A. 1993 Quantity of
community drinking water and the occurrence of late
adverse pregnancy outcomes. Arch. Environ. Health 48,
Barbone, F., Valent, F., Brussi, V., Tomasella, L., Triassi, M.,
Scognamiglio, A., Righi, G., Fantuzzi, E., Casolari, G. &
Aggazzotti, G. 2002 Assessing exposure to drinking water
disinfection byproduct in pregnant women: validity and
reproducibility of a questionnaire.Epidemiology 13, 540 – 544.
Bove, F. J., Fulcomer, M. C., Klotz, J. B., Esmart, J., Dufﬁcy, E. M.
& Savrin, J. E. 1995 Public drinking water contamination and
birth outcomes. Am. J. Epidemiol. 141, 850–862.
Bove, F., Shim, Y. & Zeitz, P. 2002 Drinking water contaminants
and adverse pregnancy outcomes: a review. Environ. Health
Perspect. 110(Suppl. 1), 61–74.
Bove, G. E. Jr, Rogerson, P. A. & Vena, J. E. 2007 Case control study
of the geographic variability of exposure to disinfectant
byproducts and risk for rectal cancer.Int. J. Health Geogr. 6, 18.
Cantor, K. P., Villanueva, C., Garcia-Closas, M., Silverman, D.,
Real, F. X., Dosemeci, M., Malats, N., Yeager, M., Welch, R.,
Chanock, S., Tardon, A., Garcia-Closas, R., Serra, C., Carrato,
A., Castano-Vinyals, G., Samanic, C., Rothman, N. &
Kogevinas, M. 2006 Bladder cancer, disinfection byproducts,
and markers of genetic susceptibility in a case-control study
from Spain.Epidemiology 17(6), S150.
Cedergren, M. I., Selbing, A. J., Lofman, O. & Kallen, B. 2002
Chlorination byproducts and nitrate in drinking water and risk
of congenital cardiac defects.Environ. Res. Section A 89,
Chen, J., Douglas, G. C., Thirkill, T. L., Lohstroh, P. N., Bielmeier,
S. R., Narotsky, M. G., Best, D. S., Harrison, R. A., Natarajan,
K., Pegram, R. A., Overstreet, J. W. & Lasley, B. L. 2003 Effect
of bromodichloromethane on chorionic gonadotrophin
secretion by human placental trophoblast cultures.Toxicol.
Sci. 76, 75–82.
Chen, J., Thirkill, T. L., Lohstroh, P. N., Bielmeier, S. R., Narotsky,
M. G., Best, D. S., Harrison, R. A., Natarajan, K., Pegram,
R. A., Overstreet, J. W., Lasley, B. L. & Douglas, G. C. 2004
Bromodichloromethane inhibits human placental trophoblast
differentiation.Toxicol. Sci. 78, 166– 174.
Chevrier, C., Junod, B. & Cordier, S. 2004 Does ozonation of
drinking water reduce the risk of bladder cancer?
Epidemiology 15(5), 605–614.
Clyma, J.-A., Dippnall, M., Baillie, H., Povey, A., Cherry, N., Moore,
H., McNamee, R., Pacey, A. & Burgess, G. 2008 Occupation
and male infertility: glycol ethers and other exposures.OEM
Crider, K. S., Whitehead, N. & Buus, R. M. 2005 Genetic variation
associated with preterm birth: a HuGE review. Genet. Med. 7,
DeMarini, D. M., Shelton, M. L., Warren, S. H., Ross, T. M., Shim,
J. Y., Richard, A. M. & Pegram, R. A. 1997 Glutathione
S-transferase-mediated induction of GC!AT transitions by
halomethanes in Salmonella.Environ. Mol. Mutagen. 30,
Dodds, L. & King, W. D. 2001 Relation between trihalomethane
compounds and birth defects.Occup. Environ. Med. 58,
Dodds, L., King, W., Woolcott, C. & Pole, J. 1999 Trihalomethanes
in public water supplies and adverse birth outcomes.
Epidemiology 10, 233–237.
Dodds, L., King, W., Allen, A. C., Armson, A., Fell, D. B. &
Nimrod, C. 2004 Trihalomethanes in public water supplies and
risk of stillbirth.Epidemiology 15, 179–186.
Dojlido, J., Zbiec
ˆwietlik, R. 1999 Formation of the
haloacetic acids during ozonation and chlorination of water in
Warsaw waterworks (Poland).Water Res. 33, 3111–3118.
Doria, M. F., Pidgeon, N. & Hunter, P. 2005 Perception of tap
water risks and quality: a structural equation model approach.
Water Sci. Technol. 52(8), 143–149.
Doria, M. F. 2006 Bottled water versus tap water: understanding
consumers’ preferences.J. Water Health 4(2), 271–276.
Dow, J. L. & Green, T. 2000 Trichloroethylene induced vitamin
B12 and folate deﬁciency leads to increased formic acid
excretion in the rat.Toxicology 146, 123–136.
Doyle, T. J., Zheng, W., Cerhan, J. R., Hong, C. P., Sellers, T. A.,
Kushi, L. H. & Folsom, A. R. 1997 The association of drinking
water source and chlorination by-products with cancer
incidence among postmenopausal women in Iowa: a
prospective cohort study.Am. J. Public Health 87,
Engel, S. A., Erichsen, H. C., Savitz, D. A., Thorp, J., Chanock, S. J.
& Olshan, A. F. 2005aRisk of spontaneous preterm birth is
associated with common proinﬂammatory cytokine
polymorphisms.Epidemiology 16, 469–477.
Engel, S. A., Olshan, A. F., Savitz, D. A., Thorp, J., Erichsen, H. C.
& Chanock, S. J. 2005bRisk of small-for-gestational age is
associated with common anti-inﬂammatory cytokine
polymorphisms.Epidemiology 16, 478–486.
Fantuzzi, G., Aggazzotti, G., Righi, E., Predieri, G., Giacobazzi, P.,
Kanitz, S., Barbone, F., Sansebastiano, G., Ricci, C., Leoni, V.,
Fabiani, L., Triassi, M. & Sciacca, S. and Collaborative Group
for the Study of Chlorinated Drinking Waters and Pregnancy
2007 Valutazione dell’esposizione a composti organoalogenati
nelle acque potabili di 9 province italiane: indagine sulla
presenza di cloriti, clorati, trialometani, trielina e
tetracloroetilene. Ann. Igiene 19, 345–354.
Fenster, L., Waller, K., Windham, G., Henneman, T., Anderson, M.,
Mendola, P., Overstreet, J. W. & Swan, S. H. 2003
Trihalomethane levels in home tap water and semen quality.
Epidemiology 14, 650–658.
Gallagher, M. D., Nuckols, J. R., Stallones, L. & Savitz, D. A. 1998
Exposure to trihalomethanes and adverse pregnancy
outcomes.Epidemiology 9, 484–489.
Garcia-Closas, M., Malats, N., Silverman, D., Dosemeci, M.,
Kogevinas, M., Hein, D. W., Tardon, A., Serra, C., Carrato, A.,
Garcia-Closas, R., Lloreta, J., Castano-Vinyals, G., Yeager, M.,
Welch, R., Chanock, S., Chatterjee, N., Wacholder, S.,
Samanic, C., Tora, M., Fernandez, F., Real, F. X. & Rothman,
N. 2005 NAT2 slow acetylation, GSTM1 null genotype,
203 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
and risk of bladder cancer: results from the Spanish
Bladder Cancer Study and meta-analyses.Lancet 366,
Gardosi, J. 2006 New deﬁnition of small for gestational age
based on fetal growth potential.Horm. Res. 65(suppl 3), 15 –18.
Gelbaya, T. A. & Nardo, L. G. 2005 Customised fetal growth
chart: a systematic review.J. Obstet. Gynaecol. 25(5), 445 –450.
Gemma, S., Vittozzi, L. & Testai, E. 2003 Metabolism of chloroform
in the human liver and identiﬁcation of the competent P450s.
Drug Metab. Dispos. 31, 266–274.
Geter, D. R., Moore, T. M., George, M. H., Kilburn, S. R., Allen,
J. W., Nelson, G. M., Winkﬁeld, E. & DeAngelo, A. B. 2005
Tribromomethane exposure and dietary folate deﬁciency in the
formation of aberrant crypt foci in the colons of F344/N rats.
Food Chem. Toxicol. 43, 1405–1412.
Gevecker Graves, C., Matanoski, G. M. & Tarfdiff, R. G. 2001
Weight of evidence for an association between adverse
reproductive and developmental effects and exposure to
disinfection by-products: a critical review.Reg. Tox. Pharm.
Golﬁnopoulos, S. K. & Arhonditsis, G. B. 2002 Multiple regression
models: a methodology for evaluating trihalomethane
concentrations in drinking water from raw water
characteristics.Chemosphere 47(9), 1007 –1018.
Golﬁnopoulos, S. K. & Nikolaou, A. D. 2005 Survey of disinfection
by-products in drinking water in Athens, Greece.Desalination
176(1–3), 13 – 24.
Havelaar, A. H., De Hollander, A. E., Teunis, P. F., Evers, E. G.,
Van Kranen, H. J., Versteegh, J. F., Van Koten, J. E. & Slob,
W. 2000 Balancing the risks and beneﬁts of drinking water
disinfection: disability adjusted life-years on the scale.
Environ. Health Perspect. 108(4), 315–321.
Hildesheim, M. E., Cantor, K. P., Lynch, C. F., Dosemeci, M.,
Lubin, J., Alavanja, M. & Craun, G. 1998 Drinking water
source and chlorination byproducts. II. Risk of colon and
rectal cancers.Epidemiology 9, 29–35.
Hinckley, A. F., Bachand, A. M. & Reif, J. S. 2005 Late pregnancy
exposures to disinfection by-products and growth-related birth
outcomes. Environ. Health Perspect. 113, 1808–1813.
Hwang, B.-F. & Jaakkola, J. J. K. 2003 Water chlorination and birth
defects: a systematic review and meta-analysis.Arch. Environ.
Health 58, 83–91.
Hwang, B.-F., Magnus, P. & Jaakkola, J. J. K. 2002 Risk of speciﬁc
birth defects in relation to chlorination and the amount of
natural organic matter in the water supply.Am. J. Epidemiol.
Infante-Rivard, C. 2004 Drinking water contaminants, gene
polymorphisms and fetal growth. Environ. Health Perspect.
IPCS 2000 Disinfectants and disinfectant by-products. Environ.
Health Criteria 216.
Jaakkola, J. J. K., Magnus, P., Skrondal, A., Hwang, B.-F., Becher,
G. & Dybing, E. 2001 Foetal growth and duration of gestation
relative to water chlorination.Occup. Environ. Med. 58,
Kallen, B. A. J. & Robert, E. 2000 Drinking water chlorination and
delivery outcome: a registry-based study in Sweden.Reprod.
Toxicol. 14, 303–309.
Kamen, B. 1997 Folate and antifolate pharmacology. Semin. Oncol.
Kanitz, S., Franco, Y., Patrone, V., Caltabellotta, M., Raffo, E.,
Riggi, C., Timitilli, D. & Ravera, G. 1996 Association between
drinking water disinfection and somatic parameters at birth.
Environ. Health Persperct. 104, 516–520.
Karagas, M., Villanueva, C. M., Nieuwenhuijsen, M. J., Weisel, C. P.,
Cantor, C. P. & Kogevinas, M. 2008 Disinfection by-products
and skin cancer: a hypothesis.Cancer Cause Control 19(5),
Kaur, S., Nieuwenhuijsen, M. J., Ferrier, H. & Steer, P. 2004
Exposure of pregnant women to tap water related activities in
London and validation of the questionnaire.Occup. Environ.
Med. 61, 454–460.
King, W. D., Marrett, L. D. & Woolcott, C. G. 2000 Case-control
study of colon and rectal cancers and chlorination by-products
in treated water. Cancer Epidemiol. Biomarkers Prev. 9,
King, W. D., Dodds, L. & Allen, A. C. 2000 Relation between
stillbirth and speciﬁc chlorination by-products in public water
supplies.Environ. Health Perspect. 108, 883–886.
King, W. D., Dodds, L., Allen, A. C., Armson, B. A., Fell, D. &
Nimrod, C. 2005 Haloaceticacids in drinking water and risk
for stillbirth.Occup. Environ. Med. 62, 124–127.
Klotz, J. B. & Pyrch, L. A. 1999 Neural tube defects and
drinking water disinfection by-products.Epidemiology 10,
Koivusalo, M. & Vartiainen, T. 1997 Drinking water chlorination
by-products and cancer. Rev. Environ. Health 12, 81 – 90.
Koivusalo, M., Hakulinen, T., Vartiainen, T., Pukkala, E., Jaakkola,
J. J. & Tuomisto, J. 1998 Drinking water mutagenicity and
urinary tract cancers: a population-based case-control study in
Finland. Am. J. Epidemiol. 148, 704 – 712.
Kramer, M. D., Lynch, C. F., Isacson, P. & Hanson, J. W. 1992
The association of waterborne chloroform with
intrauterine growth retardation.Epidemiology 3, 407 –413.
Krasner, S. W., Weinberg, H. S., Richardson, S. D., Pastor, S. J.,
Chinn, R., Sclimenti, M. J., Onstad, G. D. & Thruston, A. D.,
Jr 2006 Occurrence of a new generation of disinfection
byproducts.Environ. Sci. Technol. 40, 7112 –7113.
Larson, J. & Bull, R. 1992 Metabolism and lipid peroxidative
activity of trichloroacetate and dichloroacetate in
rats and mice.Toxicol. Appl. Pharmacol. 115,
Lewis, C., Suffet, I. H. & Ritz, B. 2006 Estimated effects of
disinfection by-products on birth weight in a population
served by a single water utility.Am. J. Epidemiol.
Lewis, C., Suffet, I. H., Hoggatt, K. & Ritz, B. 2007 Estimated effects
of disinfection by-products on preterm birth in a population
served by a single water utility.Environ. Health Perspect. 115,
204 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
Linder, R. E., Klinefelter, G. L., Strader, L. F., Suares, J. D. & Dyer,
C. J. 1994aAcute spermatogenic effects of bromoacetic acids.
Fund. Appl. Toxicol. 22, 422–430.
Linder, R. E., Klinefelter, G. R., Strader, L. F., Suarez, J. D., Roberts,
N. L. & Dyer, C. J. 1994bSpermatotoxicity of dibromoacetic acid
in rats after 14 daily exposures.Reprod. Toxicol. 8, 251– 259.
Linder, R. E., Klinefelter, G. R., Strader, L. F., Suarez, J. D.,
Roberts, N. L. & Perreault, S. D. 1995 Dibromoacetic acid
affects reproductive competence and sperm quality in the male
rat.Fund. Appl. Toxicol. 28, 9–17.
Linder, R. E., Klinefelter, G. R., Strader, L. F., Suarez, J. D. &
Roberts, N. L. 1997aSpermatotoxicity of dichloroacetic acid.
Reprod. Toxicol. 11, 681–688.
Linder, R. E., Klinefelter, G. R., Strader, L. F., Veeramachanemi,
D. N., Roberts, N. L. & Suarez, J. D. 1997bHistopathologic
changes in testes of rats exposed to dibromoacetic acid.
Reprod. Toxicol. 11, 47–56.
Luben, T. J., Olshan, A. F., Herring, A. H., Jeffay, S., Strader, L., Buus,
R. M., Chan, R. L., Savitz, D. A., Singer, P. C., Weinberg, H. S. &
Perreault, S. D. 2007 The healthy men study: an evaluation of
exposure to disinfection by-products in tap water and sperm
quality.Environ. Health Perspect. 115, 1169 –1176.
Magnus, P., Jaakkola, J. J. K., Skrondal, A., Alexander, J., Becher,
G., Krogh, T. & Dybing, E. 1999 Water chlorination and birth
defects.Epidemiology 10, 513–517.
Malliarou, M., Collins, C., Graham, N. & Nieuwenhuijsen, M. J.
2005 Haloacetic acids in drinking water in the UK.Water Res.
Meek, M. E., Beauchamp, R., Long, G., Moir, D., Turner, L. &
Walker, M. 2002 Chloroform: exposure estimation, hazard
characterization, and exposure-response analysis.J. Toxicol.
Environ. Health B Crit. Rev. 5, 283–334.
Merila¨ inen, P., Lehtola, M., Grellier, J., Iszatt, N., Nieuwenhuijsen,
M., Vartiainen, T. & Tuomisto, J. T. 2008 Developing a
conceptual model for risk-beneﬁt analysis of disinfection by-
products and microbes in drinking water (in preparation).
Min, J., Park, H., Park, B., Kim, Y. J., Park, J., Lee, H., Ha, E., Park,
E. & Hong, Y. C. 2006 Paraoxonase gene polymorphism and
vitamin levels during pregnancy: relationship with maternal
oxidative stress and neonatal birth weights.Reprod. Toxicol.
Myatt, L. & Cui, X. 2004 Oxidative stress in the placenta.
Histochem. Cell Biol. 122, 369 –382.
Ni, Y. C., Wong, T. Y., Lloyd, R. V., Heinze, T. M., Shelton, S.,
Casciano, D., Kadlubar, F. F. & Fu, P. P. 1996 Mouse liver
microsomal metabolism of chloral hydrate, trichloroacetic acid
and trichloroethanol leading to induction of lipid peroxidation
via a free radical mechanism. Drug Metab. Dispos. 24, 81–90.
Nieuwenhuijsen, M. J. (ed.) 2003 Exposure Assessment in
Occupational and Environmental Epidemiology. Oxford
University Press, Oxford.
Nieuwenhuijsen, M. J., Toledano, M. B. & Elliott, P. 2000aUptake
of chlorination disinfection by-products; a review and a
discussion of its implications for epidemiological studies.
J. Expo. Anal. Environ. Epidemiol. 10, 586–599.
Nieuwenhuijsen, M. J., Toledano, M. B., Eaton, N. E., Elliott, P. &
Fawell, J. 2000bChlorination disinfection by-products in water
and their association with adverse reproductive outcomes: a
review.Occup. Environ. Med. 57, 73–85.
Nieuwenhuijsen, M. J., Toledano, M. B., Bennett, J., Best, N.,
Konstantinou, K., Hambly, P., de Hoogh, K., Boyd, P.,
Abramsky, L., Wellesley, D., Dattani, N., Briggs, D., Lars Jarup,
L. & Elliott, P. 2008 The relationship between disinfection
byproducts in drinking water and congenital anomalies in
England and Wales.Environ. Health Perspect. 116, 216–222.
Nikolaou, A. D., Golﬁnopoulos, S. K., Arhonditsis, G. B.,
Kolovoyiannis, V. & Lekkas, T. D. 2004 Modeling the
formation of chlorination by-products in river waters with
different quality.Chemosphere 55, 409– 420.
Nissinen, T. K., Miettinen, I. T., Martikainen, P. J. & Vartiainen, T.
2002 Disinfection by-products in Finnish drinking waters.
Chemosphere 48, 9–20.
Palacios, M., Pampillon, J. F. & Rodriguez, M. E. 2000
Organohalogenated compounds levels in chlorinated drinking
waters and current compliance with quality standards
throughout the European Union.Water Res. 34, 1002 – 1016.
Pegram, R. A., Andersen, M. E., Warren, S. H., Ross, T. M. &
Claxton, L. D. 1997 Glutathione S-transferase-mediated
mutagenicity of thihalomethanes in Salmonella typhimurium:
contrasting results with bromodichloromethane and
chloroform.Toxicol. Appl. Pharmacol. 144, 183 –188.
Persson, F., La
˚ngmark, J., Heinicke, G., Hedberg, T., Tobiason, J.,
Stenstro¨ m, T. A. & Hermansson, M. 2005 Characterisation of
the behaviour of particles in bioﬁlters for pre-treatment of
drinking water.Wat. Res. 39(16), 3791–3800.
Peters, J. L., Rushton, L., Sutton, A. J., Jones, D. R., Abrams, K. R.
& Mugglestone, M. A. 2005 Bayesian methods for the cross-
design synthesis of epidemiological and toxicological evidence.
Appl. Stat. 54(1), 159–172.
Porter, C. K., Putnam, S. D., Hunting, K. L. & Riddle, M. R. 2005
The effect of trihalomethane and haloacetic acid exposure on
fetal growth in Maryland county.Am. J. Epidemiol. 162,
Potter, C. L., Chang, L. W., DeAngelo, A. B. & Daniel, F. B. 1996
Effects of four trihalomethanes on DNA strand breaks, renal
hyaline droplet formation and serum testosterone in male
F-344 rats.Cancer Lett. 106, 235 –242.
Ramon, R., Ballester, F., Rebagliato, M., Ribas, N., Torrent, M.,
Fernandez, M., Sala, M., Tardon, A., Marco, A., Posada, M.,
Grimalt, J. & Sunyer, J. 2005 Red INMA. The Environment
and Childhood Research Network (“INMA” network): study
protocol.Rev. Esp. Salud Publica. 79(2), 203 –220.
Ray, J. G. & Laskin, C. A. 1999 Folic acid and homocyst(e)ine
metabolic defects and the risk of placental abruption, pre-
eclampsia and spontaneous pregnancy loss: a systematic
review.Placenta 20, 519–529.
Reif, J. S., Hatch, M. C., Bracken, M., Holmes, L. B., Schwetz, B. A.
& Singer, P. C. 1996 Reproductive and development effects of
disinfection by-products in drinking water.Environ. Health
Perspect. 104, 1056–1061.
205 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
Richardson, S. 1998 Drinking water disinfection by-products.
In Encyclopedia of Environmental Analysis and Remediation
(R. A., Meyers, ed.), pp. 1398–1421. John Wiley & Sons,
Richardson, S. D., Plewa, M. J., Wagner, E. D., Schoeny, R. &
DeMarini, D. M. 2008 Occurrence, genotoxicity, and
carcinogenicity of emerging disinfection by-products in
drinking water: a review and roadmap for research.Mutat.
Res. Rev. Mutat. Res. doi:10.1016/j.mrrev.2007.09.001.
Rook, J. J. 1974 Formation of haloforms during chlorination of
natural waters. J. Soc. Water Treat. Exam. 23, 234–243.
Ross, M. K. & Pegram, R. A. 2004 In vitro biotransformation and
genotoxicity of the drinking water disinfection byproduct
bromodichloromethane: DNA binding mediated by glutathione
transferase theta 1-1.Toxicol. Appl. Pharmacol. 195, 166 – 181.
Sadiq, R. & Rodriguez, M. 2004 Disinfection by-products (DBPs) in
drinking water and predictive models for their occurrence: a
review.Sci. Total Environ. 321(1–3), 21–46.
Savitz, D. A., Andrews, K. W. & Pastore, L. M. 1995 Drinking water
and pregnancy outcome in Central North Carolina: Source,
amount and trihalomethane levels.Environ. Health Perspect.
Savitz, D. A., Singer, P. C., Herring, A. H., Hartmann, K. E.,
Howard, S., Weinberg, H. S. & Makarushka, C. 2006 Exposure
to drinking water disinfection by-products and pregnancy loss.
Am. J. Epidemiol. 164, 1043 – 1051.
Scholl, T. O. & Stein, T. P. 2001 Oxidant damage to DNA and
pregnancy outcome.J. Matern. Fetal Med. 10, 182–185.
Sciuto, A. M., Clapp, D. L., Hess, Z. A. & Moran, T. S. 2003 The
temporal proﬁle of cytokines in the bronchoalveolar lavage
ﬂuid in mice exposed to the industrial gas phosgene.Inhal.
Toxicol. 15, 687–700.
Shaw, G. M., Ranatunga, D., Quach, T., Neri, E., Correa, A. &
Neutra, R. 2003 Trihalomethane exposures from municipal
water supplies and selected congenital malformations.
Epidemiology 14, 191–199.
Smith, M. K., Randall, J. L. & Stober, J. A. 1989 Teratogenic activity
of trichloroacetic acid in the rat.Teratology 40, 445 – 451.
Spiegelhalter, D. J. 2003 WinBUGS: Bayesian Inference Using
Gibbs Sampling Manual Version 1.4. Imperial College,
London and MRC Biostatistics Unit, Cambridge.
Tardiff, R. G., Carson, M. L. & Ginevan, M. E. 2006 Updated weight
of evidence for an association between adverse reproductive
and developmental effects and exposure to disinfection
by-products.Regul. Toxicol. Pharmacol. 45, 185 – 205.
Toledano, M. B., Nieuwenhuijsen, M. J., Best, N., Whitaker, H.,
Hambly, P., de Hoogh, C., Fawell, J., Jarup, L. & Elliott, P.
2005 Relation of trihalomethane concentration in public water
supplies to stillbirth and birth weight in three water regions in
England.Environ. Health Perspect. 113, 225–232.
Tomasi, A., Albano, E., Biasi, F., Slater, T., Vannini, V. & Dianzani,
M. U. 1985 Activation of chloroform and related
trihalomethanes to free radical intermediates in isolated
hepatocytes and in the rat in vivo as detected by the ESP-Spin
trapping technique.Chem.-Biol. Interact. 55, 303 – 316.
Toth, G. P., Kelty, K. C., George, E. L., Read, E. J. & Smith, M. K.
1992 Adverse male reproductive effects following subchronic
exposure of rats to sodium dichloroacetate.Fund. Appl.
Toxicol. 19, 57–63.
Tuomisto, J. T. & Pohjola, M (eds) 2007 Open Risk Assessment.
A New Way of Providing Scientiﬁc Information for
Decision-making. Kansanterveyslaitoksen julkaisuja B18,
Villanueva, C. M., Grimalt, J. O. & Kogevinas, M. 2003 Haloacetic
acids and trihalomethane concentrations in ﬁnished
drinking waters from different sources.Water Res. 37(4),
Villanueva, C. M., Cantor, K. P., Cordier, S., Jaakkola, J. J., King,
W. D., Lynch, C. F., Porru, S. & Kogevinas, M. 2004
Disinfection byproducts and bladder cancer: a pooled analysis.
Epidemiology 15, 357–367.
Villanueva, C. M., Cantor, K. P., Grimalt, J. O., Castano-Vinyals, G.,
Malats, N., Silverman, D., Tardon, A., Garcia-Closas, R., Serra,
C., Carrato, A., Rothman, N., Real, F. X., Dosemeci, M. &
Kogevinas, M. 2006 Assessment of lifetime exposure to
trihalomethanes through different routes.Occup. Environ.
Med. 63(4), 273–277.
Villanueva, C. M., Cantor, K. P., Grimalt, J. O., Malats, N.,
Silverman, D., Tardon, A., Garcia-Closas, R., Serra, C.,
Carrato, A., Castan
˜o-Vinyals, G., Marcos, R., Rothman, N.,
Real, F. X., Dosemeci, M. & Kogevinas, M. 2007 Bladder
cancer and exposure to water disinfection by-products through
ingestion, bathing, showering and swimming in pools.Am.
J. Epidemiol. 165, 148–156.
Waller, K., Swan, S. H., DeLorenze, G. & Hopkins, B. 1998
Trihalomethanes in drinking water and spontaneous abortion.
Epidemiology 9, 134–140.
Waller, K., Swan, S., Windham, G. C. & Fenster, L. 2001 Inﬂuence
of exposure assessment methods on risk estimates in an
epidemiologic study of trihalomethane exposure and
spontaneous abortion.J. Expo. Anal. Environ. Epidemiol. 11,
Weber, L. W., Boll, M. & Stampﬂ, A. 2003 Hepatotoxicity and
mechanism of action of haloalkanes: carbon tetrachloride as a
toxicological model.Crit. Rev. Toxicol. 33, 105–136.
Weinberg, H. S., Krasner, S. W., Richardson, S. D. & Thruston,
A. D. 2002 The Occurrence Of Disinfection By-Products
(DBPs) of Health Concern in Drinking Water: Results of a
Nationwide DBP Occurrence Study. EPA report: EPA 600/
Westrell, T., Bergstedt, O., Stenstro¨ m, T-A. & Ashbolt, N. 2003
A theoretical approach to assess microbial risks due to failures
in drinking water systems.Int. J. Environ. Health 13(2),
Westrell, T., Schonning, C., Stenstrom, T. A. & Ashbolt, N. J. 2004
QMRA (quantitative microbial risk assessment) and HACCP
(hazard analysis and critical control points) for management
of pathogens in wastewater and sewage sludge treatment and
reuse. Water Sci. Technol. 50(2), 23–30.
206 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health
Westrell, T., Andersson, Y. & Stenstro¨m,T.A.2006 Drinking water
consumption patterns in Sweden. J. Water Health 4(4),
Whitaker, H., Nieuwenhuijsen, M. J. & Best, N. 2003 The
relationship between water chloroform levels and uptake of
chloroform: a simulation study. Environ. Health Perspect. 111,
Whitaker, H., Best, N., Nieuwenhuijsen, M. J., Wakeﬁeld, J., Fawell,
J. & Elliott, P. 2005 Modelling exposure to disinfection by-
products in drinking water for an epidemiological study of
adverse birth outcomes.J. Expo. Anal. Environ. Epidemiol. 15,
Wilcox, A. J. 2001 On the importance—and the unimportance—of
birthweight.Int. J. Epidemiol. 30, 1233–1241.
Wilkins, J. R. III & Comstock, G. W. 1981 Source of drinking
water at home and site-speciﬁc cancer incidence in
Washington County, Maryland. Am. J. Epidemiol. 114,
Wright, J. M., Schwartz, J. & Dockery, D. W. 2003 Effect of
trihalomethane exposure on fetal development.Occup.
Environ. Med. 60, 173–180.
Wright, J. M., Schwartz, J. & Dockery, D. W. 2004 The effect of
disinfection by-products and mutagenic activity on birth
weight and gestational duration. Environ. Health Perspect.
Yang, C.-Y. 2004 Drinking water chlorination and adverse birth
outcomes in Taiwan.Toxicology 198, 249–254.
Yang, C.-Y., Cheng, B.-H., Tsai, S.-S., Wu, T.-N., Lin, M.-C. & Lin,
K.-C. 2000 Association between chlorination of drinking water
and adverse pregnancy outcome in Taiwan.Environ. Health
Perspect. 108, 765–768.
Yang, C. Y., Xiao, Z. P., Ho, S. C., Wu, T. N. & Tsai, S. S. 2007
Association between trihalomethane concentrations in
drinking water and adverse pregnancy outcome in Taiwan.
Environ. Res. 104(3), 390–395.
Young, T. B., Kanarek, M. S. & Tsiatis, A. A. 1981 Epidemiologic
study of drinking water chlorination and Wisconsin female
cancer mortality. J. Natl Cancer Inst. 67, 1191 – 1198.
Young, T. B., Wolf, D. A. & Kanarek, M. S. 1987 Case-control study
of colon cancer and drinking water trihalomethanes in
Wisconsin.Int. J. Epidemiol. 16, 190–197.
First received 2 April 2008; accepted in revised form 26 July 2008. Available online February 2009
207 M. J. Nieuwenhuijsen et al.
HIWATE Journal of Water and Health