introduction and objective of the seminar - circabc industries... · 1 introduction and objective...

Scientific Synthesis ReportDrinking Water Seminar27 and 28 October 2003

December 2003

Seminar on drinking water Brussels 27-28 October 2003- 1 - December 2003

Contents

Contents 1

1 Introduction and objective of the seminar 3

1.1 Introduction 3

1.2 The revision of the Directive 3

1.3 General approach 4

2 Preparation of the seminar 5

2.1 Prepatory groups and prepatory documents 5

2.2 Position papers (original papers as they were put on the web-site) 72.2.1 Position paper on Chemical parameters 72.2.2 Position paper on Microbiology 192.2.3 Position paper on Endocrine Disrupting Compounds 242.2.4 Position paper on Risk Approach for Drinking water 322.2.5 Position paper on materials and substances in contact with drinking water 52

3 Drinking Water Seminar (day 1) 62

3.1 Presentation P. Perera and Agenda of the Seminar 62

3.2 Outcome of parallel sessions day 1 of the DWS 623.2.1 Session on chemical parameters 623.2.2 Synthesis of the discussion on chemical parameters 623.2.3 Session on microbiology 653.2.4 Synthesis of the discussion on microbiology 653.2.5 Session on endocrine disrupting compounds 683.2.6 Synthesis of the discussion endocrine disrupting compounds 683.2.7 Session on risk approach for drinking water 703.2.8 Synthesis of the discussionon risk approach for drinking water 703.2.9 Session on materials and substances in contact with drinking water 713.2.10 Synthesis of the discussion on materials and substances in contact with drinking

water 71

4 Drinking Water Seminar (day 2) 76

4.1 Presentation WHO Jamie Bartram 76

4.2 Amended presentations preparatory groups 764.2.1 Amended presentation chemical parameters 764.2.2 Synthesis of the plenary discussion on chemical parameters 764.2.3 Amended presentation microbiology 784.2.4 Synthesis of the plenary discussion on microbiology 784.2.5 Amended presentation on Endocrine disrupting compounds 794.2.6 Synthesis of the plenary discussion on endocrine disrupting compoundes 794.2.7 Amended presentation risk approach for drinking water 814.2.8 Synthesis of the plenary discussion risk approach for drinking water 814.2.9 Amended presentation material and substances in contact with drinking water 83


4.2.10 Synthesis of the plenary discussion on material and substances in contact withdrinking water 83

5 Position papers from stakeholders 87

6 Synthesis of main issues and concluding remarks 88

6.1 Introduction 88

6.2 Participation in the DWS 89

6.3 Summary and synthesis of the main outcome of the seminar 906.3.1 Chemical parameters 906.3.1.1 General issues concerning drinking water legislation 906.3.1.2 Changes in the DWD 98/83/EC 906.3.1.3 Need for guidance, research and working groups 916.3.1.4 Specific issues 926.3.2 Microbiological parameters 946.3.2.1 General issues concerning drinking water legislation 946.3.2.2 Changes in the DWD 98/83/EC 956.3.2.3 Need for guidance, research and working groups 956.3.2.4 Specific issues 966.3.3 Endocrine Disrupting Compunds (EDC’s) 976.3.3.1 General issues concerning drinking water legislation 976.3.3.2 Changes in the DWD 98/83/EC 976.3.3.3 Need for guidance, research and working groups 986.3.3.4 Specific issues 986.3.4 Risk management approach in drinking water 996.3.4.1 General issues concerning drinking water legislation 996.3.4.2 Changes in the DWD 98/83/EC 996.3.4.3 Need for guidance, research and working groups 1006.3.4.4 Specific issues 1016.3.5 Substances and materials used in the preparation or distribution of water intended

for human consumption 1026.3.5.1 General issues concerning drinking water legislation 1026.3.5.2 Changes in the DWD 98/83/EC 1026.3.5.3 Need for guidance, research and working groups 1036.3.5.4 Specific issues 104

6.4 Concluding remarks Pierre Hecq Desk Officer DGENV B1 Drinking Water Directive105


1 Introduction and objective of the seminar

1.1 Introduction

The Council Directive on the Quality of Water intended for HumanConsumption (DWD 98/83/EC) came into force in December 1998. The newDWD replaced the previous Directive from 1980 (DWD 80/778/EEC).

A major difference between the old DWD and the new one is the increasedtransparency and the fact that the number of parameters in the Directive hassignificantly decreased. This is in line with the Treaty of the European Union,and more specifically with the principle of subsidiarity, leaving moreresponsibility at the Member State level and restricting European legislationto essential and health-related parameters.

Other important aspects of the DWD are the precautionary principle, offeringthe highest level of protection to European consumers of drinking water andthe involvement of all stakeholders, with the emphasis on the involvement ofthe public. Involvement of stakeholders and the public is restricted toproviding information to the public on the quality of water and the measurescitizens can take to further decrease any adverse effects of drinking water,such as in the case of lead plumbing inside buildings

1.2 The revision of the Directive

The revision of the DWD is foreseen in the Directive, where Article 11requires the Commission to review Annex I of the Directive at least every fiveyears, in the light of scientific and technical progress, and make proposals foramendments where necessary. The previous Directive was initiated with alarge-scale consultation of all stakeholders in drinking water supply. Thisconsultation took place in the form of a Drinking Water Seminar organisedand hosted by the Commission in Brussels in 1993.

Now, ten years later, the Commission organised a Drinking Water Seminarwere all stakeholders had the possibility to give their views and input in theupcoming revision of Directive 98/83/EC. Apart from any progress in thetechnical and scientific sense, there is one important difference from theprevious consultation in 1993, i.e. the inclusion of stakeholders from the newcandidate countries. This potentially increased the number of countriesinvolved from the current 15 to a total of 25.

The new countries were expected to bring to the seminar not only a variety ofnew problems but also a world of experience and knowledge on drinkingwater supply in general. Even though this made the organisation moredifficult from a logistics point of view, the input of new and different


problems and knowledge was thought to be very interesting and a challengeto be dealt with in future regulation.

1.3 General approach

The seminar on drinking water was based on preparatory work done by anumber of selected European experts. These experts worked in fivepreparatory groups and were responsible for the production of positionpapers on identified subjects. The position papers were then distributed to allparticipants in the seminar and put on the website for every stakeholder andinterested party to read and comment on. The position papers were discussedduring the seminar in panel sessions on the same five themes. After thesepanel sessions the combined outcome of the position papers and thediscussions in the different thematic panel sessions were presented anddiscussed in the plenary meeting on day 2.

The input received through the website was not taken into account during theseminar but is processed in order to be combined with the outcome of theseminar in this scientific synthesis report.


2 Preparation of the seminar

2.1 Prepatory groups and prepatory documents

For the preparation of the seminar 5 groups were started on specific issuesidentified by the Commission. For these five preparatory groups 5 to 6experts were selected from European countries and invited to participate on apersonal basis. The objective of all groups of experts was to prepare a positionpaper on the various issues that was to be presented and discussed during thedrinking water seminar. These papers were also published on the dedicatedweb-site prior to the seminar for all members of the public (either individualsor groups/institutions to comment on. The five identified issues concerned:

o Chemical parameterso Microbiological parameterso Endocrine disrupting compounds (EDC’s)o Risk approach for drinking watero Materials and substances in contact with drinking water.

The preparatory groups on chemical and on microbiological parametersworked on the basis of position papers prepared by three expert workinggroups of the WEKNOW network (Web-based European KnowledgeNetwork On Water): papers from EWG chemistry, EWG microbiology andEWF sampling and monitoring and quality control. The preparatory group onEDC’ s worked on the basis of report produced by the Fraunhofer institute onthe study into EDC’s and drinking water, the group on risk approach usedvarious recent documents on the subject such as the Bonn paper and theWHO documents, the group on materials and substances in contact withdrinking water worked on the basis of the EAS (European AcceptanceScheme) on paper.

There was a close connection between de Drinking Water Seminar and theannual WEKNOW research conference in Kuopio, Finland that preceeded theseminar by two weeks. Also see the figure below.


WEKNOW and the DWS

WEKNOW EWG Position Papers

DWS Preparatory Groups

POSITION PAPERS DW Seminar

EAS ON PAPER

Various papersOn Risk Approach (WHO,É )

EDC Study Report

DWS WEB-SITE

WEKNOW Conference

Recommendationsto European Commission

Proposal from ECWEKNOW


2.2 Position papers (original papers as they were put on the web-site)

2.2.1 Position paper on Chemical parameters

POSITION PAPER OF PREPARATORY GROUP 1 “CHEMICALPARAMETERS

TONY LLOYD (CHAIRMAN), FRANTISEK KOZISEK (RAPPORTEUR), HELLEWESTPHAL, PIERRE LEROY, JAROSLAV SLOBODNIK, ADRIANA HULSMANN

BACKGROUNDFor the preparation of the position papers for the drinking water seminar fivegroups of experts have been formed. The various groups each focus on aspecific subject of the Drinking Water Directive (DWD), which are:

• Chemical parameters• Microbiological parameters• Endocrine disrupting substances• Risk approach for drinking water• Substances and materials in contact with drinking water.

Because of the co-existence of the other groups the Group on ChemicalParameters (the Group) restricted its activities to avoid duplication and doesnot or only briefly touches upon, endocrine disrupting substances andsubstances and materials in contact with drinking water. On the other handthe Group could not ignore the issues discussed in other groups, especially inthe group on risk approach. This group therefore assumed that a shift fromend of pipe control to process control would need to be considered by the EU.

The Group hereby presents its position paper with recommendations to theEuropean Commission (EC). The paper represents the views of its membersand not the view of the EC or the national view, although members’ ownpositions are naturally influenced by the national knowledge background.Other main sources of information used in the preparation of this paperincluded:

• WEKNOW1 position paper on chemical water quality• WEKNOW position paper on sampling and monitoring• WHO Guidelines for drinking water quality (2003 draft)• Summary Bonn workshop October 2001 “A framework for assuring

the quality of drinking water in the 21st century”• Drinking water regulations in some EU Member States and in the

USA

1 WEKNOW (Web-based European Knowledge Network on Water) is a concerted action ondrinking water supported by the European Commission under the 5th Framework Programme(contract No: EVK1-CT-2002-20004). See www.weknow-waternetwork.com.


• Blue-green algae and their toxins – Great Britain perspective (H.James& A.Lloyd)

• Directive of the European Parliament and of the Council 2000/60/ECestablishing a framework for community action in the field of waterpolicy” (EU Water Framework Directive; WFD) and Decision No2455/2001/EC (List of priority substances).

CRITERIA TO AID DECISIONS ON CHANGES TO CHEMICALPARAMETERS IN THE DRINKING WATER DIRECTIVE

The Group decided that a systematic approach is required when consideringregulatory changes and has proposed the criteria below. The Group alsorecommends these criteria be further developed and agreed upon within theEU Member States to obtain a standard tool or guideline for consideringpossible changes.

Criteria to aid decisions on changes to chemical parametersThree potential changes have been considered: setting a parametric value for achemical that is not in the Directive; deleting a parameter currently includedin the Directive; changing the parametric value for a Directive parameter. Thefollowing criteria were agreed to provide a useful checklist when consideringwhether any of the three preceding actions are justified.

1. Is there evidence for a risk to health via drinking water? Or, is thereevidence for a risk of deterioration in the aesthetic quality of water(taste, odour, appearance)?

2. Is the chemical present in water at concentrations high enough to causea risk to health, or deterioration in aesthetic quality?

3. Is the chemical frequently present in water sources, is it generatedduring the water treatment process, or is it released in the distributionsystem?

4. Can the chemical be removed, or reduced in concentration, bycurrently available treatment processes?

5. Are sufficiently sensitive analytical methods available to monitor thechemical at concentrations that cause a risk to health?

6. Can the risk to health, or deterioration in aesthetic quality, be reducedor eliminated by other means than setting a drinking water standard?

7. Does the proposed parametric value, or change in parametric valuehave a significant socio-economic impact?

8. Does the chemical have any effect on consumer perception of waterquality (taste, odour or appearance) or safety?

9. Is the chemical associated with media, consumer or political concerns?

NEW PARAMETERS CONSIDERED FOR INCLUSION

Disinfection by-products other than included in the DWD

Trihalomethanes and bromate are neither the only nor the majorrepresentatives of disinfection by-products (DBPs) in drinking water after


chemical disinfection and consequently other DBPs should also be givenattention.

Apart from chlorine and ozone, other substances may also be used for waterdisinfection such as chlorine dioxide or chloramines. For these, the DBPsparameters in the DWD are not relevant. Using chlorine dioxide disinfection,the by-product, chlorite is invariably present in treated water. As chlorite isharmful to human health, it should be regularly monitored if ClO2 is used forwater treatment. Many countries have already included chlorite in theirregulations: e.g. Czech Republic, France, Switzerland, Germany, Italy, andUK. When chloramination is practiced, the Trihalomethanes (THM) andHaloacetic acid (HAA) DBPs are at relatively low levels. On the other hand,toxic halonitriles (e.g. cyanogen chloride), halonitromethanes (e.g.chloropicrin), N-nitrosodimethylamine ( NDMA), and possibly other nitrosocompounds can be produced in treated water. As some of these substancesare harmful to human health, they should also be regularly monitored, ifchloramination is used for water treatment.

THMs have hitherto been considered to be representative of the DBPsassociated with water chlorination and therefore, if controlled as such, thisshould mean effective control of all chlorination by-products (CBPs).Nevertheless, this may not be true for raw water with a low pH value and ahigh total organic carbon (TOC) level. Chlorination of such water gives rise toCPBs other than THMs e.g., HAAs are regulated in the USA and othercountries. In this case, such contaminants should be the subject of control andan appropriate health-based limit value should be set up for them.

In the case of chloroacetic acid, it is a man-made pesticide as well as a CBP.In most cases it is not introduced in drinking water from “outside” as apesticide, but is generated during chlorination – usually at concentrationssignificantly above the 0.1 µg/l limit value for individual pesticides.

Cyanotoxins

Cyanotoxins can be generated in eutrophic surface waters under algal bloomconditions. There is evidence that many of them are dangerous to humanhealth. Their presence in treated drinking water is influenced not only by thelevel of cyanobacterial proliferation in raw water but also by the adequacy ofthe treatment technology used. In the case of inadequate treatmentcyanotoxins can be present at concentrations high enough to influenceaesthetic quality of water or even to cause a risk to health. Some countries(France, Spain) have already included the cyanotoxin, microcystin LR, in theirregulations.

Although cyanotoxins meet most criteria for inclusion on the DWD list ofparameters and require measures to be taken for their control, any attempt tocontrol the risk by routine monitoring for toxins seems not to be practicable.In respect of regulatory standards, insufficient data is available forrepresentative selection of the toxins and for determination of health-based


guidelines (with exception of microcystins). Analytical methods arecomplicated, very expensive and not sufficiently robust for routinemonitoring purposes. In addition there are almost certainly a number of asyet unidentified toxins.

A preventative approach should be preferred, starting with control ofnutrient input under the Water Framework Directive (WFD), application ofrisk assessment on the potential for cyanobacteria loadings, and provision ofadequate water treatment.

Calcium and magnesium

The results of numerous epidemiological studies have associated theconsumption of soft water with low calcium and, in particular, lowmagnesium content with increased rates of cardiovascular andcerebrovascular diseases. A certain minimum content of these elements indrinking water is therefore crucial from both the technical (water corrosivity)and public health points of view. The Group recommends reinstatement ofthe minimum hardness requirements, when water is softened or desalinated.This is especially an issue when communal softening of water is applied inpublic buildings and apartment blocks i.e. where consumers are exposedinvoluntarily to softened water.

Uranium

Evidence on chemical toxicity of uranium, which is naturally present in someunderground waters has emerged recently. Since the limit for uranium indrinking water based on its chemical toxicity (9 µg/l as suggested by theWHO Guidelines) is much lower than that based on its radioactivity, thelatter offers no protection against uranium chemical toxicity. In someEuropean regions, uranium is naturally present in underground drinkingwaters at concentrations substantially exceeding that recommended by theWHO. Nevertheless, insufficient data have been available to map thefrequency of this phenomenon within the EU countries and consequently, it isnot possible to consider the extent of the problem and whether inclusion ofuranium on the list of the DWD parameters would be justified or not.

PARAMETERS RECOMMENDED FOR DELETION FROM THEDIRECTIVE

Cyanide and mercury

Based on extensive monitoring data, it can be concluded that cyanide andmercury is currently almost never found in treated drinking water in mostEuropean countries and therefore, its widespread routine monitoring is notjustified.

Sulphate and chloride


These naturally occurring substances are common in water sources atconcentrations up to 100 mg/l; higher concentrations are rare. If present inwater at normal concentrations, these substances do not pose a risk to healthand routine monitoring does not appear to be justified.

Selenium

Currently, selenium as a pollutant of anthropogenic origin is almost neverfound in treated drinking water in Europe, its natural levels are well belowthe parametric value, and therefore, its widespread routine monitoring is notjustified either.

DIRECTIVE PARAMETRIC VALUES RECOMMENDED FOR REVISION

Pesticides: health-based limits

The limit value for individual pesticide (0.1 µg/l) in the DWD is not a health-based value (with exception of aldrin, dieldrin, heptachlor and heptachlorepoxide with parametric values of 0.030 µg/l). 0.1 µg/l is an administrativelyestablished value based on the assumption that drinking water should notcontain pesticides. However, with hundreds of licensed pesticides in use andwith many released into the environment without proper control andconsideration for water quality, a requirement for the absence of pesticides indrinking water is very difficult to achieve. Effective regulatory measures tolimit contamination of drinking water sources by these substances does notappear practicable for all biocidal products. In this case, it would bereasonable to consider whether to replace the unique limit value for allpesticides with health-based limits for individual pesticides. On the one hand,this would result in relaxation of the limit values for many pesticides, and onthe other hand, it would be possible to extend the list currently containingfour substances with the health-based limit values below 0.1 µg/l.

There is also the issue of the “polluter pays” principle. The current DWD limiteffectively transfers the cost and responsibility for pollution control from thepesticide producers and users of pesticides to the water industry and hence tothe water consumer.

Copper

In the light of recent toxicological studies in which the LOAEL or NOAELvalues ranged between 2 and 4 mg Cu/l, it can be stated that in the interimWHO guideline value, used as a basis for the parametric value (2 mg/l) in theDWD, practically a zero factor of safety has been applied, and therefore aquestion arises whether there is an adequate level of protection for sensitivesub-groups. Some countries transposing the DWD have imposed a morestringent limit value for copper (1 mg/l), e.g. Italy, Czech Republic, Slovakia,and Norway.


The EC should take advantage of the intended DWD revision and shouldreconsider whether this approach is compliant with the DWD principles.Apart from the assessment of the parametric limit value, the “adequatesampling method” for copper is to be defined and the “peak levels that maycause adverse effect on human health” (see Directive, annex I, note 3concerning the parameter copper) should be specified, as the new WHOGuidelines Draft states that “a single glass of tap-water with a concentrationgreater than 3 mg/l copper is more likely to elicit nausea than a litre of watercontaining the same amount (mass) of copper, but ingested episodically throughout aday”.

Nevertheless it may be more appropriate to protect consumers by restrictionson the use of copper, rather than by setting a stricter standard. The risk ofhigh copper levels is present only in water sources where the composition ofthe water is especially corrosive towards copper. Germany has alreadyintroduced restrictions on the use of copper in such areas. Setting a stricterstandard for copper is effectively “shutting the stable door after the horse hasbolted”. A pro-active approach by Member States to identify high-risk areasand restrict the use of copper would provide more certain protection forconsumers.

Fluoride: stricter value for artificially fluoridation

The current limit value (1.5 mg/l) for fluoride seems to be fairly safe in termsof skeletal fluorosis and other serious health effects. As the limit value hasbeen calculated on daily water consumption of 2 l, climatic conditions andpossible higher levels of water intake are to be taken into account at thenational (local) level.

However, a cosmetic effect called dental fluorosis (mottling of teeth) isobserved in sensitive sub-group of the population, even if water containsfluoride at a concentration of about 1 mg/l, i.e. below the current limit value.As naturally occurring fluoride is difficult to remove and the effect observedhas probably no health consequences, the limit value of 1.5 mg/l (for naturalfluoride) seems to be reasonable.

Nevertheless, the lower value (1.0 mg/l) recommended for artificiallyfluoridated drinking water (supplemented with fluoride to prevent caries)could also be justified as a compromise between caries prevention and dentalfluorosis prevention.

Antimony

At present anthropogenic contamination of underground sources of drinkingwater with antimony is extremely rare. If found at higher concentrations,antimony is usually of natural (geological) origin. If the limit value of 5 µg/lis exceeded, special treatment is required for the groundwater whichotherwise would not need any chemical/physical treatment. As the recentWHO revision is likely to result in a less strict guideline value (20 µg/l), it is


recommended to adopt this value in the DWD to avoid unnecessarytreatment of groundwater sources with otherwise good water quality.

The parametric values of other metals of mostly natural (geological) originshould also be kept under review to ensure consistency with the latesttoxicological knowledge.

As a general issue, the Group considered that all limit values set in the DWDshould be set on a transparent basis. It is desirable to prepare and makepublicly available a background document that would include toxicologicaldata used and describe how the limits were calculated. A clear statement oneach parameter/indicator should be given: whether the limit is health-based(and what safety factor has been used), or is based on political decision orpolitical compromise (with justification), or is an aesthetic- or operational-based limit value.

HARMONISATION OF THE DWD AND THE WFD

With regards to the foreseen harmonisation of the DWD and the WFD,extensive studies of the Expert Group on Analysis and Monitoring of PrioritySubstances (AMPS) working under the Expert Advisory Forum on PrioritySubstances and Pollution Control and their recommendations towards settingup new drinking water quality standards should be taken into consideration.

CHANGING TO SAMPLING AND MONITORING REQUIREMENTS

Despite the requirement of the DWD for samples to be taken “from taps thatare normally used for human consumption” there is little consistencybetween Member States in the selection of sampling locations and samplingpoints. Some Members States use random selection of samples, others usedesignated fixed points. Some take samples from consumers’ premises, othersmonitor fixed points in public buildings. In order to comply with the DWDsampling sites should be selected so as to be representative of the wholesupply zone and should include all properties supplied, i.e. both public andprivate buildings.Furthermore, the sampling procedure itself varies between differentEuropean countries (e.g. in time of flushing). This may compromisecomparability of results on drinking water at the international level. It isrecommended that guidance be provided on a uniform sampling procedurefor each parameter or group of parameters.

The sampling procedure has to be part of a quality assurance (QA) system,whereby sampling includes necessarily transport and storage of samples. Alink between the sampling protocol and the QA documents of the analyticalstandard methods would be helpful.

The monitoring requirements the DWD should be flexible and allow themember states and their health authorities to decide, based on available dataand risk assessment (as well as economical aspects and the responsibilities on


a national level), whether all parameters and indicators need to be monitored(and with what frequency) in a respective supply zone or not – i.e. not only tocut back the frequency to 50 % as now possible. However, any authorisationthat permits the analysis of parameters to be reduced or terminated should beavailable to the consumer public along with its justification. As this problem-related approach would yield non-harmonised data sets, a data evaluationstrategy and local data reporting have to be developed.

A good example of parameters which, after consideration of the localconditions, could be monitored with a lower or zero frequency, aresubstances of natural (geological) origin such as antimony, arsenic, boron andselenium, where anthropogenic contamination is very unusual. The samereasoning should apply to when new parameters are considered for inclusionin the Directive e.g. uranium.

Other parameters such as bromate should be monitored under specifiedcondition only. Occurrence of bromate in drinking water in detectableamounts results from water treatment/disinfection with ozone or sodiumhypochlorite, or during the electrolytic generation of chlorine by electrolysisof brine. Bromate need only be regularly monitored under these conditions.

On the other hand, the risk-based approach should involve a pro-activepolicy on the monitoring of new pollutants, which are not included in theDWD and thus cannot be monitored on either routine or screening basis indrinking water although they may possibly be present.

The monitoring of pesticides in drinking water can be used as an example of aproblematic issue of the current Directive. Pesticides are to be monitoredaccording to the following requirement: “Only those pesticides which are likely tobe present in a given supply need to be monitored” (DWD, Annex I, Part B, Note6), but reliable and accurate information on their local use in a catchment areaof a given supply zone is not available to most water producers. Furthermore,information on their relevant transformation products and adequatelysensitive analytical methods for their detection is often unavailable. Toresolve this problem it is necessary that:

a) The pesticide manufacturers be compelled to meet their obligationsimposed by Directive 91/414/EEC, i.e. to provide all required data onbreakdown products and analytical methods to be usable also for theDWD purposes and to make them available to all interested parties.

b) Local or river-basin databases or registries have to be established tocollect the data on the applied pesticides (active ingredient, area andtime of application, applied quantity); the data have to be regularlyupdated and accessible to all interested parties.

c) The obligation of drinking water producers to be informed – i.e. tostrive actively to get such information – on pesticides “which are likelyto be present” in their water sources should be enforced by law. Ofcourse, such a duty could be feasible only under the condition of anexistence of the above-mentioned databases (registries) as well as of alegal responsibility of all other parties involved.


d) The attention should be focused on the heavily used modern polarpesticides and their toxic degradation products. It is recommended tofollow provisions of the WFD when drafting the DWD list ofpesticides and include into monitoring programmes as a minimumthose selected for drinking water abstraction areas at national level.Pesticides present on the list of WFD Priority Substances (Decision No2455/2001/EC; alachlor, atrazine, chlorfenvinphos, chlorpyriphos(ethyl-, methyl-), diuron, endosulfan, gamma-hexachlorocyclohexane(lindane), isoproturon, simazine, trifluralin) should be included in allmonitoring screening programmes as “specific”.

An active approach to detection of new risk pollutants in drinking water should not be limited topesticides alone. Drinking water producers should carry out a risk analysis for possible sources ofpollution to their raw water. They should take account also of lists of river (sub)basin priority substances,to be created under WFD for each European river basin and ground water body. The legally binding WFDlist of Priority Substances includes 33 substances, which must be monitored in each European river basin.The list will be extended for basin-specific pollutants and regularly updated every four years.

Regarding heavy metals, WFD requires monitoring of not only the metal itselfbut also the organic form in which is the metal present in the aquaticenvironment (speciation analysis). Therefore the metals are listed as cadmiumand its compounds, lead and its compounds, mercury and its compounds,nickel and its compounds. The above requirements are based on the health-risk assessment studies and, therefore, a similar approach might beconsidered by DWD for those metals where health-based limit of organicform would be stricter than currentlimit in DWD.

ANALYTICAL REQUIREMENTS (DWD – Annex III)

Performance characteristics of analytical methods should be updated to be inharmony with both terminology (e.g. definitions used, trueness versusaccuracy, etc.) and common practice of accredited laboratories complyingwith standards ISO 17025 (e.g. to replace the limit of detection by the limit ofquantification).Additional performance characteristics for on-line monitoring methods neednot to be included in the DWD current revision as this type of monitoring hasbeen used up to now in drinking water production only for checking theoperational performance of the treatment process or raw water quality(conductivity, turbidity, pH, chlorine, rarely also fluoride and nitrate) and notfor compliance monitoring of water supplied to consumers.

WATER TREATMENT CHEMICALS

One of the sources of drinking water pollution may be water treatmentchemicals. These chemicals will not be covered by the proposed EuropeanAcceptance Scheme (EAS) for construction products in contact with drinkingwater (CPDW). Their use is however controlled by the European standardsmandated by the European Commission under the Public ProcurementDirective (Council Directive 93/38/EEC of 14 June 1993 coordinating theprocurement procedures of entities operating in the water, energy, transport


and telecommunications sectors). According to this Directive drinking waterproducers/suppliers have to use only the chemicals that conform with givenstandards. About 100 standards2 have now been published by CEN and theseinclude all of the chemicals commonly used in Europe, as well as the mostcommon traditional filter materials.The standards define an acceptable level of purity and include maximumconcentration levels for the toxic metals specified in the 1980 Drinking WaterDirective. The Standards also specify maximum concentrations for otherparameters or characteristics specific to the product in question. CEN hasrecently reviewed its published standards to check for consistency with therequirements of the 1998 Drinking Water Directive (DWD).These standards do not contain attestation requirements i.e. the standards donot include a mandatory requirement for checking the conformity of aproduct’s composition with the requirements of the standard. It is understoodthat the water supplier nevertheless makes checks on composition, or legallybinding requirements for conformity of composition are included in purchasecontracts.Mandatory dosing concentrations for chemicals are not included in thenormative sections of the standards but guidance on dosing and othersubjects is contained in the Informative Annex of each standard. SomeMember States regulate the dosing concentrations of water treatmentchemicals in their national approval schemes.Member States may also impose technical requirements that are stricter thanthose contained in the European standard i.e. a requirement that has not beenaccommodated within the standard by application of different classes andthat could constitute a barrier to placing a product on the market. In this casethe member state must make a notification to the European Commissionunder Directive 98/34/EC. The requirement may be maintained if theEuropean Commission does not raise objections, following its consultation onthe requirement with all Member States. However, this approach wouldnormally only be used if the review process for standards was not able tokeep up with emerging toxicological data.

On this basis, the Group concluded that where a drinking water contaminantoriginates exclusively from water treatment chemicals, e.g. chloropropanolisomers derived from the use of polyamine flocculants, it is not necessary tolist it as an individual parameter in the DWD, but it is sufficient only to referto the European control procedures explained above – the current CENstandards for these water treatment chemicals ensure that the limits are notexceeded.If any drinking water contaminant may originate from both water treatmentchemicals and materials in contact with water (e.g. epichlorohydrin fromwater treatment polyamines and from the use of certain polymeric coatings,or acrylamide from water treatment polyacrylamides and grouting agents),the most effective approach to its control is again via the use of CENstandards or materials approved under the EAS.

2 “Chemicals used for treatment of water intended for human consumption” – seehttp://www.cenorm.be/catweb/71.100.80.htm.


Only if any drinking water contaminant may also originate from othersources than treatment chemicals and contact materials, it is justified to list itas an individual parameter in the DWD, provided the criteria suggested inthe introduction of this position paper are met.

NEW APPROACHES TO DRINKING WATER QUALITY

In view of the increasing number of man-made chemicals released into theenvironment, their possible occurrence in drinking water, unclear problem ofa potential synergic effect of chemical mixtures, and the practicalimpossibility to monitor routinely hundreds of hazard substances, it isnecessary to consider the use of other complementary methods (e.g. variousbiological tests) for testing water safety, and the use of such qualitymanagement systems which would prevent or minimize the entry of foreignsubstances into drinking water.When considering risks related to different parameters and in particular theendless list of threats from new chemicals, it is to be concluded that the onlyeffective way how to control drinking water quality is to address not only thefinal product but also the process of drinking water production as a whole:this means to identify all risk points starting from the source of raw water,including its protection zones, through water treatment and distribution tothe domestic distribution system of the consumer’s premise. It wouldcontribute not only to prevention of new risks from water contamination butalso to rationalization of the monitoring programme.

CONCLUSIONS AND RECOMMENDATIONS

The Group makes the following general recommendations in respect of the scope andcontent of the Drinking Water Directive:

The Group recommends the use of the criteria it has developed to aiddecisions on possible changes to the DWD

A clear statement should be developed for all parameters, setting out thebasis of the decision to regulate a parameter and the rationale for setting theparametric value

The DWD should require Members States to follow a consistent approach,both in the specification of sampling points and in their procedures for takingsamples.

The DWD should allow greater flexibility in setting monitoring requirementsand allow Members States to go beyond the currently permitted 50%reduction in sampling frequencies, if this can be justified.

The analytical and performance characteristics terminology of the DWDshould be brought into line with modern practice.


The application of risk assessment in the control of catchments and watertreatment processes has the potential to reduce requirements for “end ofpipe” monitoring. As a specific example of the application of risk assessment,Member States should give consideration to the priority substancesdesignated under the Water Framework Directive when assessing risks inspecific catchment areas.

The Group made the following recommendations in respect of specific parameters:

New DBP parameters should be considered e.g. haloacetic acids (ifchlorination of raw water with a low pH value and a high TOC level givesrise to CPBs other than THMs) and chlorite (if chlorine dioxide is used) andcyanogen chloride (if chloramination is used)

Cyanotoxins should be controlled by limiting nutrient inputs to surfacewaters and installing appropriate drinking water treatment

A restriction on the minimum hardness in softened or desalinated watershould be reinstated

A toxicologically based standard should be introduced for uranium

The cyanide, mercury, sulphate, chloride and selenium parameters should bedeleted from the DWD

Consideration should be given to setting specific health based standards forindividual pesticides

Sensitive sub-groups may need protection in areas where water is aggressiveto copper. This is best achieved through the introduction of restrictions on theuse of copper by Member States.

A stricter standard of 1 mg/l should be introduced for fluoride, whereartificial fluoridation is practised

The standard for antimony should be relaxed to the 20 µg/l guideline valuenow proposed by WHO.


2.2.2 Position paper on Microbiology

POSITION PAPER OF THE PREPARATORY GROUP ONMICROBIOLOGY

IntroductionThe group met in Brussels on 5 September 2003. Membership is listed inappendix 1. Its remit was to identify current and future problems and makerecommendations on how they should be addressed in the EuropeanDirective on the quality of water intended for human consumption (DWD).

The group identified 6 main topics for its discussion:o comparability of microbiological analytical methodso risk management approach to microbiological safetyo molecular methodso samplingo established microbiological parameterso new microbiological parameters

To assist in improving the discussion the group considered a number ofrecent documents. These are listed in appendix 2.The group has intentionally kept this report short and, it hopes, to the pointand easy to read. Some statements will be considered controversial, but thegroup hopes that this report will stimulate lively and informed debate.

Comparability of microbiological analytical methodsThe group endorsed the conclusions and recommendations of the EuropeanMicrobiology Advisory Group (EMAG) at its meeting in Ranco, Italy, April2003. However, it considered that equivalence testing should not be restrictedto single Member States (MS); water type and quality are more importantthan geographical boundaries.Acceptance of a method by a MS does not exempt individual laboratoriesfrom the need to demonstrate method performance under their owncircumstances.ISO 17994 is the only acceptable methodology at present, but it is complicatedand does not fully define “equivalence”, “deviation” or “at least as reliableas” in terms of the DWD.More research is required on method equivalence; this may be an appropriatetask for EMAG.

Statements on comparability of microbiological analytical methods:Pan-European comparisons for equivalence testing are the most indicativeway for both established and new methods (as part of the validation),therefore a Pan-European research effort is recommended.


Pan-European equivalence testing will allow identification of theperformance of each method with regard to the different water types andwater quality.Under the terms of the DWD, a Member State can use the research done onequivalence testing, but still need to validate methods in individuallaboratories.

Risk management approach to microbiological safetyThe group acknowledged that the topic of risk assessment is the subject ofconsideration by another group but wished to endorse this concept as anappropriate additional way forward in the protection of public health.Water safety plans and risk assessment should not be seen as cost-savingalternatives for water quality monitoring or reason to cut down on samplingfrequencies; they should be used to target such programmes.Risk assessment may overlook problems not identified by current knowledgeand offers little protection against fraud.Consumers will need evidence that a risk assessment approach is anapproach towards water safety.Further advice, ideally perhaps a manual, on how to carry out amicrobiological risk approach is required.

Statements on risk management approach to microbiological safetyWe recommend the endorsement of the WHO approach on water safetyplans.The risk management approach is considered as an important step in theright direction. We raise two more aspects:We risk limiting ourselves to what we already know, and could overlook therisks of newly emerging quality threats (how can the risk approach preparefor the unexpected?).The risk of misuse of data is not addressed in the risk approach; however, itmay cause the biggest problem in implementation.It is recommended that bio films as a source of microbial contamination areincluded in the risk approach.The risk management approach should not be viewed as a full replacementfor compliance and check monitoring.We identify that the risk management approach will lead to an explanationgap for the consumer.

Molecular methodsThe group endorses the development of new and rapid methods for themicrobiological analysis of water.The information and results provided by such methods may require a newregulatory approach but this should not be a limit to their development.

Statements on molecular methodsNew molecular methods have a high potential, in particular for detectinghygienically relevant micro-organisms which are not currently covered byroutine measurements.


At present molecular methods cannot be standardised because theirdevelopment is faster than the standardisation process.Care is needed in the interpretation of data gained with these methods. Thereis a need for more research on the correlation of molecular methods withrespect to viability and virulence.Molecular methods must be integrated into the equivalence testing system.A group of specialists should be established to assess the present status andthe development potential of various molecular biological methods in orderto construct a tool box to indicate the suitability of the various methods forparticular applications. It is recommended that the extent and resources ofEMAG be extended to take up this task.Numerical standards are not necessarily appropriate for molecular methods(+/- information may be sufficient provided the volume is indicated).

SamplingThe group considered that all sampling for microbiological analysis should becarried out in accordance with ISO 19458, once this is finalised and published.Sampling points should be selected on the basis of risk assessment to provideinformation on the contamination potential of the water.The issue of flushing or not flushing taps prior to sampling should be left toMS to specify according to the purpose of sampling.The Directive minimum frequency of sampling and analysis for coliformbacteria and Escherichia coli for water supplied from a distribution network istoo low to provide meaningful data to protect public health or detectcontamination on time.For example, for E. coli the requirements are:for a supply of 999 m3/day 5 samples per year (4 check plus 1 audit);for a supply of 9.999 m3/day 38 samples per year (34 check plus 4 audit);for a supply of 99.999 m3/day 317 samples per year (304 check plus 13 audit).Results from such sampling are not statistically valid and should be adaptedto risk assessment. The group noted that some MS already specify highermonitoring requirements.

Statements on samplingAccept ISO CD 19458 when it is published.Sampling points must be selected with respect to the contamination potentialof the water.The frequency of microbiological analysis with respect to coliform bacteriaand E. coli is considered unreasonably low.

Established parametersThe group endorsed the EMAG recommendation on Clostridium perfringensbut suggested that consideration be given to a further option – return to theuse of sulphite-reducing Clostridia. There is a view that this is a more usefulgeneral microbial indicator of water quality and more reliable analyticalmethods are available.The group considered if the colony count parameter should be retained andconcluded that it should because it can provide useful information on water


quality through treatment and distribution. Its strength is in its use for trendanalysis and numerical values are not recommended.The group has no comments on other existing microbiological parameters.

Statements on established parametersIt is recommended that the inclusion of Clostridium perfringens as a checkparameter in the DWD is reconsidered with regard to its limited value as anindicator of public health protection and effectiveness of water treatment. Theoptions for consideration are:(i) removal from the list of microbiological parameters;(ii) audit monitoring frequency only for water that originates from or isinfluenced by surface water; or(iii) return to the use of sulphite-reducing ClostridiaThe group supports the retention of colony counts as a useful indicator ofwater quality.

New microbiological parametersThe group had wide ranging discussions on the value of including specificpathogens or potential indicators as parameters in the DWD. Topicsdiscussed included Legionella, Mycobacteria, Helicobacter, Pseudomonas,cyanobacteria, viruses and bacteriophage.They were not persuaded that there was potential for improved public healthprotection by the inclusion of any new parameters at present. However, thistopic should be kept under regular view and the development of new andrapid analytical techniques may make more specific monitoring feasible andbetter protect public health.Specifically on Legionella the group concluded that although monitoring atspecific locations is important it is not a relevant organism for inclusion in theDWD as a parameter.

Statements on new microbiological parametersNo health benefit can be recognised by the inclusion of specific pathogenicmicro-organisms as parameters in the DWD; risk assessment is a morevaluable tool in this respect and helps to prevent parameter hysteria.There is insufficient evidence that bacteriophage are suitable as indicators forthe presence of enteric viruses.The current use of indicators and inclusion of other micro-organisms asparameters should be kept under regular review, particularly in respect of thedevelopment of new, rapid analytical methods.Monitoring of Legionella is important but the inclusion of Legionella as aparameter in the DWD will not prevent outbreaks of legionellosis.


Members of the preparatory group on microbiology

Hans-Curt Flemming (chairman)David Drury (rapporteur)Pär AleljungLinda BaggeJean-Marie DelattreCorina de Hoogh (assistant rapporteur)

Documents considered by the group

A framework for assessing the quality of drinking water in the 21st century –Bonn workshop, October 2001

Minutes of the European Microbiology Advisory group (EMAG) meeting, 11April 2003, Ranco, Italy.

WEKNOW Expert Working Group on Microbiology position paper – draft2.7.03

WEKNOW Expert Working Group on Sampling and Monitoring positionpaper - draft 29.7.2003

EMAG Clostridium perfringens as a drinking water quality parameter – asummary paper (final draft 31.3-2003)


2.2.3 Position paper on Endocrine Disrupting Compounds

POSITION PAPER OF THE PREPARATORY GROUP ON ENDOCRINEDISRUPTING COMPOUNDS (EDCS)

Introduction

The group met in Brussels on the 19th of September 2003. Membership islisted in appendix 1. Its goal was to identify current and future problems andmake statements and recommendations on how EDCs should be addressed inthe European Directive on the quality of water intended for humanconsumption (DWD).Currently EDCs are not regulated in the DWD. No limits for individual EDCsor the total (potential) effects have been set in the DWD. For pesticides,polycyclic hydrocarbons (PACs), polychlorinated biphenyls (PCBs) andmetals, of which some are (suspected) EDCs, limit values are part of theDWD. However, these limit values have not been set based on endocrineeffects.The group identified the next main topics for discussion:

Exposure of humans to EDCs Environmental concern of EDCs Occurrence of EDCs in drinking water sources Occurrence of EDCs in drinking water Human health risk, priority EDCs Development of analytical methods and bioassays Removal of EDCs in water treatment Release of EDCs from materials in contact with drinking water

To assist in improving the discussion the group considered a number ofrecent documents. These are listed in appendix 2.The group has intentionally kept this report concise. Some statements may beconsidered controversial, but the group hopes that this paper will stimulatediscussions on this topic.

Exposure of humans to EDCs

In literature many groups of substances are described for their (potential)endocrine disrupting properties. There are strong differences in origin(natural, industrial), emission sources (manure, effluents of sewage treatmentplants, diffuse pollution of pesticides), chemical structure and behavior ofEDCs.Exposure routes are not the same for the different classes of EDCs. Humansare exposed mainly to phytoestrogens, like isoflavones, via food. Also forbioaccumulating compounds like PCBs, food is the main exposure route.Synthetic hormones, like 17α-ethinylestradiol are an exception, as drinkingwater is theoretically the main route for unintended exposure to these


compounds. However, we don’t expect these to be present in drinking waterin significant levels. Since 17α-ethinylestradiol is one of the most potentsynthetic estrogenic compounds unintended exposure must be avoided inany case.

Statement:Drinking water is not the main exposure route for most EDCs. Many EDCs,especially bio-accumulating compounds, are taken up by food (dairyproducts, meat, fish, …) in higher amounts on a daily basis.

Environmental concern of EDCs

According to our present knowledge, the effects of EDCs in the water cycleare mainly of environmental concern. The data that initiated this concernrelate to effects on wildlife like fish, amphibians and reptiles. However, goodraw water for the production of drinking water is needed as well. Ideally,production of drinking water should be achievable with only conventionaltreatment methods (for example aeration and filtration for groundwater).This need was taken into account in the Water Framework Directive(2000/60/CE), adopted by the European Parliament and the Council inSeptember 2000, which aims at improving the quality of water andecosystems in Europe. In accordance with article 16 of the Directive, a list of32 priority substances, selected among those which present a significant riskto or via the environment, was established. These substances will be subject tothe “cessation or phasing out of discharges, emissions and losses” by the year2020. EDCs such as nonylphenols, octylphenols, tributyltin compounds andseveral pesticides, are included in this list.

Statement:With regard to EDCs control in drinking water the Water FrameworkDirective will help as an “upstream” action. However, the effects of the WFDneed to be monitored with respect to the consequences for drinking waterquality.

Occurrence of EDCs in drinking water sources

Both natural and synthetic estrogenic substances commonly enter freshwatersystems through sewage treatment works effluents. Evidence for theestrogenic activity of such effluents comes from various European countriesincluding the UK.The compounds found to be responsible for the majority of the in vitroestrogenic activity of domestic wastewater treatment works effluents havebeen the natural estrogens, 17β-estradiol (E2) and estrone (E1) and thesynthetic E2 derivative17α-ethinylestradiol (EE2). The concentrations of thesesteroid estrogens that are present in wastewater treatment works effluentsdepend on the efficiency of their removal in the various wastewater treatmentprocesses. Generally removal rates are high, usually above 75% and often up


to 95%. However, even at low concentrations these compounds can beextremely potent. For example less than 1 ng/l EE2 can induce vitellogeninproduction in male rainbow trout (vitellogenin is an egg yolk protein usuallyonly found in adult females). Total concentrations of E1, E2 and EE2 reportedto be present in UK sewage effluents are generally in the range 1-10 ng/l,with E1 and E2 predominating. Concentrations reported in river waters aregenerally in the range 1-5 ng/l – again E1 and E2 are present at higherconcentrations than EE2.

Industrial compounds, such as the nonylphenols and nonylphenolethoxylates can also enter surface waters via wastewater treatment workseffluent. Although these compounds are much less potent estrogens than thesteroids (by a factor of about 105), their concentrations in such effluents can besignificantly higher, so their presence cannot be ignored. [The use ofnonylphenol ethoxylates is due to be phased out in the next few years, so thecontribution of these particular compounds to the estrogenicity of effluentsshould be considerably reduced].

Other weakly estrogenic compounds, such as some pesticides, can entergroundwaters due to agricultural usage. For example, atrazine and simazine(both of which are herbicides) have been detected in some groundwaters atconcentrations in the low µg/l range and occasionally at lower concentrationsin the derived drinking waters. Because, in comparison to 17β-estradiol, theirestrogenic potency is very low (at least 106 lower) these findings are notconsidered to present a risk to drinking water consumers.

Statement:Most of the European surface waters and some groundwaters arecontaminated with low levels of EDCs. With the presently applied sewagetreatment technologies EDCs cannot be fully avoided in surface watersreceiving sewage effluent. Without adequate drinking water treatment stepslow levels of EDCs have to be expected in drinking water.

Occurrence of EDCs in drinking water

So far, measurements have not indicated that significant levels of EDCs havebeen found in drinking water. Low concentrations in the ng/l range havebeen reported for some pesticides (atrazine, simazine, and metabolites), bis-phenol A, several phthalates and alkylphenolethoxylates. Most of thedrinking water samples are negative. Most studies did not detect natural orsynthetic hormones. However, relatively few data are available for only a fewcountries and much more representative monitoring in all Europeancountries is needed. The Commission will, in close cooperation with theMember States, establish monitoring programmes in water to estimateexposure to and effects of the substances on the priority list of endocrinedisruptors in order to gather evidence that could be used in future revisionsof different legislative instruments, such as the DWD and the WFD.


Statement:For an European wide assessment of the occurrence of EDC s in drinkingwater, more reliable data are required.

Human health risk, priority EDCs

There is not enough knowledge available to evaluate the potential health risksfrom exposure to low levels of EDCs via drinking water. For many EDCs littledata of the occurrence in the water cycle are available. Research on low doseeffects and combined effects is still going on.

As hormones are the most potent EDCs, they are at present the most relevantones for drinking water. But since we don’t know all the relevant EDCs, thereis still a need for screening of the total (potential) effect using bioassays. Atpresent, most bioassays are aimed at estrogenicity; more bioassays for otherendocrine effects (such as androgenicity) are needed.

Bioassays give an indication of the total potential endocrine effect of a sample,whereas chemical analysis gives information about individual compounds.Both techniques are complementary and are needed to study the relevance ofEDCs in the water cycle.

Statement:More toxicological data as well as more reliable drinking waterconcentrations are essential to evaluate the relevance of EDCs in drinkingwater. The contribution of drinking water to the exposure of humans seemsto be very low.

Development of analytical methods an bioassays

In order to be able to provide exposure data it is essential to have access torobust and validated analytical methods. In particular, the determination ofhormones as the most potent species in terms of endocrine disrupting effectsrequire detection limits of less than 1 or even 0.1 ng/L. Althoughsophisticated technologies (high resolution or tandem mass spectrometry)and advanced techniques (isotope dilution) are available to measurepollutants in traces, only few laboratories have the instrumental resourcesand the necessary experience to produce valid and comparable data that areneeded.Moreover, sample preparation procedures are mostly time consuming and formany compounds still no standard reference materials are available. Sincemany of the EDCs are part of our daily life they can be found ubiquitously.Hence, special procedures to avoid cross-contamination have to be taken intoaccount both for analytical and for sampling purposes.Chemical analysis determines concentrations of individual pollutants butcannot describe endocrine effects that may be provoked by the total content


of pollutants. Certain in-vitro tests are already available to measure theestrogenic of a sample but so far most of them are not sensitive enough todetermine effects of low concentrations that are expected in drinking water.Analytical measurements in the nanogram per litre range cannot be regardedas “routine work” and because of the lack of standardised methods; specialemphasis must be placed on additional quality assurance, e.g. proficiencytests for laboratories to prove the comparability of results.

Statements:Methods for monitoring individual EDCs and total (potential) effects areavailable on a limited scale. Implementation on a larger scale is needed tocarry out monitoring programmes. Special precautions for sampling methods,materials in contact with samples and pretreatment of samples must befollowed. Different laboratories can use different methods, as long as theperformance is shown to be satisfactory. However, implementation andvalidation of these methods in the (new) MS is probably a problem.Instrumentation is expensive, and technicians need to develop experiencewith these methods before reliable results can be achieved.

Removal of EDCs in water treatment

The processes used in drinking water treatment differ much according to thewater quality of the source, the endorsement of standards for drinking waterquality (WHO guidelines, national and European standards, the policy of theindividual water works on improved performance, and the willingness andability of the consumers to pay. Advanced treatment processes are ofteninstalled for disinfection of surface water intakes or for groundwatercontaminated with agrochemicals or industrial chemicals. Much research hasrecently been conducted to study the additional positive effects ofconventional and advanced drinking water treatment regarding the removalof EDCs. Ozonation, GAC, nanofiltration and reversed osmosis have provedto be very effective; coagulation, filtration and chlorination are less effective.

Statement:Advanced drinking water treatment is capable of removing EDCs at highrates.

Release of EDCs from materials in contact with drinking water

There is evidence to show that EDCs of low potency, such as organotincompounds, phthalates, alkylphenols and bis-phenol A can migrate fromcertain organic materials used in contact with water intended for humanconsumption. This is of potential concern as no action can be taken bydrinking water supply companies to remove them prior to consumption ofthe affected waters. Bottled waters can also become contaminated by EDCswhich migrate from the bottles or containers in which they are stored.


As there is a separate expert working group on substances and materials usedin the preparation and distribution of drinking water (Topic 5), there isobviously some overlap between this group and the EDCs group. The EDCsgroup consider that the potential for the introduction of EDCs into drinkingwaters from materials should be addressed by the European AcceptanceScheme (EAS). Therefore, brief details of the EAS are presented here.

In the near future, the EAS will be applied throughout the EU. This schemewill include the use of positive lists for substances used in the preparation ofmaterials to be used in contact with drinking water. The toxicology ofsubstances on these lists will have been assessed to ensure the substances aresafe to use and the body that carries out these assessments will need toconsider potential endocrine disrupting effects.

Statement:Evidence of release of EDCs from materials to drinking water / bottled wateris available. The release of EDCs from materials in contact with drinkingwater, such as for example organotin compounds, phthalates, bis-phenol Aand alkylphenols needs to be addressed in the EAS.

Recommendations

Currently EDCs are not regulated in the DWD. It is not recommended to setlimit values for individual EDCs in the DWD at this point, because:- several hundreds of substances have been identified as potential EDCs;

prioritization is still ongoing;- the knowledge to evaluate the potential human health risk is insufficient;- sufficient occurrence data for both raw and drinking water are not

available yet;- validated bioassays for screening the total potential effects of EDCs are

required.

Risk assessment and risk management approach as recommended by thethird edition of the WHO-guidelines is probably the best tool for controllingEDCs in drinking water. This approach was developed in the 1960s for thefood industry, and since then has extensively and successfully beenimplemented in this sector, in the form of HACCP (Hazard Analysis andCritical Control Points), as recommended by the Codex Alimentarius, and theDirective 93/43/CEE on the hygiene of foodstuffs. Its application to drinkingwater production in Australia and several European countries is recent. Asregards to EDCs in drinking water, risk factors (such as sewage and industrialeffluents) are known. Drinking water treatment performances for EDC-removal have been assessed. This allows for risk ranking of the drinkingwater treatment plants, and can allow for identification of those productionplants that need to be adapted to improve removal of EDCs.


More toxicological data and monitoring data are needed to evaluate the riskof human consumption of drinking water, especially at places with high riskfactors for pollution.

It is recommended that European laboratories participate in the setup of aproficiency testing scheme for EDCs.


Appendix 1: Members of the preparatory group on EDC’s

Phillipp Hohenblum, chairmanLeo Puijker, rapporteurAndrea WenzelJean-François LoretHuw JamesCorina de Hoogh, assistent rapporteur

Appendix 2: Documents considered by the group

A framework for assuring the quality of drinking water in the 21st century –Bonn workshop, October 2001.

Study on Endocrine disruptors in drinking water. Final ReportENV.D.1/ETU/2000/0083, by Andrea Wenzel, Josef Müller and ThomasTernes. Schmallenberg and Wiesbaden, Germany, February 2003.

Summary of presentations and discussions on EDC’s in drinking water at theIWA Leading Edge Technology Conference, 26-28 May 2003, Noordwijk, TheNetherlands.

Endocrine disruptors in Austria’s waters – a risk? by Renate Paumann andStefan Vetter; Federal Ministry of Agriculture, Forestry, Environment andWater Management, Vienna, Austria, July 2003.


2.2.4 Position paper on Risk Approach for Drinking water

RISK BASED APPROACHES:THE NEXT STEP IN THE EVOLUTION OF THE DRINKING WATERDIRECTIVEPrepared by (alphabetical order): Michel Gibert, Guy Howard, Adriana Hulsmann,Gertjan Medema, Oliver Schmoll, Frans Schulting and Riku Vahala

BACKGROUND

Risk management approaches in drinking waterHistorically, end-product testing has been one of the key elements withinlegislation for managing the delivery of safe drinking water, and ostensiblyfor the protection of public health. This is reflected by both the DrinkingWater Directive 98/83/EC (DWD) and its predecessor DWD 80/778 EEC,and many of the national drinking water legislations within Europe thatprimary rely on compliance monitoring against numeric water qualitystandards based on samples taken at specified minimum frequencies. Overrecent decades, this approach has given clear guidance and a great impetus tothe development of the water supply sector in European countries that hasresulted in a high level of drinking water quality and supply infrastructure.From the consumer’s perception this approach has successfully guaranteedhigh confidence in drinking water safety.While a necessary part of drinking water quality management it has,however, increasingly been recognised that reliance on end-product testinghas a number of serious limitations, particularly for microbial quality ofdrinking water. Specific limitations include:There is a limited relationship between individual pathogens potentiallypresent in drinking water and indicator organisms (e.g. E. coli) widely usedfor defining numerical standards within end-product quality based drinkingwater legislation. This is reflected by the fact that waterborne diseaseoutbreaks have occurred in the absence of E. coli. In particular, bacterialindicators have limited relationship to viral and protozoan pathogens (e.g.due to different sensitivity to disinfection) and the importance of thesepathogens is becoming increasingly apparent.Microbiological detection methods are often slow and thus have limitedcapability for early warning. If water quality monitoring provides evidence ofmicrobial contamination, in most cases, the water has already beendistributed and potentially consumed. End-product testing, therefore, oftenamounts to an “it’s already too late” approach.Water volumes tested against water quality standards are insignificantlysmall if compared to the total amount of water produced and samplenumbers are rarely statistically representative of the water being tested.Taking into account the nature of micro organisms in drinking water, thestatistic significance of the results of end-product monitoring is often limited.


Primary reliance on end-product testing is presently considered as notsufficient to provide confidence in good and safe drinking water. As a logicalstep in the evolution of an approach safeguarding the supply of safe drinkingwater that has the trust of consumers, discussions have started in the drinkingwater sector about the added value of risk assessment and risk managementstrategies analogous to the principles of the HACCP (Hazard Analysis andCritical Control Point) approach that is widely used in the food industry. Thismove is not surprising as water is considered as a food product in manycountries. Discussion forums took place at the international events AssuringDrinking Water Quality in the 21st Century (Bonn, October 2001) and WaterSafety: Risk Management Strategies for Drinking Water (Berlin, April 2003).

In Switzerland, Australia, New Zealand, France and the Netherlands, forexample, there has been a shift of emphasis within drinking water legislationtowards a preventative or quality assurance approach. It encompassescomprehensive risk assessment and quality management strategies thataccount for a holistic approach from catchment to consumer, and that moveaway from excessive reliance on end-product testing towards a preventativefocus in controlling processes and infrastructure. This is reflected by thelegislative approaches taken in Switzerland and Australia, which are brieflyoutlined in Box 1.

Box 1. Risk management in Swiss and Australian drinking water legislation.In Switzerland drinking water quality is dealt with under foodlegislation. Since 1995, the Hygiene Ordinance (Article 11) requirescontrol of microbiological hazards through the application of HACCPprinciples (though the term “HACCP” is not explicitly mentioned inthe Ordinance) from all water supplies:

a. “Identification and assessment of potential health risks (…);b. Determination of points, operations or technical steps in the production

process at which health risks can be eliminated or reduced (CriticalControl Points, CCPs);

c. Determination of critical limits and tolerance levels (CCP conditions)(…);

d. Establishment of a monitoring system that monitors compliance withCCP conditions;

e. Determination of corrective actions if monitoring detects non-compliancewith CCP conditions;

f. Determination of procedures for checking functioning of the controlsystem established (verification);

g. Documentation of measures according to letters a-f.”

In Australia a Framework for Management of Drinking Water Quality hasbeen developed for incorporation in the Australian Drinking WaterGuidelines. The Framework provides guidance for the design andimplementation of quality management systems that address the


specific requirements of drinking water supplies. The Frameworkaddresses four general areas:

• “Commitment to Drinking Water Quality Management – thisinvolves developing a commitment to drinking water qualitymanagement within the organisation. Adoption of the Frameworkwithout effective implementation and continual improvement is notsufficient. Successful implementation requires the active participation ofsenior executive and a supportive organisational philosophy.

• System Analysis and Management – this area involves understandingthe entire water supply system, the hazards and events that cancompromise drinking water quality, and the preventive measures andoperational control necessary for assuring safe and reliable drinkingwater.

• Supporting Requirements – this area involves elements needed toensure the capacity to operate and adapt a system to meet its challenges.This includes the basic elements of good practice such as employeetraining, community involvement, research and development, validationof process efficacy, and documentation and reporting systems.

• Review – this area includes regular evaluation of water quality data andaudit processes, and their review by senior executive to ensure thatmanagement system is functioning satisfactorily. These componentsprovide a basis for review and continual improvement.”

The third edition of WHO’s Guidelines for Drinking Water Quality (GDWQ)will also place great emphasis on risk assessment and managementstrategies as instrument for assuring drinking water quality and theprotection of public health. Within the GDWQ this type of approach istermed as Water Safety Plan (WSP). This WSP is intended to beimplemented by the water supplier, and it has three main components,which are guided by health-based targets and overseen through publichealth surveillance. Box 2 outlines the WHO Framework.

Box 2. WHO GDWQ’s Framework for Safe Drinking Water.

1. “Health based targets based on critical evaluation of health concerns.2. System assessment to determine whether the water supply chain (from

source through treatment to the point of consumption) as a whole candeliver water of a quality that meets the above targets (WSP component1).

3. Operational monitoring of the control measures in the supply chainwhich are of particular importance in securing drinking water safety(WSP component 2).

4. Management plans documenting the system assessment andmonitoring, and describing actions to be taken in normal operation andincident conditions, including upgrade and improvement documentationand communication (WSP component 3).


5. A system of independent surveillance that verifies that the above areoperating properly.”

Within the European Union risk assessment and management is regardedas a valuable approach to reduce the risk of adverse effects to humanhealth and the environment in several arenas (e.g. food safety, geneticallymodified organisms, ecological risk of chemicals). Activities are ongoing toharmonise the different aspects of risk assessment procedures (as outlinedin the Commission report The Future of Risk Assessment in the EuropeanUnion). The EU research project MICRORISK aims to provide theframework and the scientific basis for the introduction of the risk analysisprinciples in the area of microbiological drinking water safety.

Experience with risk management approaches in the drinking watersector

The application of risk assessment and risk management principlescomplements end-product testing by process control, and thus has thepotential to increase drinking water safety and the protection of publichealth. From current experience, a number of additional benefits for allplayers in the drinking water sector, i.e. legislators, water suppliers,authorities that undertake independent surveillance of water supply(usually, but not exclusively health bodies), and consumers have beenreported. For the legislator the approach is sensible because it:

• is proactive and presents a “push towards innovation“ as systematicand integral analysis of water supplies becomes an obligation;

• provides a consistent, objective and science-based approach that canbe applied as a overall framework to all water supply systemsirrespective of their size, source waters and technologies employed;

• is an instrument for proving that state of the art principles arefollowed and that the barriers are intact;

• introduces a sound basis for cost-benefit analysis;• stimulates sector co-operation between water suppliers and

(environmental) health authorities; and• provides a platform that formally can integrate catchment protection

aspects.

The development and implementation of a risk assessment andmanagement approach is first and foremost an activity carried out by thewater supplier. In many cases, many elements are already part of watersupplier’s good practice in quality assurance. There are, however,additional advantages for water suppliers, these include:

• for larger supplies, formalise and organise existing procedures into asystematic and accessible package that can be easily communicated(internally and externally); these can easily be integrated into existingquality management systems (e.g. ISO 9001);


• for smaller supplies, provide a tool for basic assessment of needs andhelp to promote good asset management;

• target the resources and attention toward the critical issues ofdrinking water quality and help to focus water supplier’s monitoringactivities on a smaller number of critical parameters;

• document that due diligence is in place and thus protect againstallegations of negligence;

• increase transparency towards consumers and surveillance agencies;• provide a platform for the management of existing knowledge,

competence and experience within the water supply; and• provides an opportunity to increase the knowledge and know-how of

utility personal about all aspects affecting water quality.

Many authorities that undertake surveillance of water supply may also befamiliar with the HACCP principles from their surveying activities in thefood sector. Their application in the drinking water sector has alsoadditional advantages to health authorities as they:

• provide a framework that can be audited and assessed in astandardized way;

• provide improved understanding of health risks and assessment ofpriorities as part of overall public health policies;

• provide greater confidence in the continuous management of drinkingwater quality, particularly for the small, easily “overlooked” supplies;

• provide improved understanding of operational aspects of supplyingdrinking water, the range of hazards and what can go wrong; and

• trigger more intensive cooperation and communication with watersuppliers and with environmental authorities.

First experiences with the application of risk assessment and managementstrategies in drinking water also made clear that the involvement andcommitment of all stakeholders is a major challenge and that there is asignificant need for capacity building and staff training about the aspectsof risk assessment and management approaches. Furthermore, initialinvestments and costs are associated with the implementation of thisapproach. The presence of an operational quality management systemappears to be a favourable factor for success of the approach, and mayreduce the investments needed.

POSITION OF THE PREPARATORY GROUP

The development and implementation of the EU Drinking WaterDirectives has given clear guidance and a great impetus to the watersupply sector in Europe that has resulted in an overall high level ofdrinking water quality and supply infrastructure.

In the light of the limitations presented by the end-product testingfocussed approach to guarantee safe drinking water, the developments


within other arenas of consumer safety, the approach taken by the WHOGDWQ, and the first positive experiences in using comprehensive riskassessment and management approaches in drinking water reported fromdifferent countries, the Preparatory Group is confident in recommendingthe latter for consideration within the current revision of the DWD. Thismove is seen as a natural, timely and logical next phase in thedevelopment of the DWD.

The Group clearly judges incorporation of risk assessment andmanagement strategies as to be of large added value for the DWD and forsafeguarding the supply safe drinking water that keeps the trust ofconsumers in EU Member States. The core principles given by WHO’sFramework of Safe Drinking Water (see Box 2) are in good agreement withthe principles used by the EU in other arenas of consumer safety, and arethus seen as sound basis to be included in the revision of the DWD. TheGroup assumes that the drinking water sectors within EU Member Statesdo not represent specific circumstances that would justify major deviationsfrom WHO’s approach.

The Group also clearly endorses the uphold of the use of end-producttesting though embedded in the risk based approach as verification thatprocess control is effective, and hence more targeted to the specific criticalwater quality issues in a particular water supply. Compliance with waterquality standard values – according to the recommendations of the otherpreparatory groups – should remain as a major component within theDWD as a means of verifying that safety has been achieved. Through theadoption of a risk-based approach, however, end-product testing iscomplemented by process control, and it becomes integral part of theoverall management approach. The involvement of all stakeholders in thesupply chain from catchment to consumer will strengthen the reliability ofthe system.

The incorporation of risk assessment and management strategies willchange the way the DWD is structured. The new approach should not onlybe reflected in the annexes but also in the articles of the DWD. Theimplementation of the risk-based approach should be preceded andsupported by several pilot projects to translate and evaluate the coreprinciples of this approach in the European context.

In the following sections specific issues detailing the general statementsabove are discussed. Though introduced by WHO in the context of theGDWQ, the term Water Safety Plan (WSP) is used in this paper whenreferring to water supplier’s functions within the risk based approach (i.e.steps 2-4 in Box 2).

Mode of implementation

If incorporated in the DWD, the risk-based approach represents a major shiftin philosophy for many players in the drinking water sector. Thus it shouldbe emphasized that there is an inevitable need for an adequate preparatoryphase that allows enough time for careful and intensive sector preparation in


the Member States facilitating the adaptation and implementation of the newapproach. This includes, for example, evaluation of existing experience,carrying out of pilot studies in different national supply settings,development of guidance and training materials, and coordination of capacitybuilding and training programmes.

Different scales of water supplies require different modes of implementation.For most of large utility supplies many of the elements that comprise a WaterSafety Plan are already part of the supplier’s good practice or may beintegrated in existing quality management systems (e.g. ISO 9001). Thussetting up and implementing a WSP would should not pose major difficulties.The situation may be different for small or medium sized water supplies.Here the range of skills, expertise and resources required to performcomprehensive system assessment and to implement management plans isnot necessarily covered within single supplies. Thus, the way forward usingthe WSP approach for small and medium sized supplies is differentcompared to large utility supplies, and runs in parallel. For example, thedevelopment and use of guided model and technology generic WSPs haveproven successful in supporting water suppliers implementing WSPs.Positive experience with that type of approach has particularly been reportedfrom New Zealand and Switzerland.

The preparatory phase is also essential for health authorities (or other bodies)responsible for independent surveillance as the change in the overallapproach of the DWD is likely to involve redesigning of surveillanceprogrammes and change of surveillance techniques. This will include ashifting emphasis from compliance checking to inspection and audit basedapproaches. Again, this requires sector discussions and preparatory andenabling programmes in the Member States.

If a risk based approach is considered for inclusion in the next DWD revision,the Preparatory Group recommends that all stakeholders in the MemberStates are informed about the process in good time and are encouraged tocarry out careful sector preparation. For example, in Switzerland where theHygiene Ordinance obliging water supplies to follow HACCP principlescame into legal force as early as 1995 (see Box 1), the concept practicallystarted “living” in a broader context in the year 2000, after guidance for smallsupplies has been developed and surveillance agencies had adapted theiraudit system.

Monitoring and verification requirements

Within the WSP monitoring activities can be subdivided into two parts:process or operational monitoring; and, verification of performance thatemploys end-product testing as part of its procedures.

Process monitoring is defined as a planned sequence of observations ormeasurements of control measures that are of particular importance in thesupply chain from catchment to consumer to meet the water quality targets.The objective of process monitoring is timely assessment of whether theprocess is under control. In most cases, routine monitoring will be based on


simple surrogate tests, or observations and inspections rather than oncomplex microbial or chemical tests. For example, if water is disinfected bymeans of UV radiation continuous monitoring of lamp intensity would beadequate to control the disinfection process rather than microbial testing offinished water at high frequency. If process monitoring shows that definedperformance targets or critical limits (e.g. lamp intensity) are not met, thentimely corrective action is to be taken (e.g. use of backup UV lamps) asresponse to bring back the process under control and prevent the water frombecoming unsafe. It needs to be emphasized that process monitoring is notlimited to technical procedures in the waterworks but can also includeinspection of catchment or distribution system integrity, for example.

In the context of the DWD revision, the Preparatory Group recommends aflexible and stepwise shift of end-product testing requirements. It issuggested that in principal the full range of parameters given in thecurrent DWD Annex I (or in the possibly revised annex of the futureDWD) should be kept in place and should be tested at the requiredfrequencies given in current Annex II.

As a “new” alternative to the above “traditional” monitoring approach, theGroup endorses the possibility of supply system tailored end-producttesting programmes. Those should be flexible with regard to the range ofparameters tested and to monitoring frequencies according to the specificsupply situation and to the hazards identified. Their set up, however,needs to be linked to the fulfilment of the following two prerequisites:

• the water supplier must set up and implement a WSP that comprises acomprehensive hazard analysis from catchment to consumer, definescontrol measures together with critical limits, operational monitoringand corrective actions; and

• the WSP together with the end-product testing programmeestablished in it needs to be approved and regularly audited by theresponsible health authority in the Member State.

The Preparatory Group further recommends that the DWD foresees atransition period in which both monitoring system alternatives are allowedin parallel while after the end of the transition period only the secondoption, system tailored end-product monitoring programmes, should beforeseen. It is assumed that such monitoring programmes are more (cost)effective for both water suppliers and surveillance agencies than inflexibledefault programmes that do not take into account the specifics of supplysettings.

Verification is a critical activity that should be undertaken by both thewater supplier and the surveillance agency (or designated third parties).Verification should employ both a mixture of end-product testing andaudit of performance of the Water Safety Plan to assess whether risks areappropriately controlled. Verification is the use of methods, procedures ortests in addition to those used in process or operational monitoring todetermine if the WSP is in compliance with the water quality targets. Inthis context, verification may include regular testing of micro-organisms


that provide an indication of whether microbial risks are adequatelycontrolled. This will include E. coli, but may also include indicatororganisms that demonstrate control of protozoa and viruses.

In addition to operational monitoring and verification testing carried outby the water supplier, independent surveillance of health risks associatedwith water supply is integral part of the overall approach. It is usuallycarried out by regulatory bodies, which may be local or regional(environmental) health or environment departments, or designated thirdparties. The role of verification is to assess whether the whole system isoperated properly and thus provides an independent oversight.Traditionally independent surveillance consists of different steps such aswater quality testing, review of water quality monitoring data collected bywater suppliers, sanitary inspections of treatment facilities or catchmentinspections. In the light of the new approach, surveillance programmesneed to be adapted accordingly by including regular audit procedures thatcheck and assess the completeness, adequacy and implementation of theWSP. An example for the adaptation of surveillance programmes inSwitzerland is given in Box 3. An example for the optimisation ofmonitoring programmes and their harmonisation between water suppliersand surveillance agencies in France is given in Box 4.

Box 3. Public health surveillance in Switzerland.

The design of public health surveillance programmes carried out bythe Cantonal Chemists (Swiss health authorities) was revised after theintroduction of HACCP principles into national legislation. Thesystem that has been introduced in 2001 is based on inspection andconsists of the following four domains:

A. Self-check system: This step evaluates whether the requirementsgiven by the Hygiene Ordinance (see Box 1) are fulfilled by thewater supplier. It thus assesses, for example, the adequacy ofhazard analysis performed, CCPs and corrective actions identified,as well as checking the adequacy of operational instructions andmonitoring plans.

B. Water quality: This step evaluates whether water quality standardshave been met.

C. Processes and activities: This step evaluates, for example, whetherabstraction, treatment and distribution of water is in line withrules of good practice, maintenance is carried out properly, andthe self-check system is used.

D. Buildings, equipment and devices: This step evaluates the structuralconditions and hygienic conditions at abstraction and treatmentfacilities.

Box 4. Optimisation of monitoring programmes in France.


Though the French transcription of the DWD into national law(Decree 2001-1220) in general requires independent surveillance ofwater quality through the health agencies, § 18 of the Decree stipulatesconditions under which part of the water supplier’s monitoring datacan become integral part of the regulatory surveillance and thusreduces regulatory monitoring expenditures. The optimisationrequires the existence of a quality assurance plan which mustdocument the performance of a regularly risk analysis, theidentification of critical control points together with effective controlmeasures, that operating instructions are followed and that staff isappropriately trained. Though not explicitly mentioned, these criteriaare very much based on HACCP or WSP principles. In addition, theoptimisation further requires that all analyses are made by accreditedlaboratories or within a certified quality system.

Responsibility from catchment to tap

The basic principle that underpins the risk based approach is that itprovides a platform for integrated risk assessment and managementthroughout the full supply chain from catchment to consumer’s tap. It thusenlarges the efficiency of the multi-barrier-principle.

In the specific European Union legal and regulatory background, the set-up and implementation of a WSP is generally the responsibility of thewater supplier though it typically involves supporting actions by otherstakeholders that are often regulated through other Directives. This maybe particularly important when identifying hazards and control measureswithin catchments where the water supplier does not own the land, andhence has no direct control of activities potentially polluting drinkingwater sources (e.g. in agriculture, industry). Equally important, in manycases the water supplier does not exert control over plumbing practicesand materials used in installations in private or public building thatpotentially impair drinking water quality at the tap.

In principle the water supplier’s WSP team can act as catalyst forcollaboration with different stakeholders responsible at both ends of thesupply chain (catchment and consumer’s tap) and establishing a sense ofmutual ownership for controlling contaminants at their source.Stakeholders that may be involved include, but are not restricted to,environmental protection or catchment agencies, actors involved inenvironmental issues within catchments (e.g. farmer associations), houseowners or local plumbers (or associations).

The Preparatory Group therefore recommends that the DWD

• accentuates the need for integrating the catchment and consumer’s tapends of the supply chain into the overall system assessment andmanagement; and

• defines the responsibilities at various levels (e.g. EU, Member States,regional, professional) to ensure reliable and efficient involvement andcollaboration with other concerned stakeholders in order to fulfil the


above requirement. In the context of the protection of drinking watersources, the Commission should seek for strengthening drinkingwater aspects within the Water Framework Directive (WFD). Thiswould serve as a good basis to link environmental and health issues incatchment protection from a drinking-water quality point of view, andthus trigger better coordination of individual activities andregulations.

• For installations in buildings, the provision of adequate toolsregulating the quality of construction products (like the upcomingEAS system) is essential for ensuring the efficiency of WSPs at the endof the supply chain.

An important issue that has come to the attention of the PreparatoryGroup when considering Water Safety Plans is the need to strengthen andfocus efforts on health risks from drinking water but at the same time notto compromise the high European standards based on the long establishedprecautionary principle or on the protection of the environment. TheGroup has expressed the opinion that both visions are morecomplementary than competitive.

Risk based approaches are also applicable in environmental or catchmentprotection. Risks to source water quality from human activities in thecatchment are best managed at their source instead of downstream in thedrinking water supply chain, as this often is inefficient or costly. Thus,there is an increased need for coordination between various Directives (e.g.DWD, WFD, or the coming Ground Water Directive) to adequately protectdrinking water resources and to provide safe drinking water in the mostefficient way. The Group encourages joint efforts to put DWD duringrevision in a wider context of water cycle and water quality management.

Water quality targets

In the current DWD, the general obligation of water suppliers is to providedrinking water that is wholesome and clean (Article 4). Water is considered tobe wholesome and clean if it is free from any contaminant (microbiological orchemical) in levels that constitute a risk to human health. This generalrequirement is translated to water quality targets (standards) that should bemet in tap water.

Several water quality targets are health-based targets. Water quality targetswith a clear health-base already exist for carcinogenic chemicals. Thestandards for these chemicals are set using a tolerable annual risk of 1 case ofcancer in 1 million people. This can be regarded as the reference level oftolerable health-risk in the DWD. For pathogenic micro-organisms, no specificwater quality targets are set. The absence of indicator-bacteria (E. coli,enterococci) in 100 ml water is regarded as a measure of safety.

In the DWD not all standard values are health based. Several parameters havebeen set on the basis of other criteria, such as the precautionary principle (e.g.pesticides), consumer acceptance (e.g. taste, odour, colour), or the protection


Figure 1. Target levels.

of technical integrity of supply systems (e.g. chloride, conductivity orsulphate for avoiding corrosion in distribution systems).

The introduction of a risk-based approach to manage drinking water safetyfrom catchment to tap principally requires a risk- or health-based target. Therisk level that is considered safe is the target level to be met by for the watersupply system. Water suppliers can use this target level to check whethertheir supply system is producing and delivering safe water. The currentstandards for carcinogenic chemicals are already health-based and can beused in this new setting.

For pathogenic micro organisms, health-based targets are also needed. Areference level of risk has been used to derive standards for carcinogenicchemicals (see above). However, this level cannot directly be translated to atarget level for pathogens, as the disease burden of pathogenic microorganisms is different from carcinogenic chemicals. A scale is needed thatintegrates the health-burden of different parameters and that allowstransparent and consistent standard-setting. In health-care, this type ofhealth-burden scale has been developed over the last decades. ThePreparatory Group considers the Disability Adjusted Life Years (DALYs)approach appropriate as burden of disease scale for translating the referencelevel of tolerable risk to water quality targets for pathogenic micro organisms.

It is explicitly not the intention to produce a long list ofwater quality targets for all waterborne pathogensincluding monitoring requirements for these pathogens indrinking water. Rather, the water quality targets forpathogens should be translated into performance targetsfor control measures in source protection or treatmentprocesses, for example. This also means that the generichealth-based target is translated into site-specificperformance targets. Sites with more contaminated sourcewaters need a different set of control measures than siteswith more pristine source waters. In this way, a generictarget for the protection of consumer health is translatedinto site specific performance target (Figure 1).

To determine whether a water supply system is able to meet the health-basedwater quality targets, a quantitative risk assessment is necessary. This is aformalised procedure for systematic combination of quantitative informationabout the level (or probability) of contamination of source waters and theefficacy of the control measures to assess the level of safety of drinking waterwith respect to pathogens (Box 5). This approach is also applicable to toxicchemicals.

Monitoring should be directed primarily towards determining compliancewith performance targets. This means that a water supplier should identifyperformance parameters (such as disinfectant concentration in combinationwith contact time, particle removal of filters, membrane integrity) whichdemonstrate continuous compliance with the performance targets and hencewith the health-based targets.

Health-based target(tolerable risk level of 10-7 DALY)

Water quality target(tolerable pathogen concentation)

Performance target(monitor efficacy of control measures)


This chain of target derivation helps to focus control and monitoring effortstowards pathogens that are a challenge to a specific water supply. It alsoallows the effect of short-term events (such as heavy rainfall, maintenanceand repair or process problems) to be taken into account that are importantbecause they may lead to a significantly increased health risk.

The preparatory group underlines that the protection of public healthshould be the primary basis for setting water quality targets. However, thegroup also acknowledges that other criteria such as the precautionaryprinciple, consumer acceptance, or the protection of technical integrity ofsupply systems are important factors that must be considered in standard-setting. These factors should, however, never compromise the adequateprotection of consumer health. The group therefore suggests that thereason(s) for standard designation within the DWD need to be madetransparent for each standard value.

Box 5. Five steps in quantitative microbiological risk assessment (QMRA).

Step 1: Problem formulation - This is the initialising step to establish whichspecific questions the QMRA needs to address. This requires communicationbetween the risk managers and the risk assessors to identify key issues andconstraints.

Step 2: Hazard identification - In risk assessment, hazard identification is thecharacterisation of the micro-organisms that cause human illness throughdrinking water. The ideal QMRA focuses not only on a single pathogen, buton a suite of "index pathogens", that cover the range of health risks andcontrol challenges for the particular system defined. Adequate control ofthese index pathogens would imply that the health risk of other known (andvery probably also unknown) pathogens is also adequately controlled.

Step 3: Exposure assessment - Exposure assessment is the quantitativeassessment of the probability that drinking water consumers ingest pathogensthrough this drinking water. In QMRA of drinking water, this usually requiresthe assessment of the levels of pathogens in source water and the changes tothese levels by treatment, storage and distribution, and finally the volume ofwater consumed. The information obtained in all steps of the exposureassessment is combined into an estimate of the ingested dose.

Step 4: Effect assessment - The effect assessment is the determination of thehealth outcomes associated with the (level of) exposure to waterbornepathogens, using dose-response data of the pathogens in human volunteers,preferably adjusted for the immune status of the population. The response canbe measured as risk of infection, but a burden-of-disease scale (such asDALYs) is preferred as it allows comparison of different health risks.

Step 5: Risk characterisation - The information obtained in the exposureassessment and the effect assessment is integrated to obtain a risk estimate.This can be done as a point estimation: which can be the 'best' estimate, toobtain a measure of central tendency of the risk (valuable for scenario-studies). Incorporation of the variability and uncertainty in the steps of therisk assessment chain encompasses the characterisation of the distribution ofall data used for risk assessment and to combine these distributions into adistribution of the computed risk. This approach not only provides the riskmanager with important information about the (un)certainty of the risk


estimate, but also with the relative contribution of the uncertainty andvariability in all steps of the risk assessment.

Costs involved in developing risk management approaches

Costs involved in establishing new risk assessment and managementprocesses are central in any discussion of revision of the DWD. However, thisdebate must also be balanced against benefits of the system to allow ameaningful comparison with existing approaches. Risk assessment andmanagement approaches typically emphasise improvement of operationalpractice. Therefore, additional activities to secure safety are likely to alsoimprove asset management and protect capital investment and lead toreductions in aspects such as unaccounted-for-water, thus improving thecost-return ratio of water production.

Costs may be accrued in four key areas:

• risk assessment and defining health-based targets;• developing the WSP and related quality management systems;• formal steps in auditing the WSP; and• practical expenses to adjust quality monitoring and supply system

management to meet WSP requirements.

Quantifying these costs for all water supplies is difficult because it can beexpected that local conditions will strongly influence what activities mustbe done and what existing data and practices can be co-opted into the WSP.It would appear likely that developing risk-based approaches will notresult in increased costs in larger water utilities and could lead to areduction in costs. Risk-based approaches may entail some increase in costsfor smaller water supplies in some EU countries, but this is offset againstgreatly improved water safety management.

The major costs are likely to be accrued in terms of staff time in preparing theplan to be followed and performing risk assessments. Ongoing costs may alsoinclude auditing. Equipment costs will vary and depend on currentequipment available, but it would be expected to include capital andrecurrent costs although the latter would be unlikely to exceed current costsof end-product testing. Applying the WSP approach does imply that risksfrom “emerging parameters” (in the sense of those not yet regulated) must becontrolled and this may have financial implications, although primarilythrough risk assessment than management. Tailoring mandatory end productmonitoring for verification into a larger quality monitoring framework is theonly way to offer a better risk management while limiting expenses as awhole.

The area where it is likely to be a significant new cost is in the performance ofrisk assessments of water supplies in relation to health-based targets.Although some water supplies may have undertaken these assessments for atleast some index pathogens, it is unlikely to have been performed for allwater supplies or all pathogen groups. These costs should again be offset


against the benefits of improved information about the health risks associatedwith water supply. It is also likely that any risk assessment programmewould be rolled out progressively, indicating costs could be spread overseveral years.

Box 6 provides some indication of the cost associated with experiences to datewith use of quality management systems.


Box 6. Costs: experience from three countries in using management systems.

Australia – The cost of producing a HACCP plan varies significantlydepending on the size of facility, existing information and qualitysystems and experience of staff preparing the HACCP plan. Mostutilities embarking on HACCP plans have ISO management systemsand this is noted as a useful starting point for developing a HACCPplan. Experience has shown that costs for preparation decrease with asmore experience is gained. One utility noted that for smaller suppliesabout 2-3 person months were required to develop the HACCP plan.This same utility noted no additional staff costs were involved as theprimary result was better working practices of existing staff.Additional costs (up to € 63,000) were accrued for distributionmonitoring. In a bulk supplier, the HACCP plan required a full-timestaff member to prepare the plan and involved € 17,200 for an initialrisk assessment. Six-monthly audits cost in the region of € 2,800 andevery 3 years an updated risk assessment is performed, which costs€ 5,600.

France – WSPs or HACCP plans are not routinely employed, but costscan estimated against those accrued through applying ISO 9000systems. One utility suggests that for a supply zone of 100,000inhabitants, implementing ISO involves 400 person-days from externalexperts or dedicated quality managers in the supply zone and 200person-days from personnel. Quality management, internal andexternal audit involves a further 130 person-days per year. Such costsare not considered proportional to the size of the supply zone. Whileno data is available to smaller systems (for cost-saving reasons this hasbeen implemented at a level including several supply zones, e.g. samequality manager), for larger systems an additional 15% costs areestimated for zones supplying 500,000 people and 30% for thoseserving 1,000,000. It is estimated that implementing a WSP would be30 person-days by supply zone. Were a supply zone not have a ISO9000 systems start, the overall cost would be greater, but the expenseswould cover then a larger need (and benefit) than WSP alone.

UK – Some UK water supplies have ISO 9000 series accreditation.Data from one company, which has to operate a water quality controlprogramme based on microbial testing at the same time because ofregulatory requirements, showed that the cost of staff time to run theISO management system, primarily accrued through documenting theprocess, is estimated to be € 141,000. Audit costs are in the region of€ 21,000 to € 28,000. This utility notes that they are unable to benefitfrom expected reductions in cost of microbial monitoring because ofcurrent regulations, but did not believe that risk based approacheswould increase monitoring costs and may actually lead to a reductionin costs.

Communication and reporting


Communication on the risk based approach to the public is critical to boostconsumer confidence without giving the impression that the current situationis unsatisfactory or unsafe. This implies that consumers should be convincedthat the existing approach to water quality management guarantees safe andwholesome drinking water, and that the new approach is a combination andre-enforcement of long established codes of good hygienic practicescombining water quality monitoring throughout the production process andend product testing. The risk-based approach means a more structured andmore efficient (cost-effective) approach to manage drinking water qualityfocussing expensive monitoring and sampling efforts to the most criticalpoints.

Part of communication is reporting to national/regional authorities, to theCommission and to the public at large. Reporting in the current situationprimarily concerns end product testing. In a new risk based approachreporting on final water quality will continue to play an important role as itprovides the proof for having met drinking water standards. With the newapproach, however, reporting on procedures and codes of practice in placemay become an additional requirement. This will change reportingrequirements from water suppliers to national authorities and to the public,and from Member States to the Commission.

The Preparatory Group has noted that the general public find comparing testresults with quality standards is a simple way to comprehend risk issues. Incontrast, drinking water experts and authorities have been worried by anoften too simplistic way of looking at drinking water quality that hamperscommunication about assessment and prioritisation of competing risks thatcan not be fully described in legal numbers. An example of this is the long-term risk from the consumption of disinfection by-products effects versusshort term risk through microbiological contamination.

Relation to EU policies and current DWD principles

The application of a risk assessment and management approach as a basis isconsistent with EU policies and developments in other areas of consumerhealth (such as food safety) and environmental health. It is a science-ledprocess to establish the likelihood of adverse effects to human (orenvironmental) health from exposure through risk sources.

It provides a coherent framework to allow risk management to betransparent, proportionate and based on science. In the area of food safety,the EU played an active role in the field of food safety to obtain European andinternational acceptance for risk analysis principles. This is illustrated by theWhite Paper on food safety produced by the Commission and the adoption ofa “modern, dynamic and effective legal framework for food safety, based on robustscience”.

With such an approach for drinking water, the use of the PrecautionaryPrinciple (as described in the Commission Communication in 2000) can bebased on a quantitative assessment of the risk of pathogens or toxic chemicalsoccurring in drinking water. This risk can be compared to other risks, and EU


drinking water policy can be based on a level of risk that is consistent withother areas of consumer safety. This approach is non-discriminatory, basedon cost-benefit assessments, transparent and it indicates where more scientificevidence is necessary to reduce the uncertainty in the assessment of risk.

The Standstill Principle implies that the current high level of consumerprotection is maintained but also that the current high level of environmentalprotection and consumer acceptance is upheld. It is envisaged that theimplementation of the new risk-based approach would improve the level ofconsumer safety. Risk assessment is an instrument to quantify the level ofsafety and hence to demonstrate that the Standstill Principle is met in future,using the current protection level as a benchmark.

Currently, the health risk of the drinking water consumer is very low,compared to other routes of exposure to pathogenic or toxic agents. It shouldalways be remembered, however, that this high level of safety is the result ofan huge effort in water supply and sanitation in the EU. Compromising thislevel of protection can cause wide-spread health effects. It is therefore of theutmost importance that the Standstill Principle is strongly uphold in the newDWD.

Sector preparation and development of supporting tools

It is worth the effort to make the risk assessment and managementapproach a living and accepted concept amongst the majority of thestakeholders in the drinking water sector. The lack of knowledge andexperience amongst many stakeholders creates the need for moderatedsector discussions at EU level and within Member States in order todevelop a common understanding and interest. Only if regulators, watersuppliers (associations), health departments and consumers are convincedthat the new approach will significantly increase drinking water safety andthus protect public health better, can the approach successfully betranslated into practice.

The practical implementation of risk assessment and managementstrategies requires preparatory activities that provide basic informationtools and thus support capacity building within all stakeholder groups.Examples of such activities include:

• provision of information sources in Member States that explain theprinciples of risk assessment and management strategies in drinkingwater;

• provision of information sources in Member States that describesexisting experience and lessons learned reported from countries wherethis approach has already been implemented (e.g. Switzerland,Australia, New Zealand, Uganda);

• developing pilot WSPs in the Member States for different watersupply settings (some pilot studies are already planned, i.e. in Finland,The Netherlands and Germany);


• on the basis of both consolidated experience from the pilot studies andexisting national technical rules/specifications, development ofconcrete and Member State specific guidance on ‘How to do?’ the WSP;

• developing and carrying out training programmes for allstakeholders;

• development of specific guidance for small systems, i.e. technologygeneric (‘recipe like’) guidance build on existing experience such asthe Swiss Recommendations for a simple quality assurance system for watersupplies or the New Zealand Public Health Risk Management Plans;

• commission research on the development of a knowledge base forhealth-based targets; and

• development of public health surveillance programmes in theMember States that take into account new demands of WSP audits,and conducting relevant training for public health departments.

If a risk based approach is considered for inclusion in the next DWDrevision, the Preparatory Group recommends that all stakeholders in theMember States are informed about the process in good time and areencouraged to carry out the preparatory activities outlined above. ThePreparatory Group encourage both the Commission and Member States tostrongly support these activities in order to make the shift in approach atthe legislators level a success on the ground.

KEY ISSUES FOR FUTURE STEPS

In the light of current knowledge and experience, the Preparatory Group hasidentified the following outstanding features for the risk assessment and riskmanagement approach:

• Primary reliance on end-product testing is not on its own adequate toprovide safe drinking water which has the trust of consumers. In therisk-based approach, the whole chain of water production anddistribution (from catchment to consumer’s tap) is systematicallymanaged by defined responsible bodies working under properinstitutional arrangements.

• It encourages the use of multiple barriers to provide safe drinkingwater.

• It provides better protection against microbiological hazards than end-product testing, which on its own has serious limitations.

• It is focussed on prevention rather than reaction.• Targeting the limited resources to managing (critical) control

measures is more cost-effective than end-product testing alone.• Simply relying on the water being safe when the DWD parameters are

met gives a false feeling of security. The risk based approach triggers aprocess of continuous assessment and improvement and focuses oncritical issues in specific supply settings.


• Effective quality management systems are already actively in placeand practised in many large utility supplies. Whilst further work toencourage and promote even more widespread adoption in suchutilities will be of value, it has to be recognised that small suppliesoften pose the biggest risks to European citizens. Risk basedapproaches tailored for the needs of small supplies help to improvethe situation.

Before the risk based approach can be promoted further in the EU context,there are some challenges which need to be addressed in further studies.These include:

• If risk assessment is added on the top of the requirements of existingdrinking water legislation, the costs for water suppliers are likely toincrease, especially in small communities. Ways by which theapproach can be introduced in the most cost effective way need to beinvestigated, e.g. by definition of representative supplies with roll-outbased on either prescribed WSPs or guided processes. For example,there is a need for supply system-tailored end-product testing andthus a need for more flexible monitoring of parameters than thecurrent DWD provides. In that way, the costs could be partlyreallocated from end-product testing to operational monitoring.

• The transition period needs to be sufficient for preparative activitiesand capacity building among different stakeholders. Member Statesneed to provide enough resources and training for the relevantauthorities to be able to cope with any new responsibilities.Comprehensive guidance documents are needed both in the EuropeanUnion and in the Member States.

• Across Europe there is significant variation in the type and quality ofraw water resources, water supply infrastructure, legal andinstitutional arrangements. Some elements of the risk-based approachare already in place in some Member States. While a commonEuropean framework for a risk based approach needs to beconsidered, it is essential that this allows enough flexibility for thedesign of approaches which are adequate for the national setting.Also, the development and application of a risk-based approach mustbe flexible enough to allow existing good practices to continue and toallow new Member States to adapt to the new situation.

• Key areas of the whole supply chain from catchment to theconsumer’s tap are not sufficiently regulated under the currentmandate of the DWD: specifically, source water or catchmentprotection and household installations. The integration of the DWDwith other elements of EU policy and legislation is therefore criticallyimportant. This must include stronger protection of raw watercatchments under Article 7 of Water Framework Directive and theproposed new Groundwater Directive, and also measures to minimiserisks for plumbing and other materials for example under theEuropean Approval System for construction products in contact withdrinking water.


Therefore, the Preparatory Group recommends that the risk assessmentand risk management approach described in the third edition of the WHOGuidelines for Drinking-water Quality should be gradually introduced in thecurrent European drinking water policy and legislation. A stepwiseapproach helps to adequately fill research needs and knowledge gaps, andto motivate the majority of stakeholders to make this concept living tool.

2.2.5 Position paper on materials and substances in contact with drinking water

POSITION PAPER GROUP 5: SUBSTANCES AND MATERIALS USEDIN THE PREPARATION OR DISTRIBUTION OF WATER INTENDEDFOR HUMAN CONSUMPTION.

INTRODUCTION On 10-11 September 2003, at the DG Environment Offices in Brussels, a panelof experts (see annex 1) discussed the current Drinking Water Directive(DWD), with respect to substances and materials used in the preparation ordistribution of water intended for human consumption, in the context ofpossible changes to the DWD.This position paper presents the conclusions of the discussions.

The Drinking Water Directive (98/83/EC)

Article 10 of the DWD (Quality assurance of treatment, equipment and materials)states that:

“Member States shall take all measures necessary to ensure that no substancesor materials for new installations used in the preparation or distribution ofwater intended for human consumption or impurities associated with suchsubstances or materials for new installations remain in water intended forhuman consumption in concentrations higher than is necessary for thepurpose of their use and do not, either directly or indirectly, reduce theprotection of human health provided for in this Directive. The interpretativedocument and technical specifications pursuant to Article 3(1) and Article 4(1)of Council Directive 89/106/EEC shall respect the requirements of thisDirective.”

The measures that member states must undertake are not spelt out. Furthermore, therequirements of the DWD are not specified with respect to the technical specificationsof substances and materials.Water intended for human consumption is defined in the DWD in Article 2(Definitions) as:

“1. water intended for human consumption` shall mean:


(a) all water either in its original state or after treatment, intended fordrinking, cooking, food preparation or other domestic purposes, regardless ofits origin and whether it is supplied from a distribution network, from atanker, or in bottles or containers; (b) all water used in any food-production undertaking for the manufacture,processing, preservation or marketing of products or substances intended forhuman consumption unless the competent national authorities are satisfiedthat the quality of the water cannot affect the wholesomeness of the foodstuffin its finished form;

2. domestic distribution system` shall mean the pipe work, fittings andappliances which are installed between the taps that are normally used forhuman consumption and the distribution network but only if they are not theresponsibility of the water supplier, in its capacity as a water supplier,according to the relevant national law.”

The relevant point of compliance with the DWD is defined in Article 6 (Point ofcompliance) as;“The parametric values set in accordance with Article 5 shall be compliedwith: (a) in the case of water supplied from a distribution network, at the point,within premises or an establishment, at which it emerges from the taps thatare normally used for human consumption”.

The DWD puts an obligation on member states that goes further than complyingwith a (minimum) set of quality requirements, given in Annex I of the DWD. Article4(1) requires Member States to take all measures to ensure that water intended forhuman consumption is free from any micro-organisms, parasites and from anysubstances, which in numbers or concentrations, constitute a potential danger tohuman health.Member states may set more stringent parametric values (Article 5.2) and shall setadditional requirements where the protection of human health so requires (Article5.3).

PRODUCTS AND MATERIALS IN CONTACT WITH WATERINTENDED FOR HUMAN CONSUMPTION

Current situation

Member States have developed a variety of national schemes to ensure thatdrinking water in contact with products and materials used for storage anddistribution remains safe and pleasant to drink. The nature of these schemesdepends on the type, such as plastic, cementitious and metallic, of materialand on the member states. The national schemes may be legally bindingtesting and approval systems or voluntary codes of practice or a combinationof the two. These schemes are different among Member States and, often, aproduct approved by one Member State is not accepted by another MemberState. Re-testing is needed in these cases before it can be placed on themarket. This is seen as a barrier to the ‘open market’ in Europe and incurs


significant additional testing and certification costs. This also has the effect ofreducing competition in the water sector and the water industry pays morethan it would in a competitive environment.

Such regulated products and materials used for storage and distribution ofwater intended for human consumption that are permanently incorporated inconstruction works fall within the scope of the Construction ProductsDirective (CPD, 89/106/EEC) and, as mentioned already, are referred to inthe DWD.

The CPD sets the legal framework for the CE marking of constructionproducts that proves they are fit for their intended use. Essential Requirement3 of the CPD deals with hygiene, health and environment requirements forconstruction works, e.g. water supply networks. In this respect, the MemberStates must take measures to ensure that construction works (including watersupply networks) in their territory satisfy this Essential Requirement.

In order to satisfy the legal requirements of both directives, the EuropeanCommission took the initiative and established, in 1999, the Regulators Groupfor Construction Products in contact with Drinking Water (RG-CPDW), withthe approval of the Standing Committee on Construction (SCC) and theStanding Committee on Drinking Water (SCDW).

The RG-CPDW is responsible for making proposals for a EuropeanAcceptance Scheme (EAS). The EAS is intended to be a single system ofregulatory framework and supporting test methods covering the acceptanceof construction products used in contact with water intended for humanconsumption. Construction products receiving a favourable assessment underthe EAS will carry an EAS mark in addition to a CE mark.

The RG-CPDW comprises representatives of the relevant services of theEuropean Commission, representatives of the regulatory authorities andscientific experts appointed by Member States. Observers from tradeorganisations, the Comité Européen de Normalisation (CEN) and theEuropean Organisation for Technical Approvals (EOTA) attend meetings.

The EAS is still under development. So far, the European Commission hasissued a mandate to CEN (M136), an EC Decision on the Attestation ofConformity and, recently, a Commission Communication on the EAS. Thelatest version of the scheme in the public domain is the RG-CPDW Report (i.e.the so-called EAS on Paper): November 2001 (available at the DG Enterprisewebsite:

http://europa.eu.int/comm/enterprise/construction/internal/essreq/eas/easrep.htm.

The EAS Report provides an interim report on progress made by the RG-CPDW in developing its proposals. It has been out for public consultationand is currently being updated. The EAS on Paper provides information on:


Justification for the EAS, in terms of legal obligations and the perceived costsand benefits of the proposed approach.The main features of the EAS, including the regulatory and technical aspectsand the proposed administrative and institutional framework, such as therole of notified bodies.The relationship between the EAS, CEN, EOTA and national acceptanceschemes.Proposed test requirements and arrangements for developing test methods,including the research programme launched at the European level.The basis for protection of the health of consumers and the development ofacceptance criteria and acceptance levels.

Further development of the EAS and the decision-making process, withreference to the CEN mandate and EC Decisions on the EAS and attestation ofconformity.Transitional arrangements for replacement of national acceptance schemes,following the adoption of the EAS by Member States.Information on the procedure necessary to put CPDW on the Europeanmarket.

Not all elements of the EAS are finalised.

Specific aspects for the RG-CPDW in the process of development andagreement are:EAS Positive Lists and EAS Composition ListsAssessments and acceptance levelsSpecification of assessment requirements of products that do not need fulltestingGuidance documents, including uniform principlesLegally-binding Commission decisions necessary for the implementation ofthe EASThe detailed specifications for the evaluation of conformity of constructionproducts in contact with water intended for human consumption.

POSSIBLE CHANGES TO THE DWD

Although the establishment of the EAS is in response to the requirements ofthe CPD, it is in principle intended to be a system to contribute to thecompliance of drinking water with the general and specific requirements ofthe DWD. As such, it would be appropriate to make specific reference to theEAS in the DWD. While the EAS is intended to be legally binding on MemberStates at present it is not so. Mainly because not enough detail on the EAS hasbe established. It may be that once the main principles have been specifiedand the regulatory framework has been developed, Member States will agreean EC Decision (possibly at the end of 2004). However, it could be that alegally binding EC Decision is only possible when the full EAS (i.e. theregulatory framework, guidance documents, harmonised product standardsand supporting test standards) is available for consideration (around 2006).


Thus reference to the EAS in the DWD cannot be undertaken while the EAS isnot legally binding.

The panel considered the scope of the DWD compared to that of the EAS. Inthe EAS Report the implementation of provisions is left to Member States.This seems to be in contradiction to the DWD, which definitely coversmaterials and substances used in the waterworks and distribution to theconsumers tap and, in the opinion of some Member States, transport andstorage of untreated water that is intended (after treatment) for humanconsumption.

The panel also saw possible differences in scopes in relation to hot waterapplications. However, the panel was uncertain about the exact scope of theDWD and the EAS to draw any firm conclusions on this particular aspect. InArticle 10 the term ‘protection of human health’ is used and, as an example, itwas unclear as to whether this would imply taking a shower. Furthermore,Article 2 mentions ‘all water … intended for…‘other domestic purposes’’. It isunclear what ‘other domestic purposes’ includes. The EAS includes domesticapplications such as ‘hot water’ but it is impossible to decide whether theDWD requirements are covered.”

Also considered by the panel was the level of consumer protection offered bythe EAS in relation to that offered by the DWD. It was concluded that level ofprotection was not an easily quantifiable concept but, in general, the EAS willoffer a compatible level of protection. This should be evaluated in due course.At the time of completion of the EAS, the need to specify the protection levelin the DWD should be reconsidered in view of the adoption and operation ofthe EAS.

The intention of the EAS is to include legally binding ‘Positive Lists’ forproducts made out of organic material and ‘Composition Lists’ for productsmade from metallic materials. A Positive List, for example, would listsubstances that may be used in the manufacture of a product to be used incontact with water intended for human consumption. The substances willhave been assessed for safety (using toxicity assessments similar to those usedin preparation of the DWD or by the WHO) and listed together with amigration limit that must not be exceeded by the product in the appropriateCEN test method. A Composition List would list, for example, metallicmaterial compositions that products must be complied with. The panelconsidered whether such lists that set numerical limits or compositions listsshould be a part of the DWD. Since the lists would be frequently updated, itwas not considered appropriate to include such lists, and any other similarfeatures in the EAS, in the DWD, for example as annexes,. However, theDWD could specifically refer to them if necessary. When the overallframework is further developed, it has to be reconsidered whether theprinciples and mechanisms to update these lists should be laid down in theDWD.


The panel felt that the Positives Lists/Composition Lists must have a‘European’ legal owner, for example the responsible Directorate with itsScientific Committee and a working group who would be responsible fortheir operation an updating. The panel noted that DG Enterprise, incollaboration with other European bodies, intends to nominate anappropriate ‘European’ body in due course.

RECOMMENDATIONS

For the reasons mentioned above the panel thought it inappropriate to reviseat this stage the DWD with respect to products and materials to be used incontact with water intended for human consumption. At some point in thefuture (between 2004 and 2006), reference to the EAS would be appropriate.Before then detailed examination of the scope of the EAS, compared to DWD,needs to be undertaken to identify any areas not covered by the EAS. Also,when the overall framework of the EAS is further developed, it has to bereconsidered whether, e.g. EAS-principles (related to the protection level) andmechanisms to update Positive Lists and Composition Lists should be laiddown in the DWD.

WATER TREATMENT CHEMICALS

The current situation

Water treatment chemicals are not covered by the CPD. However, a Europeanscheme for acceptance of drinking water treatment chemicals already exists inthe form of the European standards mandated by the European Commissionunder the Public Procurement Directive (89/66/EC). About 100 standardshave now been published by CEN and these include all of the chemicalscommonly used in Europe, as well as the most common traditional filtermaterials.

WG9 of CEN/TC164 drafted the European standards between 1990 and 2000.WG9 has recently reviewed its published standards to check for consistencywith the requirements of the 1998 Drinking Water Directive.

The representatives of Member States in WG9 and its sub-group includechemical suppliers, the water industry and national regulators. The standardsdefine an acceptable level of purity and include maximum concentrationlevels for the toxic metals specified in the 1980 Directive. The Standards alsospecify maximum concentrations for other parameters or characteristicsspecific to the product in question. Where the Directive does not specify aMAC for a parameter that needs to be controlled, WHO, USEPA or otherauthoritative sources of toxicological guidance have been used by WG9.

WG9 has applied the concept of the “10 per cent rule” in setting puritycriteria. The basis for the 10 per cent rule is that at the maximum dosing


concentration for a product, it may not contribute more than 10% towards alimit (e.g. in the Drinking Water Directive) in drinking water for a toxicsubstance. Where a Member State has requirements that are stricter than the10 per cent rule, WG9 has accommodated those requirements by includingdifferent classes of product within the same standard. By specifying use of anappropriate class of product, Member States can specify a requirement: forexample, a one per cent rule for a specified toxic metal. The specification ofclasses has also been used to accommodate differing traditions and practicesin relation to non-toxic constituents of chemicals e.g. the calcium hydroxidecontent of lime.

The WG9 standards do not contain attestation of conformity requirements i.e.the standards do not include a mandatory requirement for checking theconformity of a product’s composition with the requirements of the standard.It is understood that the water supplier nevertheless makes checks oncomposition, or legally binding requirements for conformity of compositionare included in purchase contracts.

WG9 has also produced guidance on purchasing chemicals and this has beenpublished as a CEN Technical Report. The guidance seeks to establish aconsistent approach towards quality assurance and quality control in thepurchase of chemicals for drinking water treatment.

Mandatory dosing concentrations for chemicals are not included in thenormative sections of the standards but guidance on dosing and othersubjects is contained in the Informative Annex of each standard. SomeMember States regulate the dosing concentrations of water treatmentchemicals in their national approval schemes. It is understood that theEuropean Commission accepts that regulation of dosing concentration (anddosing point) does not constitute a barrier to placing a product on the market.

Member States may also impose technical requirements that are stricter thanthose contained in the European standard i.e. a requirement that has not beenaccommodated within the standard by application of classes and that couldconstitute a barrier to placing a product on the market. In this case theMember State must make a notification to the European Commission underDirective 98/34/EC. The requirement may be maintained if the EuropeanCommission does not raise objections, following its consultation on therequirement with all Member States.

This approach would normally only be used if the review process forstandards was not able to keep up with emerging toxicological data.

In the current ENs the following statement is included;

“In respect of potential adverse effects on the quality of water intended forhuman consumption, caused by the product covered by this standard:this standard provides no information as to whether the product may be usedwithout restriction in any of the Member States of the EU or EFTA;


it should be noted that, while awaiting the adoption of verifiable Europeancriteria, existing national regulations concerning the use and/or thecharacteristics of this product remain in force.NOTE Conformity with the standard does not confer or imply acceptance orapproval of the product in any of the Member States of the EU or EFTA. Theuse of the product covered by this European Standard is subject to regulationor control by National Authorities.”

Existing regulations for chemicals that conflict with an EN can only bemaintained via notification according to Directive 98/34/EC.

The ENs do not contain legally binding regulatory statements.

Some aspects of the procedure for adding a new treatment chemical to thosecurrently covered by ENs were unclear. This was mainly in relation to howthe toxicology of new complex treatment chemicals and their impuritieswould be handled at the European level. In addition it was unclear how CENwould obtain the toxicological input for revision of existing ENs in the lightof new data, for example, significant changes in toxicity information onimpurities.

Possible changes to the DWD

The Public Procurement Directive requires water utilities to use ENs in theirspecifications for treatment chemicals. Member States may deviate from thePublic Procurement Directive according to reasons included in this directive.Their regulations may add more stringent requirements if they arecompatible with the EU treaty and notified via the appropriate procedures(EC Directive 98/34/EC). Thus it seems possible to conclude that ‘allmeasures’ in Article 10 of the DWD could in the case of water treatmentchemicals be clarified by adding ‘use of the relevant EN’ and a phrase thatgives the essence of the above quoted statement from the standards.However, some members of the panel were not convinced of the legallybinding nature of the Public Procurement Directive with respect to healthrelated issues of water treatment chemicals.

If reference to the EN were possible then the current DWD parametersdealing with impurities in treatment chemicals, epichlorohydrin andacrylamide, could be considered for deletion since they would be covered inthe corresponding EN. On the other hand, there should be consideration ofadding to the DWD, provisions dealing with the toxicological evaluation ofnew water treatment chemicals and their impurities as well as attestation ofconformity measures and the revision of existing ENs in the light of newtoxicological data.


It was noted that water treatment disinfectants that are placed on the market(such as chlorine) are covered also by the Biocidal Products Directive(98/8/EC). The panel concluded that the relevance of this in relation to theENs and the DWD should be checked in more detail. By analogy with theCPD, a provision in the DWD may be considered stating that ‘technicalspecifications (or uniform principles) of the BPD shall respect therequirements of this Directive’.

Recommendations

No specific changes to the current DWD are proposed. However, it issuggested to study the relation between the DWD, Public ProcurementDirective the ENs for water treatment chemicals and the Biocidal ProductsDirective.

SUMMARY OF RECOMMENDATIONS

No specific changes to the DWD are proposed.

However, the following points need consideration:

- clarification of terms such as ‘all measures necessary’, and ‘other domesticsources’ to better define the intention and scope of the DWD.

- clarification of the scope of the EAS in order to identify any requirements ofthe DWD that are not covered.

- reconsideration, at the time of completion of the EAS, of the need to specifythe protection level in the DWD with respect to the adoption and operation ofthe EAS.

- reconsideration of inclusion in the DWD of some or all of the principles andmechanisms of the EAS when the overall framework of the EAS is finalised.

- identification and installation of an “EC owner” of Positive Lists andComposition Lists.

- how the toxicology of new complex treatment chemicals and theirimpurities should be handled at the European level.

- how CEN is to obtain the toxicological input for revision of existing ENs inthe light of new data.

Annex 1:

Members TASK GROUP 5:♦ Wennemar Cramer, chair


♦ Mike Fielding, rapporteur♦ Sophie Herault♦ Ulli Schlosser♦ Hans-Jørgen Albrechtsen♦ Georgios Katsarakis♦ Nellie Slaats (contractor), assistant rapporteur

Documents for consultation: http://europa.eu.int/comm/enterprise/construction/internal/essreq/eas/easre

p.htm

http://europa.eu.int/comm/enterprise/construction/internal/essreq/eas/easrep.htm

http://europa.eu.int/comm/enterprise/construction/internal/essreq/eas/easrep.htm


3 Drinking Water Seminar (day 1)

3.1 Presentation P. Perera and Agenda of the SeminarThe Drinking Water Seminar was officially opened by Mr. P. Perera, Directorof DG ENV B. The full text of the presentation of Mr. Perera is included inAnnex I. The agenda of the DWS is attached in Annex II.

3.2 Outcome of parallel sessions day 1 of the DWS

3.2.1 Session on chemical parametersThe position paper of the expert group on chemical parameters waspresented by the rapporteur: Frantisek Kozisek. The presentation is includedin Annex III of this report.

3.2.2 Synthesis of the discussion on chemical parameters

Lecloux (CEFIC)• Harmonise WFD and DWD• Removal efficacy is not known for all priority substances

Steve White (Thames Water, UK)• You can not assume that a parameter that is relevant to the WFD is

relevant to the DWD, e.g. some adsorb to sediment and are not relevantfor water. Also they refer to the past and DWD should focus on future

Jane Forslund (EPA, DK)• Background document on parametric values is available but not to the

public. Should be made public.• Increase limit for pesticides above 0.1 µg/l is not acceptable• Advice Slobodnik: take advantage of many projects in river basins under

the provision of WFD

Fawell (UK)• There is already a link between WFD and DWD: there is a provison in the

DWD 4 (1) a . saying that no substances should be present that could beharmful to human health. Common guidance is needed, organisations likeWEKNOW network might assist

Andrea Torok (National Institute for Environmental Health, Hungary)• Regarding cyanotoxins it is not enough to monitor nutrients as blooms of

algae also occur in low nutrient surface waters. Toxins should bemonitored irrespective of the nutrient load in both surface water anddrinking water.

Frenzl, (private lab Austria)


• The detection limits should be clarified, important for comparability

Leena Hiisvirta (Ministry of Health Finland)• Cl/SO4 are important for corrosion control and should be retained in the

DWD• Voting by audience: majority is against deletion of those two parameters.

Maria Benoliel (Lisbon Waterworks)• How to measure total indicative dose. Note: will be clarified soon in the

Annex of the DWD• Agrees with change from limit of detection to limit of quantification• Requirements for the performance of methods for vinylchloride,

epichlorohydrine, and acrylamide are needed. Answer: will be addressedin EAS

• Statements like ‘No abnormal change’ and ‘acceptable for consumers’should be quantified (e.g. turbidity)

• Supports health based limits for pesticides• Agrees with remark by Torok on cyanotoxins. Should be under control of

WFD.

Torbjörn Lindberg (Regulator Sweden)• Support to keep Chloride in the DWD. Chloride salts are added for ice

control on roads and this might contaminate groundwater.

Ingrid Chorus (UBA, D)• Standard on cyanotoxins needed, both in WFD and DWD. Just looking

tatnutrients is not sufficient• Some discussion with the conclusion that a working group is proposed to

work on this topic

Tom Leahy (Dublin City Council, Ireland)• Socio-economic impact of changes in the DWD should be quantified

Dieter Jenewein (Health Ministry Austria)• There should be a limit for chlorite of 0.2 mg/l (not 0.8, WHO)• Limit for Uranium should be 9 µg /l (not 15 µg, WHO)• Keep limit for Cl/SO4• Keep limit for pesticides, as drinking water should be wholesome and

clean• Fluoride addition is not allowed in Austria. Keep the limit at 1.5• Supports the idea of a working group to elaborate on cyanotoxins

Pocock (Voice of Irish concern for the environment)• Gives information on fluorosis, see separate paper• Lower limit to 1 mg/l has no meaning because it is not protective against

dental fluorosis• Limit for total fluoride should be 0.5 mg/l


Jane Jones (National Pure Water Association, UK)• DWD should not allow fluoridisation as fluoride is no water treatment

chemical.• Total exposure to fluoride should be investigated and taken into account

before any fluoride supplementation.

Lecloux (CEFIC)• Criterium 9 is not science based, it is consumer perception

Hans-Jorg Kersten (Eurogyspum, D)• Welcomes minimum limit for calcium• Pleads to consider the ratio between Ca and sulphate. Reaction from

panel: it is the balance between Ca, Sulphate and HCO3 that determinescorrosivity

Stanislaw Witczak (Krakow University, Poland)• Recommends to take into account the draft Groundwater Directive

Steve White (Thames Water)• No revision is needed, the Annexes offer sufficient flexiblity to reduce or

cease monitoring for parameters of no concern

Francis Bourgine (France)• Why minimum calcium for treated (desalinated) water and not for all soft

natural watersReaction from the panel of experts: desalinated water undergoes alreadyradical treatment, contrary to natural water, and is extreme case of softwater, as it does not contain any minerals.

ECPA• Welcomes health based limit for pesticides

Office Public Health (France)• Need confidence of consumer, welcomes difference between health based

and technically based limits (transparency)• Why not talk about lead?

Owen Hydes (UK)• Little experience with new DWD, do not apply too many changes too

soon• Should add Radon as an important parameter

John Nuttel, European Copper• What does the position paper recommend on copper. Answer: review

limit for copper and restrict use of copper pipes in critical areas (by MS)• Recommends to follow the new WHO guideline in review.


3.2.3 Session on microbiologyThe position paper of the expert group on microbiological parameters waspresented by the rapporteur: Dave Drury. The presentation is included inAnnex IV of this report.

3.2.4 Synthesis of the discussion on microbiology

Comparability of analytical methods• Gertjan Medema (Kiwa, NL): 2.1 Equivalence testing should not be

restricted to single Member States.Response prep. group: This statement is intended to stimulate multi-MStesting, not to rule out single-MS testing.

• David Sartory (UK): UK and The Netherlands have already done somestudies, the results should be shared.

• Mihaly Kadar (HU): ISO 17994 is a good tool to provide guidance tolaboratories.

• Gertjan Medema/David Sartory (NL/UK): How are the terms“equivalence” and “better” defined? Better in terms of more detected E.coli, or more specific?

• Jean-Marie Delattre (Institut Pasteur, F): Better means quicker, lessexpensive, significantly better recovery. A better recovery means that it isanother parameter. MS can adopt more stringent regulations.

• Jean-Paul Lickes (Ministère de l’interieur, Lux): From experience in thefield it was concluded that the ISO-method is not suitable. It takes a longtime to get a result, time which is not available in case of a real problem.72 Hours is too long, a results needs to be available within 24 hours.

• David Sartory (UK): The definition of equivalence should not be too strictto keep things workable; it was deliberately done so in the DWD.

• Tommy Slapokas (National Food Administration, S): A main point is“may be accepted”; this means that it will not necessarily be accepted, itdepends on the situation in the MS. Adopting methods with a muchhigher recovery may cause problems in some MS.

• Hungary: Some uniform guidance for single laboratories on howvalidation can be done?

Risk management approach• Jean Duchemin (Agence de L’eau Seine-Normandie, F): The importance of

qualitative turbidity as indicator for viruses as a complementary methodis stressed.

• Anna Maria de Roda Husman (RIVM, NL): We need a very good studyfirst to find a good correlation between turbidity and viruses.

• David Sartory (UK): We should not try to look for a universal indicator.• Beate Hambsch (TZW, D): Turbidity is not a good parameter instead of

microbiological monitoring, but it can be used as a process parameter.• Beate Hambsch (TZW, D): The phrase “Risk assessment may overlook

problems” is unclear.• University of Paris (F): Internal plumbing systems can have influence on

microbiological aspects. How can you cover “from source to tap” in watersafety plans?


• Gertjan Medema (Kiwa, NL): 3.6.2.2 “Misuse of data” is unclear.• Regulator Belgium: Risk management approach put much more burden

on the regulators? Now the regulator can approach each water utilityuniformly, but with th risk approach the regulator needs to approach theutilities more individually.Response of chairman: We need training for all stakeholders (chair)

• Ministry of Health (P): 3.5 What do we mean exactly by risk managementfor microbiological parameters? How must sampling be adapted to riskassessment?

• Tommy Slapokas (National Food Administration, S): End-point samplingfrquencies are very low and will always be “cosmetics for the public”. Weneed to look for something in early steps, therefore, the risk managementapproach is a good method.

• (UK): Part of the risk approach is that you monitor your treatmentprocess. If you do that with continuous monitoring rather than sampling,you should be able to guarantee water quality for your customers.

Molecular methods• Michel Gibert (Veolia, F): A bridge between research and regulation is

needed, here is a role for expert working groups.• Beate Hambsch (TZW, D): It is clear that methods are not yet usable for

routine checks that should be made clear in the statements. Molecularmethods can be used for critical control poin measurements, althoughthere are limitations.

• Anna Maria de Roda Husman (RIVM, NL): PCR is very useful fornoroviruses, they would never have been found without PCR. However,there are drawbacks, too, such as interpretation problems.

• David Sartory (UK): Do we use molecular methods for indicatorparameters, or to find (new) hazardous pathogens?

• Hans Berkhuizen (VEWIN, NL): How do you communicate to the publicabout methods that may or may not have been properly validated byestablished methods?

• Jean Duchemin (Agence de L’eau Seine-Normandie, F): Molecularmethods are important to identify risks.

Sampling methods• Sampling methods are described in ISO 19458.• Jean-Marie Delattre (Institut Pasteur, F): There is a lot of discussion about

flushing. If you flush a big volume, you get water from the mains to studywater quality as provided by the distributor. If you do not flush ordisinfect, you can study how people can become infected. In between isthe responsibility of the building owner. The decision was made that itshould be left to each MS to decide. The ISO-standard does not provide a“spare wheel” and studies to review things slow down the process.

• Chairman: The ISO-standard will be delayed, should we rely on it or askthe EC to urge for some research?

• Tommy Slapokas (National Food Administration, S): The situation is notso complicated as described by France, We have references, so we canproceed with the acceptance of the ISO-standard.


• ? (P): Sampling for microbiological parameters should be seen in relationto sampling for lead, copper and nickel. Can all be done in the sameround or not? How do we explain the flushing or not flushing to thepublic?

• Guy Howard (University of Loughborough, UK): The water supplier maynot feel responsible for hygiene education and awareness rising.

• Benedikt Schaefer (UBA, D): 5.5 “changes should be avoided…” isunclear.

Established parameters• ? (LUX): Colony count is underestimated as an indicative parameter for

detecting problems in your system.• Gertjan Medema (Kiwa, NL): Coliforms can be deleted from the DWD,

because they have no additional value to E. coli.• ? (HU): Is Coliform in the DWD the same as “faecal indicator”? Should

the definition be changed?• David Sartory (UK): We should get rid of coliforms in distribution, but

not as a check on treatment efficiency.• ? (LUX): Luxembourg disagrees with NL and UK, but agrees with HU.• Tommy Slapokas (National Food Administration, S): We need to make a

decision on C. perfingens. The third option was discussed at an ISO-meeting, but it means a step backwards in the standardisation process.We have to look at it as a monitor parameters, not a pathogen.

• Jean-Marie DeLattre (Institut Pasteur, F): Option 3 is supported. The onlyaltenative is to stop with C. perfringens. If we withdraw a parameter fromthe DWD every 5 years, where do we go? We must use indicators whenwe can, although there is no universal indicator. We should not withdrawa parameter because the method is not ok.

• David Sartory (UK): We have strong reservations about replacing oneparameter with another. C. perfingens does not necessarily have to bereplaced by something else.

• Corrie Allaert (University of Lleida, ES): We agree to stop with C.perfingens and go back to SRC.

• Beate Hambsch (TZW, D): C. perfingens is in the indicator-list. (D)

New (non-established) parameters• Hans Berkhuizen (VEWIN, NL), Hans Niemann (JRC), We agree with the

statement that there is no need to take up Legionella as a parameter in theDWD. The arguments for not taking up Legionella differ.

• Hans Berkhuizen (VEWIN, NL): Do not say that Legionella is not aproblem, because it is. It is very important for drinking water quality andthere is a relation between water quality and growth.

• Hans Niemann (JRC): Legionella outbreaks cannot be prevented, butmaybe we can reduce the risk.

• Plumbing branch (D): Plumbing and inside installations are not in theDWD. The people involved in plumbing should be properly qualified, toreduce the risks of Legionella. Also critical points like hospitals could bechecked frequently to reduce the risks.


• ? (P): For “new” parameters like Legionella and cyanobacteria we need nolimit values, it is better to handle them through a risk managementapproach.

• David Sartory (UK): Better ways to use bacteriophage as an indicator forenteroviruses are coming on-line now.

• Jean Duchemin (Agence de L’eau Seine-Normandie, F): The relationbacteriophage – E. coli is better than bacteriophage – enteroviruses.

• Gertjan Medema (Kiwa, NL): Risk assessment needs an end-point. Whatis safe? WHO talks about reference pathogens. Can we agree in the EUwhich pathogens we need to choose as a reference? (NL)

• ?: It is surprising to see that Cryptosporidium is not mentioned instatement 7.1.This will be amended.

3.2.5 Session on endocrine disrupting compoundsThe position paper of the expert group on endocrine disrupting compoundswas presented by the rapporteur: Leo Puijker. The presentation is included inAnnex V of this report.

3.2.6 Synthesis of the discussion endocrine disrupting compounds

Environmetal concern of EDCs:Chairman: different pollutants have been banned last years, but not becauseof their endocrine effects.Andrea Wenzel (Fraunhofer Institute) : organotin compounds are anexception (imposex).Simon Webb (B): just industrial compounds are subject for regulation, thereis no attention for natural steroid hormones.

Regulation DWD:Chairman: no need for individual limits but what about a limit for the sum ofEDCs? Reaction: the main purpose is drinking water without any humanhealth effects, so if we know a safe level, based on toxicological data andbioassays, such a value for the total ED potency good be used. Knowing thepotential effects of an individual compound in tests, limits can also bederived for individual compounds.

Jean Duchemin (France, Agence de l’Eau Seine Normandie):The same method is used by setting limits for dioxins in food: based on therelative effects (toxicity) of the different dioxin-isomers one value iscalculated, the TCDD-equivalent.

Jane Forslund (Environmental Ministry, DK): there is a need for (a method )to summarise effects, therefore we need a method to identify and measurethis total effect.


Chairman: the sooner we get those tests, the better. In future limit values forthe total potential endocrine effects can be proposed, if validated bioassaysare available

Leo Puijker: the problem is complex which means that more endocrinedisrupting effects must be taken into account and more tests are needed forall these different effects.

Univ. of Paris: Not only estrogenic effects must be taken into account, but alsoandrogenic effects and disruption of thyroid hormones, in the same way asgenotoxicity and mutagenicity are subject of studies.

Development of analytical methods and bioassays:Jean Francois Loret (Ondeo, France):There is a need for screening methods to measure potential effects (andsources) and to investigate the removal of these effects in water treatmentprocesses.Routinely measurement of all individual EDCs is impossible.

Removal of EDCs:Jean Duchemin (Agence de l’Eau Seine Normandie): bioassays are a goodtool to investigate treatability of EDCs.Univ. of Paris: drinking water treatment performance has not beensufficiently assessed, just for some compounds information is available.Effects of byproducts from e.g. ozonation and chlorination are unknown.

Risk assessment of EDC s:John Fawell (private consultant on behalf of UK) : the combination of in vitro-assays, in vivo-assays and the occurrence of individual EDCs allows toperform risk assessment. This is very valuable to show effects of treatmentprocesses. The frequently exposure of humans to natural hormones must betaken into account. However, we don’t know everything there is to knowand research needs to be targeted and focussed on ways to be able to performproper risk assessment.Jean Duchemin France( Agence de l’Eau Seine Normandie) : agrees that foodis a main exposure route and is more important than drinking water.Chairman: it is important to find the main exposure routes for EDCs.Claudia Roncancio Pena (DG ENV European Commission): identification ofknowledge caps and more research on exposure is needed in future.Simon Webb (Belgium): Cion felt satisfied with toxicological data available,however not for all compounds data are available.

General remarks:Tony Lloyd (DWI UK): compliments for and agreement with the clearstatements in the position paper.Janne Forslund (Environmental Ministry, DK): it is disappointing that afteryears of study, we still are not sure and know enough to evaluate the risks ofEDCs in drinking water. Consumers still want this information.


3.2.7 Session on risk approach for drinking waterThe position paper of the expert group on risk approach for drinking waterwas presented by the rapporteur: Oliver Schmoll. The presentation isincluded in Annex VI of this report.

3.2.8 Synthesis of the discussionon risk approach for drinking water

Main discussion points- Broad support of the overall approach and the core principles of the

RA/RM approach

- EU should provide an overall framework of core principles and aknowledge base of health based targets

- Member States should implement programmes and plans that areconsistent with the overall framework

- Need for clear definition of responsibilities at EU, Member States andlocal level

- Need to define responsibility for the control of the catchment andhousehold installations

- Cost reduction is not the driving force but optimising costeffectiveness is the ultimate goal

- RA/RM is supportive of establishing targets for chemicals andmicrobiological parameters and provides clarity regarding the basis ofstandard established

- RA/RM will not solve old problems (e.g. Pb), but offers a processapproach including all relevant stakeholders leading towards possiblenew solutions and transparency of responsibilities

Other discussion points- Need to maintain end-product testing

- Question was raised whether other policy instruments than legislationcan achieve the implementation of the approach. One responseexpressed the need for legislation particularly with respect to thesetting of health based targets and to surveillance requirements

- Question was raised whether substantial changes are indeed neededin the text of the DWD, or whether it isn’t quite simple to introducethe WSP approach quite briefly and leave the rest to the subsidiaryprinciple

- Reinforcement on the aspect of integrating household installationmaterials and plumbing into the WSP (WG response is that this issomething where water supplier takes initiative, perhaps by


providing households with generic guidance in relation to the waterquality)

- The implementation of a risk based approach in the DWD will requirea level of risk that is considered tolerable, to be set by the EuropeanCommission. A level of tolerable risk is already used for thecarcinogens in the DWD, but this level cannot be directly applied tomicrobiological risks. The use of a generic burden-of–disease measure(such as Disability adjusted life years, DALY’s), is used by WHO forthis purpose

- Water quality targets: “tolerable” is on the societal level; “acceptable”is perceived as related to the individual and opens the debate ofwhether any level of risk is acceptable

- Need for harmonisation with the WFD

- Potential benefit for small systems was emphasised as usuallysampling frequencies are much too low

- Quality of construction products should become integral part of theWSP, i.e. by requirements for suppliers to inform house owners aboutrisks of certain materials

- Difficulty to integrate catchment protection as generic guidance willnot address problems with behaviours of single farmers´

- Challenge in communication to explain the benefits of the approach tothe public.

3.2.9 Session on materials and substances in contact with drinking waterThe position paper of the expert group on materials and substances in contactwith drinking water was presented by the rapporteur: Mike Fielding. Thepresentation is included in Annex VII of this report.

3.2.10 Synthesis of the discussion on materials and substances in contact withdrinking water

Questions for clarification:

George Glasser, UK: Fluoridation of drinking water poses a hazard becausefluoridation chemicals are toxic.

Birgit Mendel, Germany: Germany has assumed that the DWD covers hotwater applications. Therefore hot water is included in the EAS.

Peter Arens, Germany: How are materials being put on the European Market?Protection of the market is needed against unsatisfactory products comingfrom outside the EU, e.g. China etc.

Reply of Georgios KatsarakisProducts have to fulfil the Essential Requirements of the CPD.


Products have to comply with the product standards.It is the responsibility of member state market surveillance authorities to check.

Georgios KatsarakisNote that Essential Requirements relate to construction works not products.

Discussion

Michel Rapinat, Position of Eureau• Eureau supports development of the EAS.• DG Environment should consider removing from the DWD, chemicals

which are specific for materials and treatment chemicals, i.e.controlled by product specification.

• Metals in contact with drinking water are difficult area. In the domesticnetwork less has been done on control and management than in thewater supply system. Not only is the selection of materials importantbut also design and maintenance of the plumbing installation.

• Design and maintenance of domestic installation is not covered by theEAS and the DWD, but by Member States themselves.

Birgit Mendel• Reference to technical Guidelines have to be included in DWD (like in

Germany)

Dominique Oliver, CEN• What is the position of the preparatory group on hot water? In the

development of the EAS hot water has to be included. This needs to beconsidered in choosing test conditions of materials (EN-standards).

Reply Georgios Katsarakis and Birgit MendelThe EAS deals with hot water applications.

Christian Legros, Belgium• prevention of backflow from domestic installation to the public network

has to be regulated. At the moment it is not regulated with enoughprecision in each Member State.

• new DWD: obligation on regulation on backflow prevention

Remark Peter Arens, GermanyIn Europe available: EN 1006 and EN7770 deal with the design of plumbinginstallation.In a new DWD there should be associated technical guidelines.

Francois Depagne, BelgiumHot water is consumed, e.g. via cooking, therefore it has to be considered asdrinking water and covered by the DWD.

Sophie Herault, France• The CPD covers raw water distribution materials.


Riku Vahala, Finland• In practise the inclusion of raw water is impractical. Proposal is to use any

material while Member States have to decide on it. Some materials arenot covered yet by the EAS, such as wood.

Tony Newson, Belgium• How to test raw water materials?

Stephan Merckx, The Netherlands• Positive List is a ‘moving’ list. Changes will sometimes take place within

one year. Therefore any combination with the DWD will be difficult.

ReplyMike Fielding: It is inappropriate to combine a positive list with the DWD, althoughthere should be reference in the DWD.Wennemar Cramer: a quick practical procedure is necessary in order to enablemodification of the list as necessary.

Treatment chemicals

Birgit MendelThree substances (parameters) in DWD refer to materials and chemicals.These have to be removed from the DWD. These substances are covered bythe acceptance procedures (EAS and CEN ENs). What is the sense of havingDWD parameters as well? Further, there are no methods for their analysis indrinking water and no good idea on how to handle them in practise.

Michel Rapinat, Eureau• It is more efficient to control these substances via materials and chemicals

specifications than via the DWD.

Christian Legros, Belgium• The EAS covers substances and deals attestation on conformity. A similar

approach to water treatment [?] chemicals is wantedGood and efficient mechanism by working with Positive Lists andComposition Lists. Is it meant to develop a list with treatment chemical?

Reply Georgios Katsarakis• The EAS deals with construction products. The CPD does not cover treatment

chemicals.• The new proposal of EC DG Enterprise, REACH (Registration, evaluation and

authorisation of chemicals) may be relevant.

Dominique Oliver:• Better balance between monitoring and system management (risk

assessment) is necessary.

Peter Arens, Germany


• Chemicals used in treatment plants have EN’s. Can these chemicals alsobe used in household applications, for example, disinfectants?

Bob Tanner• Question on lab testing and product certification (?),• Clarification is needed if hot water is potable water and covered by the

DWD and EAS.• Information is needed on whether products such as mixer-taps used in the

kitchen are covered and if so how they should be tested. Also to theapplication of materials in dish washers, swimming pools, etc

• Why is bottled water mentioned in the DWD and therefore covered by theDWD?

• Is it left to standardisers how to control and test materials?

Comment Birgit MendelIn the case of bottled water, the DWD only refers to drinking water quality prior tobeing put into bottles.

Michel Rapinat• Standards to test hot water applications need to be developed. Hot water

applications are part of the plumbing systems.

Christian Legros• Is rainwater to be part of the EAS. In Belgium rainwater can be used in

household applications as a feed for hot water production.

??Attestation of conformity of treatment chemicals is important• suggestion to DWD writers• DG ENV need to look at this to see what changes are necessary• DG ENV has to propose something.

Bob Tanner, UK• The EAS gives excellent possibilities to control the materials and not to

control the DWD.

Dominique Oliver, France:• What are the regulations on Notified Bodies in testing materials?

Reply of Georgios KatsarakisEAS: group of notified bodies, notified by MS.An EAS manual will be prepared and adopted by the Regulators.

Peter Cutler, United Kingdom (Nickel Development Association)• Composition of treatment chemicals is described in the EN standard.

How is dosing and maximum concentrations covered? What arecodes of practise? Terms of approval?

• Operation guidelines and codes of practise are needed. Can be given inGuidance documents.


Final comment Georgios Katsarakis:• Do not wait to 2006 (day 1 of EAS implementation?) to approve your

materials!


4 Drinking Water Seminar (day 2)

4.1 Presentation WHO Jamie BartramMr Jamie Bartram of WHO presented the latest developments in the WHOapproach to the quality of drinking water. The presentation is included inAnnex VIII of this report.

4.2 Amended presentations preparatory groupsOn the basis of the presentation and discussions in the parallel sessions of day1 of the seminar, the rapporteurs of the five expert groups adapted theirpresentations. A short presentation, including these changes were presentedin the plenary session on day 2.

4.2.1 Amended presentation chemical parametersThe amended presentation on chemical parameters was presented by therapporteur of the expert group: Frantisek Kozisek. The amended presentation(shorter version of the original presentation) is included in Annex IX of thisreport.

4.2.2 Synthesis of the plenary discussion on chemical parameters

After the presentation of the rapporteur, EUREAU presented its position onthe subject. The complete position of EUREAU on chemical parameters can befound in the position paper of EUREAU position paper (Annex XIV). Theissues can be summarised as follows:

• EUREAU generally supports the position paper, but has a number ofcomments

• Any changes to the DWD should only be made when it results in asignificant improvement

• The use of criteria is strongly supported• Changes should be based on available scientific evidence i.e. WHO

and CSTEE, but should be practical, achievable taking into accountproper timescale and costs involved

• Only cold water supplied for normal domestic uses should becovered

• Essential parameters across Europe should be included and MemberStates should add national standards (subsidiarity principle)

• Socio-economic aspects have to be taken into account• EUREAU supports the need for research for individual parameters• A third category of parameters for monitoring purposes only might be

useful• Product specified parameters such as acryl amide, epichlorohydrin

and vinyl chloride are to be covered by the EAS and can then bedeleted from the DWD


• There is no need for more DBP in the DWD other than chlorite (alsosee WHO)

• HAA is not necessary nor are other chlorine by products , followWHO with respect to these parameters

• Uranium parameter there is a need for more information and research• The proposed removal of cyanide and mercury from the DWD is

supported• As for pesticides and parametric values, EUREAU states that such

substances should not be present in drinking water. A link is to bemade with Article 7 of the WFD.

• The view on copper, fluoride and antimony is supported.

Other comments and opinions• Robert Pocock (Voice of Irish concern for the environment): there is no

science behind artificial fluoridation of drinking water and addition ofthis contaminant is inconsistent with Article 4 of the Directive.Children are not protected and artificial fluoridation infringes withhuman rights. No medication should be given through drinkingwater, the DWD needs to protect human beings. The position paperof Voice of Irish Concern for the Environment can be found in AnnexXV.

• John Fawell (private consultant on behalf of the UK): Member Statesare not yet used to the new DWD 98/83/EC.

• Jean Duchemin (France): No harmful substances should be present indrinking water. Global toxicity bio-assays should be used to look fore.g. EDC’s and when there is a problem then more detailed researchshould follow.

• Janne Forslund (Ministry of the Environment Denmark): Anintegrated approach is need for all possible sources that can affectwater quality, e.g. source, materials, treatment.

• Jane Jones (Pure Water Association) UK: Ask consent before addingany chemicals to water. The position paper of the Pure waterAssociation can be found in Annex XVI.

• Steve White Thames Water: Changes in the DWD take a long time.Greater transparency on standards in the current DWD is needed e.g.radio-activity and also on monitoring. No formal changes are needednow. There are non-relevant parameters for specific Member States.

• Hans-Jorg Kersten (Eurogypsum) Sulphate has a positive effect onhuman health.

• Stanislaw Witczak (University of Krakow Poland) With respect to theharmonisation with the WFD, most water is produced fromgroundwater so there is also need for a link with the GWD.

• Janne Heimann Environmental Institute Germany: The current valuesfor pesticides 0.1 ug/l should be retained for the protection of humanhealth and for the protection of ground- and surface water.

• R. Pocock Caution is needed for the parametric value of aluminium asvery little is known about this parameters, but what is known is veryworrying. The level should be reduced to 0.05 mg/l; values should bebased on science.


• Steve White Thames Water: The attendance at the seminar is toonarrow, there are no consumers. Do they want to pay for the changes?Also other stakeholders are involved and affected in catchment areas.All parties need to be involved.

• Daniel Villessot (EUREAU): The parametric value or lead of 10 ug/lhas to be complied with in 2013, which costs 12 billion euros in Franceto comply. Can WHO investigate the interim value of 25 ug/l and theeffects of this interim value,, taking into account all other sources oflead.

• EU Chemical Industry: There is a statement on simple watertreatment, appropriate treatment and efficiency of treatment and thelink between WFD and DWD. Simple treatment might be a goal but isnot easily achieved. How will the Commission achieve harmonisationbetween the WFD and the DWD?

• Steve White Thames Water: The surface water for drinking waterdirective 75/440/EEC already gives a requirement for water treatmentin relation to the source water quality, so there is nothing new there.

• Michel Gibert Veolia: Include requirements in the DWD for what isneeded from the WFD. But it is still possible that we might not getfrom the WFD what we need from the point of view of drinking waterproduction.

4.2.3 Amended presentation microbiologyThe amended presentation on microbiological parameters was presented bythe rapporteur of the expert group: Dave Drury. The amended presentation(shorter version of the original presentation) is included in Annex X of thisreport.

4.2.4 Synthesis of the plenary discussion on microbiologyAfter the presentation of the rapporteur, EUREAU presented its position onthe subject. The complete position of EUREAU on microbiological parameterscan be found in the position paper of EUREAU (Annex XIV). The issues canbe summarised as follows:• There is general support from EUREAU for the recommendations in the

position paper.• EUREAU supports molecular methods, but they need to be practical and

affordable.• There is no support for the increase of sampling frequencies; the risk

based approach is regarded as much more efficient.• The removal of Clostridium perfringens from the Directive is supported.• Legionella is important and need to be controlled, among others through

the control of plumbing installations.

Additional comments from the audience:• Ingrid Chorus (UBA, D): What are the concerns for the mis-use of data?

Verification is needed.


• Birgit Mendel (Federal Ministry of Health and Social Security, D): Colonycounts should be included in the DWD again, a limit value should beconsidered, and also the frequency should be reviewed.

• Torbjörn Lindbergh (National Food Administration, S): The definitions of“equivalent”, “better” and “as reliable as” need clarification.

• Dieter Jenewein (OVWG, AU): Validation should be done using the ISO-methods. Either Clostridium perfringens or spores should be kept as aparameter in the DWD. We need to track bacteriophages in the treatmentprocess.

• Jean Duchemin (Agence de l’Eau Seine-Normandie, F): We support amore qualitative approach of turbidity, as an indicator for viruses.

• Tommy Slapokas (National Food Administration, S): Use presumptiveClostridium perfringens as an audit parameter. The analysis takes a longtime to confirm the results, we need something faster and better.

• Gertjan Medema (Kiwa, NL): In the risk assessment approach there is aneed for verification if the process is under control.We know that E.colitesting alone is not sufficient for this and we need additionalmicrobiological indicators (such as phages and Clostridium spores) toverify the efficacy of our treatment processes for viruses and protozoa.

• David Holt (UKWIR, UK): The parameters monitored do not necessarilyinclude all micro-organisms of interest.

• Torbjörn Lindbergh (National Food Administration, S): In Swedenenteroviruses are important, Cryptosporidium is not. This may be dfferentin other Member States.

• Jean Duchemin (Agence de l’Eau Seine-Normandie, F): Is it a good idea toboth flush and disinfect systematically when sampling at the tap?Response David Drury: This very much depends on the purpose of yoursampling.

• Steve White (Thames Water Utilities, UK): We need to include operatorsin the further discussion about sampling at the consumer’s tap.

• Anna-Maria de Roda Husman (RIVM, NL): Molecular methods should beincluded in the DWD, mainly for the detection of infectious viruses andnew pathogens.

• Philippe Harmant (Ministry of Health, F): How can criticial end-points forLegionella be identified for the Water Safety Plan?

• Corrie Allaert (University of Lleida, ES): In Spain different methods thandescribed in the DWD are used for E. coli and coliforms.

4.2.5 Amended presentation on Endocrine disrupting compoundsThe amended presentetation on endocrine disrupting compounds waspresented by the rapporteur of the expert group: Leo Puijker. The amendedpresentation (shorter version of the original presentation) is included inAnnex XI of this report.

4.2.6 Synthesis of the plenary discussion on endocrine disrupting compoundes


After the presentation of the rapporteur, EUREAU presented its position onthe subject. The complete position of EUREAU on endocrine disruptingcompounds can be found in EUREAU’position paper (Annex XIV) The issuescan be summarised as follows:

• EUREAU endorses the recommendations from the preparatory group• It is not necessary and practical to include parameter(s) for endocrine

disrupting substances in the DWD• EUREAU supports further research into EDC’s make sure drinking

water is kept safe• However, research into EDC’s should be in perspective with the

exposure risk, as the main route for exposure to these substances isthrough food, with a minor contribution from drinking water.

Other opinions and remarks made during the discussion:• Christian Legros Belgaqua: avoid using the word waste water

treatment plant as the main source of EDC’s, it is not the plant but thesewage that is a source.

• John Fawell (private consultant on behalf of UK): Warns for themisinterpretation of results from bio-assays, that measure activityrather than effects, which could easily be wrongly interpreted andcommunicated. Bio-assay results should not be used as a standard tojudge drinking water in Member States. There is no accuracy betweenlaboratories. Bio-assays are screening tools and indicators and notregulatory tools.

• Michel Gibert (Veolia, France): Bio-assays are only a way to avoidendless lists of substances, but a risk evaluation is quite a differentstory. Removal in water treatment provides good information ofexposure but not of risk.

• Institute for food safety (Denmark) There is no need for moretoxicological data, use existing data and limits (e.g. for bisphenol,.organotin compounds and phthalates). Look at WHO for limits. Thereis a need for knowledge about general exposure through drinkingwater and information on multiple exposures and the associated risk.

• Daniel Villessot (EUREAU): Water resources are outside the scope ofthe DWD. Recommend Member States to prevent mis-use ofchemicals that pollute water sources. In general a reduction in use inall compartments and not only in drinking water. Care is needed inusing categories of treatment in the position papers that mentionconventional, adequate and advanced treatment, because thesecategories do not mean the same thing in various countries.Borderlines are not clear with respect to EDC’s . Definitions should beclear.

• Steve White (Thames Water): A better integration between WFD andDWD is needed. Some substances are natural or are pharmaceuticalse.g. the contraceptive pill, what can water suppliers do to preventsuch substances?

• European Crop Protection Association. Extensive treatment willremove not only pesticides, but also EDC’s and other compounds.


4.2.7 Amended presentation risk approach for drinking waterThe amended presentation on risk approach for drinking water waspresented by the rapporteur of the expert group: Oliver Schmoll. Theamended presentation (shorter version of the original presentation) isincluded in Annex XII of this report.

4.2.8 Synthesis of the plenary discussion risk approach for drinking waterAfter the presentation of the rapporteur, EUREAU presented its position onthe subject. The complete position of EUREAU on a risk approach fordrinking water can be found in EUREAU’s position paper (Annex XIV) Theissues can be summarised as follows:

• EUREAU firmly supports a risk approach for drinking water, manywater suppliers already use risk approach

• Clear definitions are needed• All stakeholders should be involved not only water suppliers, but also

health administrations and water source managers• Do not use HACCP but WSP• For distribution systems there is a need for a matrix to identify critical

contamination points• QMS should not be a prerequisite for WSP• Only health based targets should be considered• Training of all stakeholders is necessary• Verification has to be done independently• Approach with care and allow sufficient time.

Position WECF Romania (Mrs Michaela Vasilescu)The organisation supports a risk approach to drinking water. There is a needfor training of staff. If this approach results in supplementary costs forconsumers it is not acceptable, the water price in Romania is already too high.

Main discussion points- Prioritisation of RA/RM approaches would be very helpful for

accession countries

- Need for stop thinking in HACCP terminology. HACCP can be toonarrow and fails to consider the multiple barrier principle.Distribution systems, whilst difficult, can also be brought into theRA/RM approach.

- Endorsement of need to “rewrite the Directive” to accommodate forWSP: example for EU, national and local level of responsibility: e.g.aesthetical parameters which typically should be local responsibility(keep or gain consumer confidence), whereas assessment of e.g. EDCshould happen at EU or national level.

- Support of pilot projects. Need for collecting such experience.Suggestion that commission should facilitate exchange of experience


- Be careful about “new management instruments” in times of intensediscussions about deregulation and cost reduction.

- Key approach and way forward in the way we think about drinkingwater quality. Caution in relation to health-based targets – need to beused with great care in the way they are developed. Theoretical riskassessments don’t automatically achieve a health benefit (actual risksare often much lower).

- RA/RM should assist finding cost-effective solutions, maximumbenefit in minimising risks

- Communication issue: credibility gap with consumers is feared if wego forward and explain the limitations we now see with our end-product based approach – taking safety further forward rather thancondemning the last century of success

- No need for putting the approach into the Directive now – keypriority now is to start collecting experience, to run pilot trials, to use100 years of experience. Useful approach for long-term assetmanagement of water sector. Deregulation propositions, introducingmarket mechanisms into management of water sector – has nothing todo with the approach taken here

- Concerns about how it will work and which way of implementationwould be best. Guidelines for small supplies are needed. Feelsreminded of start of HACCP in food sector – concerns at first, nowproven to be very useful.

- Endorses the RA/RM approach. Good for “regulator parameters” and“new parameters”. Accidents and problems happen, RA is moreappropriate approach to address them. Culture change necessary –training and research important. Pilot projects important.

- Any RA has problems with gaps of knowledge – don’t only do riskassessment, measurements (i.e. verification) is also crucial. Concern ofoverlooking processes in RA – which is why comprehensivemonitoring is necessary.

- Reference level of risk (or safety?) is needed and should be decidedupon at EU or national level. DALY approach (from health caresetting).

- Concept of WSP is to go down to a very local level in the choice ofcontrols. Time scales of compliance by catchment control will bedifferent from those by a more conventional – e.g. treatment approach.Will the regulator accept such a delay in favour of a more cost-effective, but longer-term intervention?

- EU chemical industry: (i) cost-effectiveness question – importance oflooking at the full chain, e.g. including improvement of STP. (ii)Doesn’t understand difference between WSP and ISO 9001. (iii) Do weknow today how much disease is due to DW in Europe, and whichkind of improvement would we get with the new proposal.


- Response to concern about new member states’ ability to follow thischange: on behalf of Czech Republic – process started many years agoand is well developed by the point of accession, and maybe even morestringent than in “old Europe”. Are really looking forward to the newversions (e.g. bathing water directive), not eager to implement the oldones. Yes, Czech Republic would be ready to endorse WSP.

- How do we make sure that the outcome of this session reaches WFDgroup? The other way round didn’t work so well.

4.2.9 Amended presentation material and substances in contact with drinkingwaterThe amended presentation on materials and substances in contact withdrinking water was presented by the rapporteur of the expert group: MikeFielding. The amended presentation (shorter version of the originalpresentation) is included in Annex XIII of this report.

4.2.10 Synthesis of the plenary discussion on material and substances in contactwith drinking waterAfter the presentation of the rapporteur, EUREAU presented its position onthe subject. The complete position of EUREAU on materials and substances incontact with drinking water can be found in EUREAU’s position paper(Annex XIV) The issues can be summarised as follows:

Materials• EUREAU supports development of the EAS.• Implementation date for EAS (2006) is too optimistic; Notified Bodies

need to be set up and operational.• The EAS is a useful tool to control installations inside buildings• The main problems in existing installations are not covered by the EAS• Maintenance and usage of installations need EU standardisation• There must be strict consistency of Notified Bodies• And equal practise, good practise of Notified Bodies• Who will monitor the Notified Bodies

Treatment chemicals• EUREAU fully support of recommendations• a quality assurance scheme covering delivery of chemicals is needed.

Now the control of chemicals is at the point of use.• attestation of conformity is important.• hot water, must be treated as heated water prepared from drinking

water. Only the drinking water needs to meet the requirements of theDWD.

• Raw water: no opinion.

Reply Mike Fielding


• EAS will describe transparent harmonised operating principles for theNotified Bodies. The group of NBs has recently started discussing itsoperational procedures.

Reply Georgios Katsarakis• In DG Ent. there is some concern over the capacity of NBs to carry out the

requirements of the EAS. It is a very important issue.• At the moment there is exchange of knowledge between Regulators and

Notified Bodies.• There needs to be a sufficient long transitional period

Discussion:

Janne Forslund (Denmark )• Positive List is good starting point to control components in products, but

it is necessary to look not only what is put into the product, but alsowhat comes out

• EN standards for chemicals:• toxicological aspect may be not included• reactions with pipes and chemicals in the drinking water are not included• EN Standards may be have to be adapted• What is contribution of organic chemicals; what comes out from

chemicals, can be measured f.i. by AOC content of the drinking water

Reply Mike Fielding• Migration tests cover what comes out of organic products in contact with

drinking water. The EAS includes GCMS to control unsuspectedsubstances.How to deal with reaction products (particularly in the case of ‘cured’

site applied products) needs further consideration.

Leena Hiisvirta ( Ministry of Social Affairs and Health, Finland)• involved in the implementation of the current DWD:• in current DWD water for personal hygiene is excluded;• in current DWD hot water is excluded.

Philippe Harmant (Health Ministry of France)Including hot water in DWD: solubility of metals: increases with temperature;• different levels my be defined• differentiation in values

Robert Pocock(Ireland)• About 100 standards for chemicals in contact with drinking water are

available; why spend valuable time to develop standards for fluorideaddition, only used in 2% of the total drinking water in Europe.

• Who is providing toxicological data to judge chemicals within the EAS

Michel Rapinat ( EUREAU)


Emphasis of the work has to be on the domestic network, not on the rawwater which is well controlled.

Summary by the rapporteur (Mike Fielding)None of the recommendations presented in the position paper were rejected.Some were more strongly supported and/or discussed than others.There was generally strong support for the EAS and supporting CENstandards and for the system for water treatment chemicals and that thereshould be reference to these in the DWD at the appropriate time.The various terms in the articles (especially 10 and 2) of the current DWDneed clarification. No doubt the people who put together the DWD rememberhow the terms were arrived at and perhaps where the clarification can befound but there was confusion and disagreement about the exact meaning.There seemed to be concern about domestic situations. It was thought thatwhile the water supply side was well controlled many problems can occur indomestic situations where controls and checks can much less rigorous inpractice. Proper design and maintenance was mentioned in addition to otheraspects.There was support for clarifying the scopes of the DWD and the EAS,especially in the area of hot water applications and raw water.There was a feeling that some or all of these matters were perhaps notappropriate for the DWD itself but the idea of guidance documents (coveringthese aspects) to support the DWD (similar to guidance documents thatsupport the CPD) was thought to be a possibility for the future.Strong support was evident for some system of attestation of conformity ofwater treatment chemicals.It was suggested that the REACH (Registration, evaluation and assessment ofchemicals) proposal might assist in the issue of who will provide theassessment of toxicity data for impurities (at the European level) for any newcomplex water treatment chemicalsThere was support for deleting parameters from the DWD that were coveredby product specification, i.e. vinyl chloride, acrylamide and epichlorohydrin,particularly when the EAS is complete and some issues relating to ENs forwater treatment chemicals are resolved. This is justified for these substancesbecause other sources are negligible (but some uncertainty aboutepichlorohydrin from epoxy resins). This is mentioned because it wassuggested outside the meeting that it could be used as a reason for deletingthe copper parameter. However, there are other sources of copper (thoughsmall).


5 Position papers from stakeholders

They were various ways to bring a position paper to the seminar. A limitednumber of stakeholders presented a position paper during the seminar. Therewas also the possibility to enter a position through the dedicated web-site.And finally the European Commission allowed interested parties to sendwritten positions to the Commission by 7th of November 2003. All positionpapers that were received through any of these channels are included in theAnnexes of this report. The contents have been taken into account in chapter6. No written position paper was received from the Romanian women’sorganisation.

The position papers were received and included in the Annexes.


6 Synthesis of main issues and concludingremarks

6.1 IntroductionThis chapter aims to summarise the main deliverables of the Drinking WaterSeminar. The detailed comments and positions can be found in the previouschapters as well as in the Annexes that include all written and electroniccontributions from stakeholders.This chapter follows the five issues identified by the Commission as a basisfor the seminar and includes the following sources:

1. Position papers as prepared by the 5 expert groups2. Comments made during the parallel sessions on day 1 of the seminar3. Comments made during the plenary session on day 2 of the seminar4. Position papers received prior to the DWS, during the DWS and

position papers send to the Commission before the deadline setduring the DWS of 7th of November 2003.

5. Positions and statements received through the talkback boxes on thewebsite.

Text Box on Position papers received

Position papers (total 16) were received from:o EUREAU European water and waste water operatorso Voice of Irish concern for the Environment, Robert Pococko Pure Water Association United Kingdom, Jane Joneso Eurogypsum Dr. Hans-Jörg Kersteno CENORM Be Ima Gómez Lópezo AEAS Spanish Water Supply Associationo Avril Doyle MEPo IWW Germany, Ulrich Borcherso UBA German Dieter Hermanno SWDE, Vincent Laurent, Belgiumo Government Hungary, Mihály Kádáro Nickel Development Institute, Peter Cutlero Direction des Expertises, France, Jean Marie Delattreo Drinking Water Inspectorate, UK, Jeni Colbourneo Danish EPA, Janne Forslundo Government Spain, Mz Elena Ordaz


Text Box on Website input

The five issues are addressed in four sections, which areo General issues concerning drinking water legislationo Changes in the DWD 98/83/ECo Need for guidance, research and working groupso Specific issues.

This summary highlights the main issues discussed and the main outcome ofthe Drinking Water Seminar. It is therefore not a complete overview of allpositions and statements. It is necessary to read this summary together withthe previous chapters that include the full position papers of the preparatorygroups and of the various stakeholders as well as a detailed overview of thediscussions in the parallel and the plenary meetings.

During the seminar, various research topics were identified; they areaddressed in specific sections of this report. They are worth to be consideredby the research community.

6.2 Participation in the DWSSpecial attention was paid to representatives from the adhesion and accessioncountries and EFTA countries, in order to have them participate in the DWS.Attention was also paid to have all stakeholders present such asenvironmental groups and industry, from pipe manufacturers to foodprocessing. In close co-operation with the desk officer, selection criteria wereagreed from which to determine who should be allowed to attend and whoshould not. The selection criteria were transparent and fair. A standard letterfor this purpose, where mention was made of the alternative possibility ofparticipation through the website.

But is was clear early on that even if interest was high in the AccedingCountries, they did not have the financial resources to come to Brussels. Thisnecessitated an important review of the list of invitees and resulted insending out three invitation letters and two follow-up letters to new

The input through the website was very disappointing. A total of 7 boxes onspecific issues were opened, but only 7 comments was received:

o John Nuttall on chemical issueso Women in Europe for a Common future on general issueso Rudolf Ziegelbecker on Chemical issueso Fatima Coimbre on Microbiological issueso Leena Hiisvirta on Chemical issueso M. Maquennehan AFPR on Materials and Substanceso Roy and Carol Smith on Materials and Substances


candidates who had been added to the lists by Kiwa in cooperation with thedesk officer.Despite the significant additional input and attempts to have participationfrom the 10 acceding countries as well as the pre-accession countries Bulgariaand Romania, there was a profound under-representation from these regions.This resulted in a seminar that was dominated by stakeholders from thecurrent 15 Member States. After inquiring it turned out that no representationwas sent because of the relatively high cost involved in travelling to Brusselsand the fact that the stakeholders in the new countries are as yet unfamiliarwith prioritisation and allocation of money to the appropriate activities. Asecond disappointment of the seminar was the fact that despite considerableeffort the largest stakeholder, the European consumer- was highly under-represented. Various national and European consumer organisations as wellas environmental action groups were invited but did not react to multipleinvitations.

6.3 Summary and synthesis of the main outcome of the seminar

6.3.1 Chemical parameters

6.3.1.1 General issues concerning drinking water legislationThere should be more harmonisation between EU water related legislationsuch as the WFD, GWD and the DWD. There is already some reference in the“old” Surface Water Directive 75/440/EEC, where categories of surface waterare distinguished and appropriate treatment levels required are described.From a drinking water point of view raw water for the production ofdrinking water should be of such a quality that relatively “ simple/adequate”treatment will produce water intended for human consumption that:

o Meets the requirements of the DWDo Is wholesome and cleano Complies with Article 4(1)a of the DWD with respect to the absence of

substances and micro-organisms.

A link between the WFD, GWD and DWD is best made throughrecommendations in the DWD, specifying what is needed in other waterlegislation to enable the production of drinking water.Care should be taken in copying parameters such as priority substancesmentioned in the WFD directly into the DWD, as the WFD often refers to thepast (history of contamination), where the DWD refers to the future. Alsosubstances of particular importance for the WFD might adsorb to sedimentand therefore be of little importance for drinking water.

6.3.1.2 Changes in the DWD 98/83/ECThere is a need for transparency on the current parametric values in theDirective. Even though, there is a memorandum of explanation that explainsthe reasoning behind the parametric values, this document is not in the publicdomain as it perhaps should be. In order to propose and justify any changes


in the DWD the reason behind current parametric values (be it health based,precautionary, practical, political compromise or socio-economic), should beknown.Changes in the DWD, if any, should be based on scientific evidence, and oron the opinion of bodies such as WHO and CSTEE. On top of that anychanges should also be considered from a practical point of view, taking intoaccount a reasonable period for change and socio-economic aspects. Otherstakeholders that are often financially involved, such as consumers andpeople operating in the catchment areas should be more closely involved inany changes.

When changes are made to drinking water legislation this should be done onthe basis of transparent and harmonised criteria, such as proposed in theposition paper on chemical parameters. The issue of inclusion of other than“scientific” criteria such as consumers’ perception and consumer confidenceare subject to further discussions.

In general there was agreement that there is no urgent need to change theDWD. The main reason being that there is as yet very little experience withthe implementation of the ‘new’ DWD. Especially the adhesion countries butalso many of the Member States need some time for the implementation andgaining experience. Also the Annexes of the DWD offer sufficient flexibilityfor the coming years.

The DWD should remain restricted to parameters that are essential to protecthuman health in Europe, while leaving Member States free to add parametersat a national level in line with the Subsidiarity principle.A change that might be considered is the omission of the product specifiedparameters (acryl amide, vinyl chloride and epichlorohydrin), once the EAShas become operational.Also on the basis of the Subsidiarity principle the DWD should only apply tocold water as supplied to the consumers for domestic purposes. In the caseMember States would like to widen the scope of the DWD this should be leftto subsidiarity.

6.3.1.3 Need for guidance, research and working groupsA number of areas were identified where the DWD needs more clarificationor where more research is needed. This could either be done throughguidance documents on specific issues, working groups or a combination ofboth. There might be a role for the WEKNOW network. Examples are:

o Sampling and monitoring in generalo Radio-activity parameterso Cyano toxins in raw water and drinking watero Union wide survey on occurrence of uraniumo Radon and daughters, standard or guidance?o Occurrence of chloramination by products and haloacetic acids (Union

wide as was done for THM’s)


6.3.1.4 Specific issueso Fluoride: The position paper of the anti-fluoride lobby proposes a

lower parametric value for situations were artificial fluoride is addedto drinking water for dental reasons. Any addition of fluoride todrinking water is strongly opposed by representatives from the anti-fluoride lobby and a Member of the European Parliament as isexplained in the respective position papers and statements.[Remark: The review of standards in the Drinking Water Directive willtake into account the best possible health advice and will thereforefollow changes in the WHO guidelines. The matter of artificialfluoridation of drinking water within the limits of the Directive is amatter of subsidiarity.}

o Copper: The parametric value for copper in the DWD should bereviewed and the conclusions of the third edition and any futureeditions of the WHO guidelines should be taken into account. In someareas (specific types of water) the effect of the aggressivity of coppercan be prevented through a restriction in the use of copper pipes orthrough water conditioning.

o Calcium (and magnesium): A minimum limit for calcium in water thathas been softened or desalinated is supported, but this could also bedone on a case by case basis according to some. Setting a minimumlevel for naturally soft waters is another matter. Health benefits needfurther attention as will be done by WHO.

o Pesticides: yes or no health-based standards in stead of the 0.1 µg/l(and issue of 0.5 µg/l for total pesticides). The opinions in Europe arevery mixed. Some were in favour of health-based standards such asthe WHO approach uses, while others strongly objected any changebecause the current values are a substitute for zero. Pesticides shouldnot be present in drinking water, but values are also necessary toprotect groundwater and surface water. There are already 4 pesticideswith a lower value in the DWD, because there toxicological limitvalues were below the general limit value. At the time the DWD wasprepared no limit values lower than the general value were known forother pesticides. This could be done again. The issue of pesticides willneed a lot of discussion in the future, as both points of view havestrong supporters.

o Disinfection-By-Products (DBP’s): There is general support to includechlorite but only in the case chlorine dioxide disinfection is used.Some also mentioned chlorate. However, in general there was littlesupport to include more DBP’s other than the feeling thatdevelopments within WHO with respect to these substances(cyanogen chloride and halo acetic acids) should be followed. Whensetting more and stricter standards for DBP’s, the effect on themicrobiological safety should be taken into consideration.

o Antimony: the proposed less strict value is supported.o Selenium, mercury and cyanide: there was much support for deletion

of these parameters (support was not unanimous), when necessarythey can be used in specific areas. However, there was a request not to


remove cyanide and mercury until the WFD has been implementedand the source water is no longer polluted.

o Sulphate and chloride: the deletion of these parameters was supportedby many but not by all. Objection was also made to the fact that nodistinction was made between both substances and the adverse healtheffect of sulphate was questioned.

o More information is needed on the various sum parameters such astotal pesticides, THM, PAH and tri- and tetra. Where possibleindividual parameters should de used.

o Analytical requirements: the DWD needs to be changed to include thedevelopments in this field such as definitions used (e.g. limit ofquantification and not detection), and performance characteristics(levels and definitions). A clear distinction should be made betweenLimit of detection and Limit of Quantification, as these two limits arenot the same.

o Water treatment chemicals are addressed in another section.


6.3.2 Microbiological parameters

6.3.2.1 General issues concerning drinking water legislationSome of the microbiological methods defined in the DWD are not alwayssuitable e.g. E.coli and coliforms, Clostridium perfringens and spores. Manyquestions did arise concerning the comparability of methods and subsequentequivalence testing. The position of EMAG on the issue of equivalencetesting, in accordance with ISO 17994 is endorsed. However, testing shouldnot be restricted to single Member States or geographical borders, but shouldtake more account of differences in drinking water quality. Member Statesshould be allowed to use results of research done elsewhere in Europe, butstill need to prove methods under the terms of the DWD. Pan-Europeancomparisons for equivalence testing are a most indicative way for bothestablished and new methods (as part of the validation) and will allow theidentification of performance of each method with regard to the differentwater types and water qualities.

The concept of risk management approach is an appropriate additional wayforward in the protection of human health. Care should be taken not tooverlook the risks asociated with newly emerging threats (how can the riskapproach prepare for the unexpected). Biofilms should be taken into accountas a source of microbiological contamination.Risk management approachshould not be viewed as a full replacement for compliance and checkmonitoring. A problem is to cover the whole water supply cycle from sourceto tap, especially with respect to the effect of internal plumbing installations.Sampling strategies must be adapted to the risk approach. In the currentDWD the sampling frequencies are too low to provide meaningful data toprotect public health or detect contamination on time. This could beaddressed by monitoring earlier in the production process. There is a need forverification to ascertain the process is under control. Proper information tothe consumer is needed.

The participants endorse the development of new and rapid methods as longas they are practical and affordable. Even though the methods and theinterpretation of the results may require a new regulatory approach thisshould not be an excuse to limit their development. Especiallystandardisation is a difficult issue as development of such methods is fasterthan the standardisation process. Molecular methods have a high potentialfor the detection of hygienically relevant micro-organisms which arecurrently not covered by routine measurements. Molecular methods shouldbe integrated into the equivalence testing system.

Sampling and monitoring should be carried out in accordance with ISO19458, once it is finalised and published. Sampling points should be selectedon the basis of risk assessment to provide information on the contaminationpotential of the water. The sampling method will depend on the objective ofthe sampling exercise (issue of flushing and disinfection).


If, in the future, EMAG is to play a significant role to the Commission and tobe involved in the setting up of the registrar of equivalence tests, itscomposition and functioning should be re-assessed.

6.3.2.2 Changes in the DWD 98/83/ECISO 19458 should be accepted when it is published. The sampling frequencyof microbiological analysis with respect to coliform bacteria and E.coli isconsidered rather low in the DWD. Consumers suffer from infections ratherthan intoxications and microbiological surveillance should be given moreattention.

On the parameter (Presumptive) Clostridium perfringens, different opinionsexist with respect of retaining it and the question of the method used. It isrecommended that the inclusion of Clostridium perfringens as a checkparameter in the DWD is reconsidered.The experts were not persuaded that there was potential for improved publichealth protection by the inclusion of any new parameters at present.However, this topic should be kept under regular view and the developmentof new and rapid analytical techniques may make more specific monitoringfeasible and better protect public health. No health benefit can be recognisedby the inclusion of specific pathogenic micro-organisms as parameters in theDWD. Risk assessment is a more valuable tool in this respect and helps toprevent parameter hysteria. Also Legionella should not be included eventhough monitoring at specific locations is important.

6.3.2.3 Need for guidance, research and working groupso There is a need for clear definitions and explanations of some wording

used in the DWD such as “equivalent”, “better”, and “as reliable as”.Does better mean more detected or more specific.

o More research is needed on viruses in drinking water, occurrence,significance, effect of disinfection and epidemiology.

o Pan-European research on comparisons of equivalence testing andidentification of performance of methods, including molecularmethods, Multi-Member State equivalence testing will help somesmaller Member States who find the requirements daunting.

o Uniform guidance for single labs on how validation can be done.o Advice (e.g. a manual) on how to carry out a microbiological risk

assessment approach.o Training needs for all stakeholders on risk management as the

regulator can no longer approach utilities in a uniform way but has todo this more individually.

o There is a need for more research on the correlation betweenmolecular methods with viability and virulence.

o A working group of specialists should be established to assess thepresent status and the development potential of various molecularbiological methods. Also a bridge is needed between research andregulation, interpretation problems.


6.3.2.4 Specific issueso Legionella should be controlled among others through the control of

plumbing installations. Critical end-points for Legionella need to beidentified in risk management approach.

o People involved in plumbing should be properly qualified to reducethe risk of Legionella, critical points like hospitals could be checkedfrequently to reduce the risks.

o Colony Counts should be included, a limit value should be consideredand also the frequency should be reviewed.

o The importance of parameters can vary between Member States suchas e.g. Cl. Perfringens and enteroviruses.

o There is insufficient evidence that bacteriophages are suitable asindicators for the presence of enteric viruses, but they can be useful asindicators of treatment performance for enteric viruses.

o In how far can a qualitative turbidity be used as an indicator forviruses. The idea does not find much support as there is no researchavailable on the correlation between the both. But turbidity can beused as a process parameter.

o Operators need to be included in further discussions about samplingat the consumers’ tap.

o Should Coliforms be kept in the DWD because they have noadditional value to E.coli. Or should they be deleted as parameter forthe distribution system but kept as a check on treatment efficiency.

o Cryptosporidium should also be mentioned in the position paper as apotential microbiological parameter.


6.3.3 Endocrine Disrupting Compunds (EDC’s)

6.3.3.1 General issues concerning drinking water legislationCurrently EDC’s are not regulated in the DWD. No limits for individualEDC’s or the total (potential) effects have been set in the DWD. For pesticides,polycyclic aromatic hydrocarbons (PAHs), chlorinated hydrocarbons andmetals, some of which are (suspected) EDC’s, limit values are part of theDWD. However, these limit values have not been set based on endocrineeffects.

Most European surface waters and some groundwaters are contaminatedwith trace amounts of EDC’s. However, surface waters that are subject todischarges of treated industrial or sewage effluent and with low dilution,levels of EDC’s may be higher. With the presently applied sewage treatmenttechnologies EDC’s cannot be fully avoided in surface waters receivingsewage treatment plant effluent. In this respect the Water FrameworkDirective will help as an upstream action to control EDC’s in drinking water.However, the effects of the WFD need to be monitored with respect to theconsequences for drinking water quality. As water resources are outside thescope of the DWD, the mis-use of chemicals that pollute water sources shouldbe prevented by other European legislation and a harmonisation betweenDWD, WFD and GWD is needed to achieve this.

The contribution of drinking water to the exposure of human beings is verylow, many EDC’s especially bio-accumulating compounds are taken up byfood in higher amounts on a daily basis.

6.3.3.2 Changes in the DWD 98/83/ECIt is not recommended to set limit values for individual EDC’s in the DWD atthis point, because several hundreds of substances have been identified aspotential EDC’s and prioritisation is still ongoing. Also the knowledge toevaluate the potential human health risk is insufficient, sufficient and reliabledata on the occurrence in both raw water and drinking water are not yetavailable and validated bioassays for screening the total potential effects ofEDC’s are lacking. More data are needed to evaluate the risk of humanconsumption of drinking water, especially at places with high risk factors forpollution.There is evidence of the release of EDC’s from materials to drinkingwater/bottled water. This needs to be addressed within the EAS.Risk assessment and risk management is probably the best tool for controllingEDC’s in drinking water.Future limit values for the total potential endocrine effects can be proposed,once validated analytical tools are available. Such a safe level for the sum ofEDC’s should be based on toxicological data and bioassays, value for totalEDC potency.


6.3.3.3 Need for guidance, research and working groupso More toxicological data and monitoring data are needed to evaluate

the risk for human health through drinking water.o More reliable data are required for a European wide assessment of the

occurrrence of EDC’s in drinking water.o It is recommended that European laboratories participate in the set-up

of a proficiency testing scheme for EDC’s.o There is an urgent need for the development of analytical tools and

validation of tools that assess total potential endocrine effects as wellas for various tests for different effects such as estrogenic, thyroideffects and androgenic effects.

6.3.3.4 Specific issues

o Advanced drinking water treatment is capable of removing EDC’s.However, it is not always clear what ‘advanced’ or ‘adequate’treatment means, as this can differ from country to country.

o Methods for monitoring individual EDC’s and total (potential) effectsare available on a limited scale. Implementation on a larger scale isneeded to carry out monitoring programmes. Implementation andvalidation of these methods in the Member States is probably aproblem.

o Bioassays only measure activity rather than effects, so the results fromsuch tests should not be used as a standard to judge drinking water.The assays should be used as a screening tools and not as a regulatorytool. It is a way to avoid endless lists of substances, but risk evaluationis quite a different story.

o There is no need for more toxicological data but a need for moreknowledge about the general exposure through drinking water andinformation on multiple exposure and the associated risk.

o Routine measurement of all individual EDC’s is impossible.


6.3.4 Risk management approach in drinking water

6.3.4.1 General issues concerning drinking water legislationA risk assessment and management approach is consistent with EU policiesand developments in other areas of consumers health (such as food safety)and environmental health. The use of the Precautionary Principle for drinkingwater can be based on a quantitative assessment of risks of pathogens andtoxic chemicals occurring in drinking water. Risk assessment and riskmanagement approach is an instrument to quantify the level of safety andhence to demonstrate that the Standstill Principle is met in future, using thecurrent protection level as a benchmark. It is of vital importance that theapproach encompasses the whole water supply process from raw watersource to tap, where all responsibilities are properly defined. It is thereforenecessary that harmonisation with other EU legislation (especially WFD andGWD) occurs and that the outcome of this seminar is brought to the attentionof the various groups working on these Directives. The involvement of allstakeholders in the supply chain from catchment to consumer will strengthenthe reliability of the system and will lead towards possible new solutions andtransparency of responsibilities.

Incorporation of risk assessment and management strategies are of largeadded value for the DWD and for safeguarding the supply of safe drinkingwater that keeps the trust of consumers in the European Member States. Thecore principles given by WHO’s Framework of Safe Drinking Water are ingood agreement with the principles used by the EU in other areas ofconsumer safety, and are thus seen as a sound basis to be included in therevision of the DWD. For many water suppliers RA/RM is already commonpractice, but a more consistent approach formalises existing practices andmakes them more rigorous and transparent.

6.3.4.2 Changes in the DWD 98/83/ECThere is broad support for the overall concept and the core principles of theRA/RM approach to be included in the revision of the DWD. Prioritisation ofsuch an approach would be very helpful for accession countries. In a revisionof the DWD the Water Safety Plans could be accommodated, where the EUshould provide an overall framework of core principles and a knowledgebase of health-based targets and Member States (in line with the SubsidiarityPrinciple) should implement programmes and plans that are consistent withthe overall framework. Inclusion of RA/RM strategies will change the waythe DWD is structured, where the new approach should not only reflect in theAnnexes but also in the Articles of the DWD.

There is a need for careful and adequate sector preparation and anappropriate transition period where the old and the new approach areallowed to function in parallel. It is recommended that the RA/RM approachas described in the third edition of the WHO Guidelines for Drinking-waterQuality is gradually introduced in the current European drinking water


policy and legislation. A flexible and step-wise approach helps to adequatelyfill research needs and knowledge gaps, and to motivate the majority ofstakeholders to make this concept a living tool.

Responsibilities have to be identified and defined at various levels such asEU, national and local level. In addition there is a need for re-design ofsurveillance programmes and a major change in surveillance techniques. Ingeneral there will be a shift from compliance checking (often by HealthAuthorities or other bodies responsible for independent surveillance) towardsinspection and audit based approaches.

The introduction of a RA/RM approach to manage drinking water safetyfrom source to tap requires risk- or health-based water quality targets In thecurrent DWD not all standard values are health-based. The setting of thetargets for chemicals and micro-organisms will provide clarity regarding thebasis of standard values.

RA/RM approach is not only useful for regulated parameters but also fornew parameters. As is the case with the current approach accidents andproblems will happen, but RA/RM is a more appropriate approach toaddress them.

The experts support the continued use of end product testing as being part ofthe whole risk approach. It should be used as verification that process controlis effective and hence more targeted to the specific critical water qualityissues in a particular water supply. Compliance with water quality standardvalues according to the recommendation of the other preparatory groupsshould remain as a major component within the DWD as a means of verifyinghat safety has been achieved.

Many questions are raised about the cost involved. Even though costreduction is not the driving force, optimisation of effectiveness is the ultimategoal and the approach should assist in cost-effective solutions. There is aworry however, that costs might increase substantially in particular for smallsupplies. Costs benefit analyses need to be carried out.

The present DWD offers some flexibility to include (some of) the principles ofRA/RM, existing policy instruments can be used but ultimately the DWD willneed to reflect the shift to a risk based approach for consistency.

6.3.4.3 Need for guidance, research and working groupso Implementation should be preceded and supported by several pilot

projects to translate en evaluate the core principles of this approach inthe European context. Support for pilot projects is necessary and theCommission should facilitate exchange of knowledge and experiencein this respect.


o Evaluation of existing experiences inside and outside Europe isnecessary.

o There is a need for the development of guidance and trainingmaterials.

o Special attention is needed for small and medium-sized watersupplies, e.g. a more generic guidance document. Small supplies needa different approach from large utility supplies.

o Co-ordination of capacity building and training programmes.o Reference level of risk (or safety) is needed and should be decided

upon at EU or national level (DALY approach).

6.3.4.4 Specific issueso Communication to all stakeholders and especially the consumers is

needed. Communication should not focus on limitations in endproduct testing as this will result in a credibility gap with theconsumers. Communication should emphasise the fact that theapproach takes safety further forward rather than condemning the lastcentury of success.

o Clear definitions are needed about RA/RM approach.o In the RA/RM approach only health-based parameters should be

taken into account.o Quantitative RA for micro-organisms and chemicals as a means of

setting health targets can give a misleading impression that there willbe real reductions in adverse health effects. Quantitativemicrobiological RA is not appropriate for setting regulatory standards;it is an information tool to be used with care.

o Avoid using HACCP as this can be too narrow and fails to considerthe multiple barrier principle. Also distribution systems, whilstdifficult and a challenge to identify CCP, can also be brought into theRA/RM.

o Quality of materials and substances in contact with drinking watershould become integral part of the Water safety Plans.

o RA/RM will not solve old problems as e.g. lead.


6.3.5 Substances and materials used in the preparation or distribution of waterintended for human consumption

6.3.5.1 General issues concerning drinking water legislationRegulated products and materials used for storage and distribution of waterintended for human consumption that are permanently incorporated inconstruction works fall within the scope of the Construction ProductsDirective (CPD 89/106/EEC) and are referred to in the DWD (Article10). In1999 the RG-CPDW was established with the approval of the StandingCommittee on Construction (SCC) and the Standing Committee on DrinkingWater (SCDW). The RG-CPDW is responsible for making proposals for aEuropean Acceptance Scheme (EAS). The EAS is intended to be a singlesystem of regulatory framework and supporting test methods covering theacceptance of construction products used in contact with water intended forhuman consumption. The EAS is still under development, the latest versionof the scheme the so-called EAS on paper has been out for public consultationand is currently being updated. Not all elements of the EAS are finalised.

Water treatment chemicals are not covered by the CPD, but a Europeanscheme for acceptance of drinking water treatment chemicals already exists inthe form of the European standards mandated by the European Commissionunder the Public Procurement Directive (89/66/EC), the ENs for watertreatment chemicals.

Water treatment disinfectants that are placed on the market (such as chlorine)are covered also by the Biocidal Products Directive (98/8/EC).The relevance of the Biocidal Products Directive (98/8/EC) in relation to theENs and the DWD should be checked in more detail. By analogy with theCPD, a provision in the DWD may be considered stating that ‘technicalspecifications (or uniform principles) of the BPD shall respect therequirements if this Directive’.

An EC owner of Positive Lists and Composition Lists has to be identified andinstalled. The toxicology of new complex treatment chemicals and theirimpurities and how to handle them should be decided at the European level.It should also be made clear how CEN is to obtain the toxicological input forthe revision of existing ENs in the light of new data.

6.3.5.2 Changes in the DWD 98/83/EC

There was general support for the recommendations in the position paper,including support for the EAS and supporting CEN standards as well as forthe system for water treatment chemicals. There should be a reference tothese in the DWD at the appropriate time.No specific changes to the DWD are proposed. The EAS in principle intendsto be a system to contribute to the compliance of drinking water with the


specific and general requirements of the DWD, and as such, it would beappropriate to make specific references to the EAS in the DWD. However,while the EAS is intended to be legally binding on Member States at present itis not so. Thus reference to the EAS in the DWD cannot be undertaken. Atsome point in the future reference to the EAS would be appropriate (between2004 and 2006, according to some too optimistic). However, before that adetailed examination of the scope of the EAS, compared to the DWD needs tobe undertaken. Also it has to be reconsidered whether, e.g. EAS principles(related to the protection level) and mechanisms to update Positive Lists andComposition Lists should be laid down in the DWD.

It would not be appropriate in future to include so-called Positive Lists forproducts made out of organic material and Composition Lists for productsmade from metallic materials as Annexes to the DWD, as these would befrequently updated. However, the DWD could specifically refer to them ifnecessary.

With respect to treatment chemicals a possible reference could be made to‘use of relevant EN’, but not all experts were convinced of the legally bindingnature of the PPD with respect to health related issues of water treatmentchemicals. If a reference to the EN is possible then the current DWDparameters dealing with impurities in treatment chemicals, epichlorohydrinand acryl amide (and vinyl chloride) could be considered for deletion sincethey would be covered in the corresponding EN.

There should be consideration of adding to the DWD, provisions dealing withthe toxicological evaluation of new water treatment chemicals and theirimpurities as well as attestation of conformity measures and the revision ofexisting ENs in the light of new toxicological data.

6.3.5.3 Need for guidance, research and working groupso Study the relationship between the DWD, Public Procurement

Directive the ENs for water treatment chemicals and the BiocidalProducts Directive.

o Clarification of terms (especially in Articles 2 and 10) such as ‘allmeasures necessary’, and ‘other domestic uses’, to better define theintention and scope of the DWD.

o Clarification of the scope of the EAS in order to identify anyrequirements of the DWD that are not covered (such as materials andsubstances used in the water works and distribution to the consumerstap, but also hot water applications).

o Specify protection level in the DWD with respect to the adoption andoperation of the EAS.

o How should the toxicology of new complex treatment chemicals andtheir impurities be handled at the European level.


6.3.5.4 Specific issueso A statement was made on fluoridation of drinking water as according

to the speaker these chemicals are toxic and pose a hazard to humanhealth.

o Some Member States assume that the DWD covers hot waterapplications and therefore hot water is included in the EAS. There aremany questions on whether products are covered by the EAS (such asmixer taps). Not all experts agree that hot water needs to meet therequirements of the DWD. It is also stated that hot water should betreated as heated water prepared from drinking water.

o Protection of the European market is needed against unsatisfactoryproducts coming from outside the EU. The answer is that all productshave to fulfil the essential requirements of the CPD (productstandards), and it is the responsibility of Member States to check this.The essential requirements apply to construction works and not toproduce.

o Metals in contact with drinking water are a difficult area, with littlecontrol over domestic networks. Not only is the selection of materialsimportant but also the design and maintenance of the plumbinginstallation. Maintenance and usage of installations might need EUstandardisation. The suggestion is also made to include in the DWDan obligation to regulate backflow from domestic installations to thepublic network. This has been regulated in EN 806 and EN 1717 on thedesign of plumbing installations. There should be associated technicalguidelines in the DWD.

o Materials used in the transport of raw water for the production ofdrinking water and whether they are covered by the EAS is left toMember States to decide.

o Attestation of conformity of treatment chemicals is important. Thenew proposal of EC DG Environment, REACH (Registration,Evaluation, and Authorisation of Chemicals) might assist in the issueof who will provide the assessment of toxicity data for impurities (atthe European level) for any new complex water treatment chemicals.

o Notified bodies: there are questions about the notified bodies as towho will monitor them. There should also be strict consistency ofnotified bodies and both equal and good practice of such bodies.Answer: the EAS will describe transparent harmonised operatingprinciples for the notified bodies. The group of NBs has recentlystarted discussing its operational procedures. There is some concernover the capacity of notified bodies to carry out the requirements ofthe EAS. At the moment there is an exchange of knowledge betweenregulators and notified bodies.

o There is concern about the temperature setting fixed for thedevelopment of some tests included in the EAS. Higher safety marginsare requested considering the climatic conditions existing in someSouthern countries, where temperatures in domestic systems arethought to be higher than 32 degrees and the effects of materials (alsobecause of their smaller pipe diameter) are expected to be moreintense.


6.4 Concluding remarks Pierre Hecq Desk Officer DGENV B1 DrinkingWater Directive

The European Commission is very pleased with the success of the drinkingwater seminar, because of the number of countries, experts and stakeholdersthat attended the seminar. The seminar was well organised and the outcomeof the various group meetings and the discussions is promising. As majorconclusions the following needs have been identified:

• no need the revise the Drinking Water Directive at this stage.• learn from experience with the current DWD in the existing Member

States and in the enlargement countries.• guidelines and guidance documents on various subjects.• workshops on specific issues.• clarity and transparency.• more harmonisation between EU Directives e.g. WFD, GWD and

DWD.• more involvement of the consumers as they are more or less absent

from the seminar.

The Commission invites all stakeholders to send written comments to theCommission before the 7th of November. The web-site was closed after thesecond day of the DWS.

The synthesis report on the DWS will be ready at the end of 2003 and will besend to all participants of the seminar and to all that participated through theweb-site.

The Commission will use the outcome of the DWS to set priorities and todefine the work programme. Working groups will be established andworkshops on identified issues will be organised. The Commission will usethe expertise available within the various groups of stakeholders to take theconclusions of the seminar forward. However, this will not be done before theend of next year (2004). There will be a liaison with groups working under theWFD.