Guidelines for Drinking Water Quality - Training Pack (WHO)

Preface

Between 1993 and 1997, the World Health Organization (WHO) published the second edition of Guidelines for Drinking-Water Quality in three volumes: Volume 1: Recommendations (1993); Volume 2: Health Criteria and Other Supporting Information (1996); and Volume 3: Surveillance and Control of Community Supplies (1997). As with the first edition of the Guidelines, their development was organised and carried out jointly by WHO headquarters and the WHO Regional Office for Europe.

This pack is intended to provide information for use in the planning and delivery of seminars, workshops and training courses in water quality surveillance, control and improvement, especially where these concern the WHO Guidelines for Drinking-Water Quality. The pack contains 23 different sessions, including both presentation and practical exercises.

It is hoped that the availability of this pack will encourage and assist local, national and regional authorities to implement events of this type with or without the assistance of expert institutions or individuals. It is hoped that the pack will facilitate the development of expertise and thereby promote the organisation of further events.

The pack is designed to cover a broad range of water-related topics in order that appropriate elements can be selected in response to local circumstances and priorities. The pack includes sessions addressing the scientific basis of the Guidelines; the establishment of national standards; the ways in which water supplies may be improved; and some broader issues such as human resource development.

Each section of the pack addresses a single session and includes the objectives, a session plan, a background paper and overhead transparencies. The materials are intended to provide a resource person with information to assist in the review of what they might reasonably expect to achieve in a session, and to plan the structure and layout of the session. The background papers can, where appropriate, be provided to participants. The pack also addresses practical sessions. The materials provided to support these give guidance as to how such sessions could be delivered and the materials required to implement them.

It is unlikely that all sessions would be necessary for a single seminar, workshop or training course. It is therefore important that the overall objectives of an event are defined, taking local priorities into account when selecting which sessions will be of most practical use.

The sessions in the pack can be divided into a number of groups which could be of value when planning its use, defining the target audience and selecting sessions. The groups are described in the table over leaf.

In order to develop a well-balanced seminar, workshop or training course, at least one session would normally be required from each group. Discussion of drinking-water quality and the use of the Guidelines for Drinking-Water Quality should, for example, generally be accompanied by sessions considering monitoring and assessment and the means to secure improvements; and, possibly, a practical session illustrating issues of particular local relevance.

It is hoped that this pack will be a useful addition for those implementing training courses in water quality. The pack should assist in building capacities to provide and facilitate initial, further and ongoing training for staff of diverse disciplines. Any comments that users of the pack may have on experience with its use and which might assist in its further development would be gratefully received and should be addressed to:

Jamie Bartram
WHO
Water and Sanitation for Health Unit
1211 Geneva 27
Switzerland

Water and Public Health

Session Objectives

· To demonstrate the link between water and health and show the profound influence of water supply and quality on public health.

· To describe the basic classification of water-related disease.

· To describe the concept of the faecal-oral route of disease transmission and the classic waterborne disease cycle.

· To describe how improvements in water supplies will lead to improvements in health and a reduction in morbidity and mortality rates.

Introduction

Water has a profound influence on human health. At a very basic level, a minimum amount of water is required for consumption on a daily basis for survival and therefore access to some form of water is essential for life. However, water has much broader influences on health and wellbeing and issues such as the quantity and quality of the water supplied are important in determining the health of individuals and whole communities.

The first priority must be to provide access for the whole population to some form of improved water supply. However, access may be restricted by low coverage, poor continuity, insufficient quantity, poor quality and excessive cost relative to the ability and willingness to pay. Thus, in terms of drinking-water, all these issues must be addressed if public health is to improve. Water quality aspects, whilst important, are not the sole determinant of health impacts.

The quality of water does, however, have a great influence on public health; in particular the microbiological quality of water is important in preventing ill-health. Poor microbiological quality is likely to lead to outbreaks of infectious water-related diseases and may causes serious epidemics to occur.

Chemical water quality is generally of lower importance as the impact on health tend to be chronic long-term effects and time is available to take remedial action. Acute effects may be encountered where major pollution event has occurred or where levels of certain chemicals are high from natural sources, such as fluoride, or anthropogenic sources, such as nitrate.

Microbiological drinking-water quality and human health

The microbiological quality of drinking-water has been implicated in the spread of important infectious and parasitic diseases such as cholera, typhoid, dysentery, hepatitis, giardiasis, guinea worm and schistosomiasis.

Many other diseases are associated with water in other ways. Water may act positively in the control of some through its use in hygiene, and may act as a source or vector for others where contact with water is required for disease transmission or where agents of disease or insect vectors require water in which to complete their life cycle. The various relationships between water and disease are summarized in Table 1.

Water-related disease incidence worldwide

Water-related disease places an excessive burden on the population and health services of many countries worldwide and in particular those in developing countries. Table 2 shows estimates of the morbidity and mortality rates of some major water-related diseases worldwide, figures which are likely to be conservative estimates.

Table 1: Diseases related to water and sanitation

adapted from Bradley, D J, London School of Hygiene and Tropical Medicine, various

Forty per cent of mortality in children under five years of age is related to diarrhoeal disease and it has been estimated that in 1995 more than 1,500,000,000 episodes of diarrhoea occurred in children under five years of age in the developing world (excluding China) and that some 4,000,000 of these resulted in death.

Table 2: Morbidity and mortality rates of some important water-related diseases (after WHO, 1995)

These diseases are caused by the ingestion of contaminated faecal material transmitted by the transmitted by the faecal - oral route. Infectious agents of all types may be transmitted by the faecal - oral route via water, including viruses (such as infectious hepatitis, rotavirus and Norwalk agent); bacteria (such as cholera, typhoid and dysentery); and parasites (such as Giardia, Cryptosporidium and Entamoeba).

Faecal pollution of drinking-water may be sporadic and the degree of faecal contamination may be low or fluctuate widely. In communities where contamination levels are low, supplies may not carry life-threatening risks and the population may have used the same source for generations. However, where contamination levels are high, consumers (and especially the visitors, the very young, the old and those suffering from immuno deficiency-related disease, for instance through malnutrition or AIDS) may be at a significant risk of infection.

Improving water and sanitation and improvements in health

Results of epidemiological studies into the relationship between the quality of water supply and sanitation versus human health vary widely and there are severe methodological difficulties involved in undertaking such studies. Nevertheless there is sufficient evidence to support the conclusion that improving water supply and sanitation can have a significant impact on human health. Table 3 summarizes the findings of an extensive review of studies of this type.

Table 3: Percentage reduction in the diarrhoea morbidity rate attributed to improvements in water supply or excreta disposal

Source: after Esrey, Feachem and Hughes, 1985

One of the reasons for the difficulty in undertaking studies on the health impact of improvements in water supply quality is that the faecal - oral route includes several and multiple routes to infection as summarized in Figure 1 below.

Figure 1: Principal elements of faecal - oral disease transmission

This complexity of routes also demonstrates the importance of various aspects of hygiene as complementary actions to water quality improvements.

Clearly, the likelihood of acquiring a waterborne infection increases with the level of contamination by pathogenic (disease-causing) microorganisms. However, the relationship is not necessarily a simple one and depends very much on factors such as infectious dose and host susceptibility.

Moreover there remains some doubt as to the relative importance of drinking-water quality and other aspects of water supply on the prevalence of infections with a faecal-oral route of transmission. For example, some agents with a low infectious dose may be transmitted primarily from person to person and thus improving the quality of drinking-water may not make a dramatic impact on their prevalence in the community. Human rotavirus and some species of Shigella fall into this category. Bacteria which are capable of multiplication in food may follow a food-borne transmission route more readily than waterborne.

Conversely there are other agents for example Salmonella typhi, Vibrio cholerae, Giardia lamblia and hepatitis A virus which are frequently transmitted via contaminated drinking-water. Where this is the case, improvements in water quality may result in substantial reductions in prevalence.

In those cases where transmission is not primarily water borne, improvements in water availability and personal hygiene may be much more important in reducing morbidity from diarrhoea and other water-borne infections.

The relative importance of drinking-water quality to the maintenance of public health may vary with respect to a number of geographical, social, seasonal and microbiological factors. It is not possible to state with any confidence which aspect of water supply is the most important at any one time or in any one location. What is becoming increasingly clear however is that all factors relating to the quality and availability of drinking-water are potentially important and must be taken into consideration. In this context it is worth emphasizing that one of the few general conclusions that may be drawn about drinking-water quality is that if faecally-derived pathogens are not present, then endemic or epidemic waterborne disease will not occur.

Other aspects of microbiological quality

As noted above, water borne disease is not exclusively transmitted by the faecal-oral route, although this route of disease transmission is of overwhelming importance globally. Some other microbiological aspects of importance are as follows:

Opportunistic and other water-associated pathogens

Opportunistic pathogens are naturally present in the environment and normally present no risk to human health. They are able to cause disease in people with impaired local or general immune defences. These people include the elderly and the very young; persons with extensive burns; persons undergoing immuno-suppressive therapy (such as following transplant surgery) and those with immuno deficiency-related diseases (such as AIDS). Examples of opportunistic pathogens of this type include Pseudomonas aeruginosa, certain species of Flavobacterium, Acinetobacter, Klebsiella, Serratia, Aeromonas and some 'slow growing' mycobacteria.

Inhalation of water containing certain infectious agents may also cause disease. This is the case with, for example, Legionella spp (Legionnaire's disease) and Naeglaria fowleri (an occasional cause of primary amoebic meningoencephalitis).

Cyanobacterial Toxins

Some cyanobacteria ('blue-green algae') are capable of producing toxins, including hepatotoxins, neurotoxins and lipopolysaccharides. Few epidemiological studies have been undertaken and little information is available regarding the true importance of this problem. Where blooms of cyanobacteria occur in lakes and reservoirs used for drinking-water supply a potential risk to health exists and therefore impounded surface waters used for drinking-water supply should be protected from contamination with nutrients.

Nuisance organisms

A number of organisms of no public health significance are undesirable because they produce turbidity, taste or odour or because they are visible to consumers of drinking-water. Their presence indicates that water treatment and supply system maintenance may be defective. These include: tastes and odours from Actinomyces and Cyanobacteria; and infestation of water mains by animal life feeding upon microbial films, such as the crustacean Gammarus pulex, Nais worms and the larvae of chironomids.

Chemical contamination and health

Chemical contamination of drinking-water may also have effects on health, although in general these tend to be chronic rather than acute, unless a specific pollution event has occurred and are therefore generally considered of lower priority than microbiological contamination.

Chemical pollutants which affect health include nitrate, arsenic, mercury and fluoride. In addition, there are an ever-increasing number of synthetic organic compounds released into the environment whose effect on human health is poorly understood, but which it appears may be carcinogenic.

Some details are given below on the four substances noted above, however, it must be recognized that raised concentrations of any chemical known to have an impact on human health may lead to long-term problems. In general, water sources used for drinking-water supply should be protected from chemical contamination through land-use control, definition of protection zones and application of adequate wastewater treatment.

Nitrate

Excess nitrate in drinking-water has been linked to methaemaglobinamenia in infants, the so-called 'blue-baby' syndrome. Nitrate leads to the oxidation of normal haemoglobin to methaemoglobin which is unable to transport oxygen to the tissues. This may result in cyanosis (a dark blue coloration) and in some cases, asphyxiation and death.

The Guideline Value (GV) for nitrate of 50 mg/l has been set on the basis of the acute health risk to infants and is unusual for this reason as most GVs are set for long-term risks. Many countries are now experiencing problems with elevated nitrate, particularly in groundwaters caused through poor treatment and disposal of excreta, intensification of animal husbandry and large-scale applications of inorganic and organic fertilizers.

In some countries, notably in the Countries of Central and Eastern Europe (CCEE) such as Moldova and Romania, levels have been recorded in shallow groundwater at up to 1000 mg/l, whilst in India anecdotal evidence suggest levels of up to 1500 mg/l. At these levels, more widespread chronic effects are likely to be noted including a possible greater likelihood of gastric cancer.

Nitrate is a conservative element in natural groundwaters and therefore once large-scale nitrate contamination has occurred, it will take a considerable period of time before it is naturally attenuated through de-nitrification or diluted. In these circumstances, short term measures will include identifying alternative sources of water, for instance deeper boreholes, or through blending with low-nitrate waters. Removal of nitrate by ion exchange in treatment plants is expensive as most anion exchangers are non-selective for nitrate and therefore nitrate specific resins must be used.

Long-term solutions must involve the reduction in the release of nitrate into the environment through, for example, control of fertilizer application and improvements in human and animal excreta treatment and disposal.

Arsenic

A provisional GV of 0.1 mg/l has been set for arsenic on the basis of an excess cancer risk of 6 × 10^-4. In some parts of the world, natural sources of arsenic may contaminate water supplies and lead to poisoning of the users. The most well-documented cases of arsenic poisoning from drinking-water have come from India, where there is arsenic contamination of large numbers of rural water supplies. Common symptoms include inflamed eyes and skin lesions. Arsenic contamination has also been noted in southern Thailand and the CCEE.

Most natural arsenic comes from the reduction of arsenic complexes caused through changing redox and pH conditions and from the oxidation of arsenic containing minerals exposed by falling groundwater tables induced through over abstraction or reduced recharge.

There is also increasing evidence that there is a tendency for arsenic levels to increase in shallow groundwaters under urban areas. This has been particularly noted where conditions become anoxic, organic rich sediments are present and arsenate compounds associated with iron are common. This has significant implications for water supply in these areas, particularly in low-income areas where community-based water projects may involve the sinking of dug and wells and shallow tube wells. Arsenic may also be discharged in effluent from a variety of industrial processes.

Control options for arsenic contamination will vary according to the source. Arsenic derived from industrial effluents should be controlled through proper treatment of wastes and monitored by the pollution control agency. The control of arsenic from natural sources must include sustainable groundwater resource management. Many of the problems noted in India result from over-abstraction of groundwater, primarily by the agricultural sector. Arsenic problems noted under urban areas may be more difficult to control given the range of factors which influence whether arsenic is released.

In all cases, short-term options will include treatment of water in home using, use of alternative sources or a switch to an alternative source, such as deep groundwater unaffected by arsenic contamination. Arsenic may be removed at treatment plants through a variety of processes, although like most treatment aimed at chemical removal, increase the costs of producing drinking-water.

Fluoride

Fluoride in drinking-water can have toxic effects in both excess and deficiency, although WHO only set a GV of 1.5 mg/l for excess fluoride as susceptibility in deficiency is highly dependent on nutritional status.

Excess fluoride may lead to dental or skeletal fluorosis, the latter being a crippling disease which affects a number of areas including the Rift valley of East Africa and parts of India, Mexico and the former Soviet Union. However, a lack of fluoride may cause dental caries, a weakening of the teeth, thus in some circumstances fluoride may be added to the drinking-water supply.

The acceptable concentration of fluoride in water is in part related to climate, as in warmer climates the quantities of water consumed are higher thus leading to a greater risk of fluoride related problems as overall intake increases. Susceptibility of individuals to fluorosis may also be determined by renal impairment.

Control options for fluoride contamination of water include blending of fluoride-rich waters with waters of low fluoride content, selection of low-fluoride sources and removal of fluoride by treatment at public water supply or household level. Fluoride can be successfully removed by precipitation by use of coagulants (commonly an alum-lime mix), adsorption on activated carbon substrates, osmosis or ion exchange. Fluoride removal is often more effective at a water supply level and the Nalgonda technique, developed in India, has been proven as a low-cost techniques which can operate on a variety of water supply options ranging from piped water supplies to handpump units.

References

Anon. World Health Organization Guidelines for Drinking-water Quality, Volume I, 2nd Edition, WHO, Geneva, 1993

Anon., World Health Organization Guidelines for Drinking-water Quality, Volume II, 2nd Edition, WHO, Geneva, 1996 Anon. Report of a National Water Quality Seminar, Romania, WHO-EURO, Rome, 1995

Esrey, S.A., Feacham, R.G. & Hughes, J.M., Interventions for the control of diarrhoeal diseases among young children: improving water supplies and excreta disposal facilities, Bulletin of the World Health Organization, vol. 63, No. 4, pp. 757-772, 1985.

Hofkes, E.H. (Ed), Small Community Water Supplies, John Wiley & Sons. Chicester. 1986.

Presentation Plan

Water Quality

"All people, whatever their stage of development and social and economic condition, have the right to have access to drinking water in quantities and of a quality equal to their basic needs."

(UN Conference at Mar del Plata, 1977)

Global Morbidity and Mortality Rates

World Health Report, 1995

Potential Reductions in Morbidity for Different Diseases as a Result of Improvements in Water Supply and Sanitation

Infant Mortality versus Access to Safe Water

Source: Regli et. al 1993

Water and Sanitation-related Diseases

The Faecal-Oral Route of Disease Transmission

The Classical Waterborne Infection Cycle

Source: Tebbutt, T.H.Y., 1992

The WHO Guidelines for Drinking-Water Quality

Session Objectives

· To introduce the latest edition of the Guidelines; identifying all three volumes and the information contained within each.

· To emphasise the basic concept and the advisory nature of the Guidelines and to describe the difference between scientific risk assessment and risk management.

· To provide an outline of the consultation process that resulted in the revised 2nd edition of the Guidelines.

· To discuss the reasoning behind the prioritisation of microbiological quality of drinking water in the Guidelines.

· To provide a basic overview of the criteria used in the selection of contaminant substances that are contained within the Guidelines.

· To explain the nature of Guideline Values, highlighting substances and parameters to which they apply.

· To explain the process of the rolling revision of the Guidelines.

Introduction

An established goal of WHO and its Member States is that:

all people, whatever their stage of development and their social and economic conditions have the right to have access to an adequate supply of safe drinking-water.

In this context, 'safe' refers to a water supply which is of a quality which does not represent a significant health risk, is of sufficient quantity to meet all domestic needs, is available continuously, is available to all the population and is affordable. These conditions can be summarised as five key words: quality; quantity; continuity; coverage; and, cost.

The importance of these key words cannot be over-emphasized since the impact of contaminated drinking-water on health has been well documented and range from massive outbreaks of infectious and parasitic diseases to subtle chronic toxicological effects. It is vital that all these key issues are addressed, if clear policies and programmes on water supply and quality are to be established and maintained.

To assist governments in dealing with these and related issues regarding water quality, WHO has over the years, been involved in the review and evaluation of information on health aspects of drinking-water supply and quality and in issuing guidance material on the subject.

The first WHO publication dealing specifically with drinking-water quality was published in 1958 as International Standards for Drinking-Water. It was subsequently revised in 1963 and in 1971 under the same title. Because of the ever-continuing research on water quality, the 1971 standards were again reviewed, and in 1984 the WHO Guidelines for Drinking-Water Quality were published.

The philosophy and content of these Guidelines constituted a significant departure from the old International Standards as they were designed as advisory in nature based solely on the impacts on human health of the various substances and organisms considered. Standards have, by their nature, to take other considerations into account such as social, economic, environmental, political and financial considerations and have to balance a number of criteria.

In 1989, work was started on a second edition of the Guidelines. These new Guidelines which were published in 1993-97 rely to a great extent on the pioneering concepts of the 1984 Guidelines.

The purpose of this paper is to briefly describe the second edition of the Guidelines, the revision process and the scope and new concepts incorporated into the Guidelines for the 1990s.

Presentation

The Guidelines have been published in three volumes:

Volume 1 - Recommendations describes the criteria used in selecting the various microbiological, chemical and radiological contaminants considered, the approaches used to derive the guideline values, and brief information supporting the values recommended, or explaining why no health-based guideline value was recommended.

Volume 2 - Health Criteria and Other Supporting Information is essentially an environmental health criteria document covering the contaminants that were examined with a view to recommending guideline values. Volume 2 elaborates greatly on the health risk assessment of microbial and chemical contaminants presented in Volume 1 and should be considered as a vital companion document.

Volume 3 - Surveillance and Control of Community Supplies deals specifically with small communities, predominantly those in rural areas of developing countries.

Preparation

At the time the Guidelines for Drinking-Water Quality were published in 1984, it was recognized that as new information on the potential health risks of contaminants in drinking-water became available, the basis of the recommended guideline values would need to be reviewed and revised. New or changed guideline values would therefore have to be recommended.

In 1988, the decision was made within WHO to initiate the revision of the Guidelines. As with the 1984 Guidelines, responsibility for carrying out this revision was shared between WHO's Headquarters and the Regional Office for Europe (EURO). Within Headquarters, both the Urban Environmental Health Unit (UEH) and the International Programme on Chemical Safety (IPCS) were involved; IPCS providing a major input to the health risk assessment of chemicals in drinking-water.

From the onset, it was agreed that the general philosophy of the 1984 Guidelines remained sound and valid and should therefore not be changed.

A series of planning and co-ordination meetings took place to establish the scientific approach and mechanism for the preparation of evaluation documents, substance by substance, for the revision of the Guidelines. This was followed by a series of Review Group Meetings dealing with specific subject areas. A total of 19 meetings were held involving the participation of numerous institutions, and over 200 experts from some 40 different countries.

The preparation of the Guidelines required intensive human and financial resources. The Guidelines could not have been developed without the scientific and/or financial support of the following organisations and countries: DANIDA, NORAD, SIDA, ODA (UK), Belgium, Canada, Denmark, Finland, France, Germany, Italy, Japan, Netherlands, Norway, Poland, Sweden, United Kingdom, and the United States of America.

Microbial contaminants and some 128 chemicals were selected for evaluation. For each selected chemical, a lead country prepared a draft evaluation document examining its occurrence in drinking-water, exposure from food and air, effects on laboratory animals and humans. Based on the evaluation of available data, a guideline value was also proposed. The outline of such an evaluation document is given as Annex 1. These evaluations constitute Volume 2 of the Guidelines.

The draft evaluation document was then circulated for review by the Co-ordinator to the "support countries" and selected experts. The Co-ordinator worked with the lead countries to incorporate the comments received and prepared overviews of scientific issues to be resolved. This documentation was then submitted for evaluation to a Review Group meeting which took a decision as to the health risk assessment and recommended a guideline value. The role of the seven Co-ordinators was crucial in the revision process.

During the preparation of draft evaluation documents and at the Review Group meetings, careful consideration was always given to previous risk assessments carried out by the WHO/ILO/UNEP International Programme on Chemical Safety in its Environmental Health Criteria Monographs, by the International Agency for Research on Cancer, the Joint FAO/WHO Meeting on Pesticide Residues and the Joint FAO/WHO Expert Committee on Food Additives which also evaluates contaminants such as lead and cadmium in addition to food additives.

Basic Concept

As reflected in the title, the Guidelines are of an advisory nature and are intended to be used by national or regional authorities as a basis for the development of drinking-water standards and regulations appropriate to their own socio-economic and exposure situation. The Guidelines clearly recognize the desirability of adopting a risk-benefit approach (qualitative or quantitative) to national standards and regulations. The establishment of drinking-water quality standards by individual governments must follow a very careful process in which the health risk is considered alongside other factors, such as technical and economic feasibility. Standards achieve nothing unless they can be implemented and enforced. When establishing national standards, consideration must be given to the practical measures that will need to be taken with respect to finding new sources of water supply, instituting certain types of treatment, and providing for adequate surveillance and enforcement.

Priorities

Since water is essential to life, the first priority is that it must be made available to consumers even if the quality is not entirely satisfactory.

As with the 1984 Guidelines, the new 1993 Guidelines place the greatest emphasis on the microbiological quality of drinking-water. The microbial contamination of drinking-water has been implicated, directly or indirectly, in the spread of major infectious and parasitic diseases such as cholera, typhoid, dysentery, hepatitis, giardiasis and guinea-worm infection. In 1992, the United Nations Conference on Environment and Development (UNCED) estimated that '..80 per cent of all diseases, and over one-third of deaths in developing countries are water-associated, and on average as much as one-tenth of each person's productive time is sacrificed to water-related diseases' (Agenda 21, UNCED, Chapt. 18, p175). Diseases associated with water are heavily concentrated in the developing world, and within the developing world, among the poorer urban and rural households of the poorer countries.

Diseases arising from the ingestion of pathogens in contaminated water have the greatest impact worldwide. Table 1 shows the morbidity and mortality rates of the major water-related diseases. These figures provided to WHO by Member States are in many cases underestimated. For instance, no figures are available for certain diseases such as hepatitis which are often waterborne, some countries with numerous cases of typhoid do not report any to WHO, whilst others do not have the infrastructure to conduct the necessary surveys. There can be little doubt that true annual morbidity and mortality rates are well over these figures. It would be erroneous to ascribe these diseases exclusively to unsafe drinking-water. With the exception of dracunculiasis which is transmitted solely by drinking-water, a variety of non-water sources are also important.

Table 1: Morbidity and mortality rates of some important water-related diseases (after WHO, 1995)

The toll of human suffering from the microbial contamination of drinking-water is indeed heavy. As with the 1984 Guidelines, the 1993 Guidelines, justifiably, stress protection of water supplies from microbial contamination and call for uncompromised disinfection of drinking-water despite the potential formation during this process of compounds with potentially harmful long-term health effects.

Selection Criteria

Thousands of organisms and substances have been identified in drinking-water supplies around the world. It is neither necessary nor feasible to develop recommendations for all these.

Microorganisms selected for evaluation were selected through an international consultation process, on the basis of the presence in water and likely risk to human health. Particular emphasis was given to developing guidance on selection of indicator organisms that can give early warning of faecal contamination and likely potential risks of disease. The Guidelines adopted a clear policy from the outset that microbiological quality must be the key water quality priority.

Chemicals for evaluation were selected through an international consultative process, guided by three main criteria:

· The substance presents a potential hazard for human health;

· The substance was detected relatively frequently and at relatively high concentrations in drinking-water indicating that there may be significant exposure to humans;

· The substance was of major international concern (i.e. of interest to several countries).

On this basis, some 128 priority chemicals were selected for evaluation in the Guidelines and health-based acceptable levels of exposure from drinking-water (Guideline Values) recommended for 95 of these, taking into account all sources of exposure. Guideline values were not recommended for certain substances because they were found to be not hazardous to health, because of inadequate health effects information, or because the concentration of the chemical normally found in drinking-water does not represent a hazard to human health. Contaminants evaluated included chlorinated alkanes, ethylenes and benzenes, aromatic hydrocarbons, pesticides, inorganic chemicals, disinfectants and disinfectant by-products.

The Guideline Value

The recommendations made concerning water quality are expressed as Guideline values (GVs). Guideline values are not formal standards or regulatory limits and are not to be taken as strict limits such as "maximum permissible concentrations". They are intended to provide quantitative risk assessment information for regulatory authorities, risk managers, and others to make decisions concerning human health protection and to be adapted to national requirements and situations in prescribing limits and standards.

Guideline Values require adaptation because they relate to a "reference" human in a specified exposure environment. National populations and exposure situations will be different.

What is a guideline value?

· Guidelines are set for indicator bacteria - E.coli or thermotolerant (faecal) coliforms and total coliforms. These have been selected as they give a good indication of the likelihood of faecal contamination and the integrity of a water supply.

· Unlike chemical guideline values, the presence of indicator bacteria will always represent a health risk. However, when faecal contamination is indicated, water supplies should not be closed off unless a better source of water is available for use. The microbiological Guidelines should be used as a desirable end-point and improvement in microbiological water quality should be the priority for water supply.

· A guideline value represents the concentration of a chemical constituent that does not result in any significant risk to the health of the consumer over a lifetime of consumption.

· Short-term deviations above the guideline values do not necessarily mean that the water is unsuitable for consumption. The amount by which, and the period for which, any guideline value can be exceeded without affecting public health depends upon the specific substance involved.

· Although the guideline values describe a quality of water that is acceptable for life-long consumption, the establishment of these GVs should not be regarded as implying the quality of drinking water may be degraded to the recommended level. Indeed, a continuous effort should be made to maintain drinking-water quality at the highest possible level.

· When a guideline value is exceeded, the authority responsible for public health should be consulted for advice on suitable action, taking into account the intake of the substance from sources other than drinking-water (for chemical constituents), and the practicability of remedial measures.

· When developing national drinking-water standards based on these guideline values, it will be necessary to take account of a variety of geographical, socioeconomic, dietary and other conditions affecting potential exposure. This may lead to national standards that differ appreciably from the guideline values.

The recommended GVs must be both practical and feasible to implement as well as protective of public health. Guideline values are therefore not set at concentrations lower than the detection limits achievable under routine laboratory operating conditions. Moreover, guideline values are recommended only when control techniques are available to remove or reduce the concentration of the contaminant to the desired level.

Contrary to the 1984 Guidelines, the 1993 Guidelines do not propose guideline values for substances and parameters that affect the acceptability of drinking-water to consumers. The Review Groups were of the opinion that guideline values should be recommended only for those substances that are directly relevant to health.

Many of the inorganic and aesthetic constituents evaluated in the Guidelines are known to be essential for life. No attempt was made in the Guidelines to define minimum desirable concentrations of essential elements in drinking-water.

Contaminants derived from water treatment chemicals, construction materials, paints or coatings were not specifically addressed. The control of such contaminants is best accomplished by appropriate specifications for and control of the quality of the products themselves rather than the quality of the water.

The recommended guideline values are set at a level to protect human health; they may not be suitable for the protection of aquatic life.

The Guidelines apply to bottled water and ice intended for human consumption but do not apply to natural mineral waters, which should be regarded as beverages rather than drinking-water in the usual sense of the word. The Codex Alimentarius Commission has developed Codex standards for such mineral waters.

Future Revision

Understanding of water quality and the health risk from microbes and chemicals is constantly increasing and the knowledge base expanding. As a result, it has been agreed that there will be a continuing process of updating of the Guidelines with a number of substances or agents subject for evaluation each year. New editions of the Guidelines will be published at about ten-year intervals. For the 3rd edition of the Guidelines, the protection and control of water quality will be prioritised and issues such as development of monitoring and assessment methodologies in urban areas, resource and source protection and control of chemicals and materials used in water treatment fully addressed. This will lead to the preparation of a volume 4 of the Guidelines, either as a single volume or in the form of a series of documents in the Guidelines series.

Biennial addenda to the Guidelines are to be issued, beginning in 1997 containing evaluations of new substances or substances already evaluated for which new scientific information has become available. Substances for which provisional guideline values have been established will receive high priority for re-evaluation. Table 2 overleaf summarises the priorities for the first addendum in 1997.

Table 2: Priorities for the first addendum, 1997

Conclusions

The Guidelines are based on international consensus assessment of the risks to human health from the presence of microbial and chemical contaminants in drinking-water and provide a sound scientific basis for establishing standards with respect to health protection.

It is the hope of the Organization that the Guidelines will be utilized by governments at all levels to set new drinking-water quality standards where they do not yet exist, or to update and expand existing ones. Thus, legislators and policy makers now have access to more comprehensive and detailed information to match health criteria with economic and technological when establishing drinking-water quality standards.

Table 1: Priorities for the first addendum, 1997

Presentation Plan

WHO Guidelines for Drinking-water Quality

· Volume 1 - Recommendations
· Volume 2 - Health Criteria and other supporting information
· Volume 3 - Surveillance and control of community supplies

Consultation Process for Setting Guideline Values (Part 1)

Consultation Process for Setting Guideline Values (Part 2)

What is a Guideline Value?

· For microbes: no significant risk of pathogen presence at infectious dose.

· For most chemicals: no significant risk to health over a lifetime of consumption.

· Some chemicals (e.g. nitrate): no significant risk of acute intoxication of vulnerable group.

· National standards may be appreciably different from guideline values.

Priority Microbes considered in the GDWQ

· Orally transmitted pathogens of high priority (microbes associated with human faeces)

· Opportunistic and other water associated pathogens (moderate priority)

· Toxins from cyanobacteria

· Nuisance organisms causing rejection

· Guideline values for indicator bacteria and operational parameters

Selection Criteria

1. Adverse effects
2. Magnitude, frequency and duration of exposure
3. Population exposed
4. International concern

IARC Groups

IARC overall evaluation of chlorinated drinking-water: Group 3

Guideline Values

· No GV for individual pathogens: use indicator bacteria, turbidity and chlorine residual

· No GV for aesthetic parameters

· Treatment chemicals and construction materials not addressed

· No environmental effects

· Not for mineral water

· No minimum desirable level

Provisional Guideline Values

· Limited health effects information and/or UF>1000

· Health-based GV below quantitation level

· Health-based GV cannot be achieved through practical treatment methods

· Disinfection likely to result in health-based GV being exceeded

· GV at 10^-5 lifetime excess cancer risk not feasible

WHO Guidelines for Drinking-Water Quality

Protection and Control of Water Quality

Aim to include balanced, integrated guidance on monitoring and assessment of drinking-water supply and quality and on the elements of risk management in the Guidelines in 2003.

Monitoring and assessment of water supply and quality:

» Volume 3 coverage good for rural areas
» Guidance for urban settings will be developed, field tested and revised

Risk management:

» resource and source protection
» water treatment
» chemicals and materials
» significant expansion

Microbiological Aspects

Session Objectives

· To highlight the number and range of pathogens that may be found in water.

· To describe some of the key preventative and monitoring actions which maintain and improve microbiological water quality.

· To introduce the concept and use of indicator bacteria in water quality monitoring.

· To describe the principal indicator bacteria used and their key characteristics which make them suitable for use as indicators.

· To emphasise the value of E.coli and thermotolerant faecal coliforms as routine indicators.

Summary

The wide variety of waterborne diseases is the most important concern about water quality, and their public health impact has far-reaching implications. The pathogens concerned include many types of viruses, bacteria, protozoa and helminths, which differ widely in size, structure and composition. This implies that their survival in the environment and resistance to water treatment processes differs significantly. However, the waterborne transmission of infectious diseases can be controlled effectively by practical and economic methods. The approach must be based on protection of the source, selection of appropriate treatment methods, fail-safe application of the treatment methods, well protected distribution networks and appropriate quality monitoring. Relatively simple and inexpensive indicator methods are available for routine monitoring of the microbiological safety of water and the efficiency of treatment processes. Most reliable results are obtained by high frequency testing for indicator organisms selected for particular purposes. For instance, routine monitoring programmes for drinking-water may be based on tests for faecal streptococci, thermotolerant coliform organisms or Escherichia coli. Under certain circumstances, tests for the heterotrophic plate count and coliphages may be included. These tests are simple, inexpensive and yield results in a relatively short time. More complicated and expensive tests such as those for human viruses and protozoan parasites are required only for particular purposes, including research and assessment of the efficiency of treatment processes.

1. Introduction

Waterborne diseases are the most important concern about the quality of water. Developing countries and rural communities are particularly vulnerable. In developed countries the mortality due to waterborne diseases is low, but the socio-economic impact is phenomenal (Avendano et al, 1993; Payment, 1993).

Waterborne diseases are typically caused by enteric pathogens which belong to the group of organisms basically transmitted by the faecal-oral route. In other words, they are mainly excreted in faeces by infected individuals, and ingested by others in the form of faecally contaminated water or food. Some of the pathogens may be of animal origin. Some may also be transmitted by personal contact, droplet transfer, or inhalation of contaminated aerosols. Water may also play a role in the transmission of pathogens which are not faecally excreted. These include opportunistic pathogens which are members of the normal flora of the external human body. Some of these pathogens are natural inhabitants of certain water environments. Most waterborne pathogens are distributed world-wide, but outbreaks of some, for instance cholera and hepatitis E, tend to be regional. Dracunculiasis is geographically limited to rural areas in India, Pakistan, and sixteen countries in sub-Saharan Africa.

1.1 Enteric pathogens typically transmitted by the faecal-oral route

1.2 Helminths

Infections contracted by exposure to, or ingestion of, infectious larval stages of human parasites released by specific snails or cyclops:

Schistosoma spp (schistosomiasis, bilharziasis) and Dracunculus medinensis (dracunculiasis guinea worm). The latter is not faecally excreted but typically transmitted by water and of major public health importance in some countries.

1.3 Opportunistic pathogens

Infections of the skin and mucous membranes of the eye, ear, nose and throat:

Pseudomonas aeruginosa, Aeromonas, and species of Mycobacterium.

Infections contracted by the inhalation of contaminated aerosols:

Legionella spp (legionellosis), Naegleria fowleri (primary amoebic meningo-encephalitis) and Acanthamoeba spp (amoebic meningitis, pulmonary infections).

1.4 Toxins from cyanobacteria

Toxins released by blooms of cyanobacteria (blue-green algae) such as Microcystis aeruginosa may adversely affect the health of animals and possibly also humans.

1.5 Nuisance organisms

A variety of non-pathogenic micro-organisms, and small plants and animals, may under undesirable conditions thrive in water supplies and cause turbidity, taste and odour, or visible animal life, which are aesthetically objectionable.

Bacterial contamination of drinking-water has resulted in numerous cases of infectious disease. The massive cholera epidemic in Latin America, which spread from Peru to several other countries, and the recent one in Rwanda, are reminders of the speed with which certain waterborne diseases can spread.

Viruses feature prominently among the wide variety of waterborne pathogens. Examples include the 1991 outbreak with 70,000 cases of hepatitis E caused by polluted drinking-water in Kanpur (Grabow et al, 1994a). Reasons for the high incidence of waterborne viral infections include excretion in exceptionally high numbers by infected individuals, relatively high resistance to unfavourable environmental conditions including water treatment and disinfection processes, and a minimal infectious dose which may be as low as a single viable viral particle (Payment, 1993). The impact of viral infections is aggravated by secondary and even tertiary transmission by routes other than the water which caused the original infection (Morens et al, 1979). Epidemiological studies on waterborne viral infections are complicated by the absence of clinical symptoms in many individuals, particularly children, while all infected individuals excrete viruses at similar rates.

Recent years have seen a substantial increase in the number of waterborne Giardia and Cryptosporidium outbreaks. The cysts and oocysts of these protozoan parasites are extremely resistant to water treatment and disinfection processes, and their minimal infectious dose is low (Casemore, 1991; Craun, 1991).

Despite modern technology and know-how, waterborne diseases continue to have a major public health and socio-economic impact, and at least in parts of the world their incidence may even increase (Craun, 1991). Challenges to control waterborne diseases are complicated by continuous changes in the composition and priority of waterborne pathogens. Factors which affect the occurrence of pathogens include changes in population densities, socio-economic situations, standard of living, education, vaccination, climate, geography, urbanisation, migration and travelling, and public health policies.

The role of microbiological analysis is very important in a strategy for the control of waterborne diseases based on appropriate treatment systems, appropriate operation of the treatment systems, and appropriate quality monitoring.

2. Water Treatment and Disinfection Technology

A wide variety of treatment systems and disinfection processes are available to ensure the safety of water supplies. At the low technology and inexpensive end of the range there are methods such as simple sand filtration of water, the addition of household bleach to a bucket of drinking-water, storage of water, the exposure of water to sunlight, or boiling of drinking-water. At the other end of the range there are multiple-barrier treatment trains capable of the direct reclamation of drinking-water from waste water. All of these systems are capable of producing safe water. Consideration of the quality of available raw water sources is an integral part of the selection of appropriate treatment methods. The challenge is to select the appropriate system for each particular situation. Each situation has to be evaluated in its own merit, based on considerations such as the raw water quality, intended use of the water, financial resources, and technological capabilities.

3. Operation of Water Treatment Systems

The wide variety of treatment systems capable of producing safe water mentioned above, are without exception subject to potential breakdown and human failure in operation, supervision and quality surveillance. There is not even an indication that in the foreseeable future we can hope for a practical fail-safe water treatment system. Successful operation and supervision of treatment systems, improvement of technical capabilities, and training programmes aimed at meeting water quality requirements, are very important. The production of safe water is not possible without fail-safe operation and supervision of treatment systems (Bellamy, 1993).

4. Microbiological Water Quality Monitoring

Transmission of diseases by treated water supplies can be ascribed to inappropriate treatment methods, failure in operation and supervision, or shortcomings in quality monitoring. In fact, it can theoretically be argued that all waterborne diseases can be prevented by appropriate monitoring and corrective measures taken in good time. Since there is no indication that we can expect to see practical fail-safe treatment systems, or an elimination of human failure or error in the operation and supervision of treatment systems, appropriate microbiological quality monitoring will remain an indispensable component of strategies for the control of waterborne diseases.

Regular inspection of sanitary and hygienic aspects of raw water sources, treatment facilities and distribution networks is an important component of quality monitoring programmes, and is particularly important with regard to pathogens such as viruses and protozoan parasites which are not readily detectable in water.

4.1 Indicator organisms

Since it would be practically impossible to test water for each of the wide variety of pathogens that may be present, microbiological water quality monitoring is primarily based on tests for indicator organisms. There is no single indicator organism that can universally be used for all purposes of water quality surveillance. Each of the wide variety of indicators available for this purpose has its own advantages and disadvantages, and the challenge is to select the appropriate indicator, or combination of indicators, for each particular purpose of water quality assessment.

Indicators most commonly used are of faecal or sewage origin, and the following are some of the most important requirements of such indicators:

a) Present whenever pathogens are present.

b) Present in the same or higher numbers than pathogens.

c) Specific for faecal or sewage pollution.

d) At least as resistant as pathogens to conditions in natural water environments, and water purification and disinfection processes.

e) Non-pathogenic.

f) Detectable by simple, rapid and inexpensive methods.

Ideally, various other properties are desirable, such as counts which are directly related to those of pathogens. However, the fundamental and most important requirement is that pathogens should be absent or inactivated whenever indicators are absent or inactivated.

Many indicators have been studied and recommended for water quality assessment (ISO, 1990; Standard Methods 1992). Evaluation of the reliability of indicators is carried out by comparison of their incidence and survival in water and treatment processes with that of selected pathogens, by epidemiological studies on the consumers of water supplies, by calculations based on the minimal infectious dose of pathogens, and by experiments with human volunteers (Regli et al, 1991). The following is a summary of the most important features of commonly used indicators:

4.1.1 Escherichia coli

This species is a member of the group of faecal coliform bacteria. Escherichia coli has the important feature of being highly specific for the faeces of man and warm-blooded animals. For all practical purposes these bacteria cannot multiply in any natural water environment and they are, therefore, used as specific indicators for faecal pollution. They are generally distinguished from other thermotolerant coliforms by the ability to yield a positive indole test within 24 hours at 44.5°C. More recently, E. coli is also identified by possession of the enzyme �-glucuronidase, which hydrolyses the fluorogenic substrate 4-methyl-umbelliferyl-�-D-glucuronide (MUG) with release of the fluorogen which can be observed in liquid media under ultraviolet light. Media based on hydrolysis of MUG are commercially available under names such as "Colilert". Such complex sets of tests for the final confirmation of E. coli are not recommended as a routine.

4.1.2 Thermotolerant coliform bacteria

This term refers to certain members of the group of total coliform bacteria which are more closely related to faecal or sewage pollution, and which generally do not readily replicate in water environments. This group of bacteria is also known as faecal coliforms, presumptive E. coli, faecal E. coli, faecal coli, etc. Thermotolerant coliforms are primarily used for the assessment of faecal pollution in waste water and raw water sources. They are detectable by simple and inexpensive tests, and are widely used in routine water quality monitoring. The test methods used are the multiple tube and membrane filtration using mFC medium and incubation for 24 hours at 44.5°C. In the membrane filtration individual colonies can be identified, and the presence of Escherichia coli provides strong evidence of faecal pollution.

4.1.3 Coliform bacteria (total coliforms)

The term "coliform bacteria" refers to a vaguely defined group of Gram-negative bacteria which have a long history in water quality assessment. In outdated literature these bacteria go by all sorts of names, including coliforms, colis, etc. Some of the bacteria included in this group are almost conclusively of faecal origin, while other members may also replicate in suitable water environments. These bacteria, which can be determined by simple and inexpensive tests, are primarily used for assessment of the general sanitary quality of finally treated and disinfected drinking-water. Methods used are multiple tube or membrane filtration using LES Endo agar and incubation for 24 hours at 35-37°C. More recently coliform bacteria are also identified by their possession of the enzyme �-D-galactosidase, which hydrolyses chromogenic substrates such as ortho-nitrophenyl-�-D-galactopyranoside (ONPG), resulting in release of the chromogen and a colour change in liquid media.

The primary purpose of coliform tests is not to detect faecal pollution but to screen the general sanitary quality of treated drinking-water supplies.

4.1.4 Enterococci

Enterococci, sometimes referred to as faecal streptococci, is a group of bacteria more closely related to faecal pollution than total coliforms because most members of this group do not replicate as readily in water environments. These Gram-positive bacteria tend to be more resistant than faecal coliforms (Gram-negative), and are detectable by practical techniques, such as membrane filtration using m-enterococcus agar and incubation at 44.5° or 37°C for 48 hours. Presently the group is considered to primarily include only Enterococcus faecalis, E. faecium, E. durans and E. hirae. More recently enterococci are identified by the ability to hydrolyse 4-methyl-umbelliferyl-�-D-glucoside (MUD) in the presence of thallium acetate, nalidixic acid and 2,3,5-triphenyl-2H-tetrazolium chloride (TTC) resulting in release of the fluorogen which in liquid media is readily detectable under ultraviolet light.

4.1.5 Sulphite-reducing clostridia

An important advantage of these Gram-positive anaerobic bacteria is that their spores are more resistant to conditions in water environments, as well as treatment and disinfection processes, than most pathogens, including viruses. Clostridia are sometimes considered as too resistant, and their inclusion in water quality guidelines as too stringent. One of the members of the group, Clostridium perfringens, is like E. coli highly specific for faecal pollution. Clostridia generally occur in lower numbers in waste water than coliform bacteria. Detection methods are relatively expensive and time-consuming.

4.1.6 Heterotrophic plate count

This test is also known as the total or standard plate count. The test detects a wide variety of organisms, primarily bacteria, which give an indication of the general microbiological quality of water. The test is simple and inexpensive, yields results in a relatively short time, and has proved one of the most reliable and sensitive indicators of treatment or disinfection failure. The generally used test method is pour plates using a rich growth medium such as yeast extract agar and incubation for 48 hours at 37°C.

4.1.7 Other indicators

A variety of other indicators has been used in water quality assessment, including cytopathogenic human viruses, Pseudomonas aeruginosa, Staphylococcus aureus, acid-fast bacteria, Legionella species, Candida albicans, and endotoxins. All of these have advantages for certain purposes.

4.2 Protozoan parasites

The cysts and oocysts of intestinal parasites such as Giardia and Cryptosporidium species are exceptionally resistant, and they generally occur in low numbers in raw and treated water supplies (Casemore, 1991; Bellamy et al, 1993). In addition, they are not readily detectable, and their behaviour in water treatment and disinfection processes differs extensively from that of commonly used indicators. Quality control is, therefore, generally based on specifications for raw water quality and the efficiency of treatment processes rather than indicators or testing for cysts and oocysts (Regli et al, 1991).

4.3 Human viruses

The incidence and behaviour of human viruses in water environments and treatment processes may differ extensively from that of faecal indicators for reasons such as:

a) Viruses are excreted only by infected individuals, and coliform bacteria by almost all people and warm-blooded animals. Numbers of viruses in water environments are, therefore, generally lower than those of indicators such as faecal coliforms by several orders of magnitude.

b) Viruses are excreted for relatively short periods in numbers of up to 10¹²/g of faeces, while coliform bacteria are excreted fairly consistently in numbers of about 10⁹/g of faeces.

c) The structure, composition, morphology and size of viruses differs fundamentally from that of bacteria, which implies that behaviour and survival in water differ.

In view of the above differences it is not surprising that bacterial indicators such as coliform bacteria have shortcomings as indicators for viruses. These shortcomings have been confirmed in epidemiological studies and research on the incidence of indicators and viruses in water supplies. Ideally water quality surveillance should, therefore, include tests for viruses. Unfortunately, however, tests for viruses are relatively expensive, complicated and time consuming, and require sophisticated facilities and know-how. In addition, the great majority of viruses concerned are not detectable by conventional virological cell culture techniques. Control of the virological safety of water is, therefore, as in the case of protozoan parasites, often based on raw water quality and specifications for purification and disinfection processes rather than testing of the treated water (Regli et al, 1991).

4.4 New developments in microbiological water quality monitoring

4.4.1 Bacteriophages

Bacteriophages (phages) are viruses which infect bacteria. In terms of size, structure, morphology and composition they closely resemble human viruses. The behaviour of phages in water and related environments, and their resistance to unfavourable conditions, treatment systems and disinfection processes do, therefore, more closely resemble those of human viruses than bacterial indicators of faecal pollution.

Phages can replicate only in specific host bacteria, which implies that the phages of E. coli (coliphages) are, like their hosts, related to faecal pollution. Phages commonly used in water quality assessment include the groups of phages known as somatic and male-specific coliphages, which each have their own indicator advantages and disadvantages. Phages which infect Bacteroides fragilis strain HSP40 are highly specific for human faeces, and can be used to distinguish between faecal pollution of human and animal origin (Grabow et al, 1994b). Evidence supporting the indicator value of phages is accumulating, and their inclusion in quality monitoring protocols is gaining ground rapidly.

4.4.2 Virological analysis of water

Although desirable, virological analysis is not included in many routine surveillance protocols because of cost, complexity and time. In addition, the great majority of viruses concerned are not detectable by conventional techniques. However, progress is being made in the development of more practical and meaningful techniques, and virological monitoring for certain purposes is becoming more feasible. Challenges include the recovery of small numbers of viruses from large volumes of water, the detection of a wider variety of viruses, and reduction in the cost of testing (Standard Methods, 1992).

4.5 Indicator strategies in water quality surveillance

Since no single indicator can fulfil all the needs of water quality surveillance, best results are obtained by using appropriate combinations of indicators for various purposes. Each of these indicators offers certain information, which in combination yields a reliable picture of the quality of the water under investigation. For instance, indicators selected for monitoring the quality of treated drinking-water supplies may primarily be based on tests for Escherichia coli to detect faecal pollution. However, under certain circumstances indicators such as the heterotrophic plate count, total coliforms and somatic coliphages may yield valuable additional information.

Breakdown in treatment plants, and human error in operation and supervision, generally take place without warning, in fact, like a thief at night they tend to strike when least expected. This implies that quality surveillance programmes should make provision for microbiological monitoring at the highest possible frequency, in order to detect problems at the earliest possible stage. Since monitoring programmes are subject to many variables and considerations, including the raw water source and treatment system concerned, as well as available financial resources, facilities and manpower, it is not possible to formulate universal sampling protocols. Each case has to be evaluated in its own merit. With regards to sampling frequencies, it is important to keep in mind that it is better to run simple and inexpensive tests at high frequency than complicated and expensive tests at low frequency.

Important principles in sampling procedures include aseptic collection in sterile containers, and delivery at the laboratory for testing preferably within two hours of collection. The inclusion of samples collected at the consumer's tap is advisable, and so is the collection of samples at different times of the day and different days of the week.

5. Microbiological Water Quality Guidelines

Water quality guidelines and standards recommended by various authorities are similar in that they intend to ensure the minimum risk of infection. However, they differ in detail because of considerations such as economic and technical capabilities, and perceptions of acceptable risks of infection.

The Guidelines state that drinking-water must not contain waterborne pathogens. More specifically, E. coli or thermotolerant coliforms should not be present in 100 ml samples of drinking-water at any time, for any type of water supply, treated or untreated, piped or unpiped. In the case of large supplies, where sufficient numbers of samples are examined, total coliforms are acceptable in the distribution system in a maximum of 5% of samples taken throughout any 12 month period (Annex 1).

If guideline values are exceeded, immediate investigative action must be taken, including repeat testing, and thorough inspection of the treatment plant and its operation, the raw water source, and general hygiene of the water distribution system.

It is recognised that, in the great majority of rural water supplies in developing countries, faecal contamination is widespread and achieving the guideline values for E. coli or thermotolerant coliforms is often not possible. Under these conditions, the national surveillance agency should set medium-term targets for the progressive improvement of water supplies, as recommended in Volume 3 of Guidelines for drinking-water quality (Surveillance and control of community supplies).

Because routine monitoring techniques are not available for viruses, protozoa and helminths of health significance, the Guidelines recommend protection of the source and treatment techniques to ensure their absence. The degree of treatment required is a function of the nature (ground or surface water) and level of faecal contamination of the source.

To ensure the absence of viruses, the Guidelines recommend that the following conditions of disinfection with chlorine be met:

The control of pathogenic protozoa and guinea-worm requires efficient filtration since these organisms are rather resistant to disinfection.

6. Conclusions

· Waterborne diseases have a major public health and socio-economic impact, and are the most important concern of water quality.

· Strategies for the control of waterborne diseases must be based on the selection of appropriate water sources and treatment systems, fail-safe operation of these treatment systems, and reliable bacteriological quality monitoring.

· A wide variety of treatment and disinfection systems is available for reliable production of microbiologically safe drinking-water.

· Reliable guidelines for the microbiological safety of drinking-water are available.

References

Avendano P, Matson D O, Long J, Whitney S, Matson C C and Pickering L K (1993). Costs associated with office visits for diarrhea in infants and toddlers. Pediatric Infectious Diseases Journal 12, 897-902.

Bellamy W D, Cleasby J L, Logsdon G S and Allen M J (1993). Assessing treatment plant performance. Journal of the American Water Works Association 85, 34-38.

Casemore D P (1991). The epidemiology of human cryptosporidiosis and the water route of infection. Water Science and Technology 24, 157-164.

Craun G F (1991). Causes of waterborne outbreaks in the United States. Water Science and Technology 24, 17-20.

Grabow W O K, Favorov M O, Khudyakova N S, Taylor M B, and Fields H A (1994a). Hepatitis E seroprevalence in selected individuals in South Africa. Journal of Medical Virology (in press).

Grabow W O K, Neubrech T E, Holtzhausen C S and Jofre J (1994b). Bacteroides fragilis and Escherichia coli bacteriophages: excretion by humans and animals. Water Science and Technology. (Submitted for publication).

ISO (1990). Detection and enumeration of coliform organisms, thermotolerant coliform organisms, and presumptive Escherichia coli.

Part 1: Membrane filtration method (9308-1:1990).
Detection and enumeration of coliform organisms, thermotolerant coliform organisms, and presumptive Escherichia coli.
Part 2: Multiple tube (most probable number) method (9308-2:1990).

Morens D M, Zweighaft R M, Vernon T M, Gary G W, Eslien J J, Wood B T, Holman R C and Dolin R (1979). A waterborne outbreak of gastroenteritis with secondary person-to-person spread: Association with a viral agent. The Lancet i, 964-966.

Payment P (1993). Prevalence of disease, levels and sources. In: Safety of Water Disinfection: Balancing Chemical and Microbial Risks. International Life Sciences Institute, Washington DC.

Regli S, Rose J B, Haas C N and Gerba C P (1991). Modelling the risk from giardia and viruses in drinking-water. Journal of the American Water Works Association 92, 76-84.

Standard Methods (1992). Standard Methods for the Examination of Water and Wastewater, 18th Edition. Eds.: A E, Clesceri L S, Eaton A D. American Public Health Association, Washington DC. pp 9-1 - 9-147.

Annex 1
BACTERIOLOGICAL QUALITY OF DRINKING-WATER

Presentation Plan

Problems in Setting Guideline Values for Individual Pathogens

· Pathogens are discrete and not in solution

· Pathogens often in clumps or adhere to suspended solids

· Cannot predict likelihood of infectious dose from average concentration

· Infection and disease development dependent on invasiveness, virulence and immunity

· Dose-response not cumulative

Ensuring Microbiological Quality

· Water quality analysis
· Sanitary inspection
· Source protection
· Minimum treatment requirements

Need for Faecal Indicator Organisms

· Pathogen analysis:

» expensive
» impracticable
» techniques may be time consuming
» reactive

· Reliance is therefore placed on relatively simple and more rapid tests for the detection of certain intestinal bacteria which indicate that faecal contamination could be present.

Characteristics of the Ideal Faecal Indicator

1. Should be present in wastewater and contaminated water when there are pathogens

2. Should be present when there is a risk of contamination by pathogens

3. Should be present in greater numbers than the pathogens

4. Should not multiply in environmental conditions under which pathogens cannot multiply

5. The indicator population should correlate with the degree of faecal contamination.

6. The survival time in unfavourable environmental conditions should exceed that of pathogens

7. Should be more resistant to disinfectants and other stresses than the pathogens

8. Should present no health risk

9. Should be easy to enumerate and identify by simple methods

10. Should have stable characteristics and give consistent reactions in these analyses

Examples of Indicator Organisms

Principal Indicator Bacteria

Principal indicator bacteria are:

· Escherichia coli
· Faecal coliforms (95% are E.coli, ± 44^oC)
· Faecal streptococci

Table 1: Examples of Pathogens Considered in the Guidelines

Disinfectants and Disinfection By-Products

Session Objectives

· To describe the importance of disinfection in providing safe drinking water.

· To describe the key disinfectants evaluated in the Guidelines and describe their principal characteristics and effectiveness.

· To describe the key by-products formed by the principal disinfectants and describe the likely health risk from their presence in water.

· To describe the balance between microbiological and chemical health risks and emphasise the need to prioritise microbiological quality.

Introduction

Disinfection of drinking-water is essential if we are to protect the public from outbreaks of waterborne infectious and parasitic diseases. The main disinfectants evaluated in the Guidelines are free chlorine, chloramines, chlorine dioxide and ozone.

As much as the perfect indicator organism does not exist, each of the commonly used disinfectants has its advantages and disadvantages in terms of cost, efficacy, stability, ease of application and formation of by-products.

Table 1 summarizes the Ct values for the four main disinfectant,

where

C = disinfectant concentration in mg/litre, and
t = the contact time in minutes required to inactivate a specified percentage of microorganisms.

Table 1: Summary of C.t values (mg/L. min)for 99% inactivation at 5°C (Clark et al, 1993)

^a Values for 99.9% inactivation at pH 6-9.
^b 99% inactivation at pH 7 and 25°C.
^c 90% inactivation at pH 7 and 25°C.

From the Ct values, ozone is the most efficient and chloramine the least efficient, particularly for viral agents. Free chlorine is more effective than chlorine dioxide with regard to E. coli and rotavirus. Chlorine dioxide is more effective than free chlorine with regard to the protozoa Giardia lamblia and muris. Ozone is the most efficient disinfectant for cryptosporidium parvum. As the temperature increases, the Ct values decrease for all disinfectants. The effect of pH varies with the nature of the disinfectant and is most pronounced for chlorine.

Chlorine and its by-products

Chlorine is the most widely used drinking-water disinfectant. When added to water the following reaction occurs within a second or less:

Cl₂ + H₂O = HOCl + H⁺ + Cl^-

The magnitude of the equilibrium hydrolysis constant is such that hydrolysis to hypochlorous acid, HOCl, is virtually complete in fresh water at pH > 4 and at chlorine doses up to 100 mg/litre (Morris, 1982).

Hypochlorous acid is a weak acid that dissociates partially in water as follows:

HOCl = H⁺ + OCl^-

The value of the acid ionization constant is about 3 × 10^-8. As shown in Figure 1, at 20°C and pH 7.5, there is an equal distribution of HOCl and OCl^-. At pH 8, about 30% of the free chlorine is present as HOCl, and at pH 6.5, 90% is present as HOCl (Morris, 1982). The term free chlorine refers to the sum of hypochlorous acid and hypochlorite ion. Since HOCl is a considerably more efficient disinfectant than OCl^-, and free chlorine, even as hypochlorite, is more effective than combined chlorine (e.g. chloramines), the Guidelines recommend that disinfection be carried out at pH less than 8 and at a free chlorine concentration ³ 0.5 mg/litre.

Of all the disinfectants, the chemistry and toxicity of the reaction by-products of chlorine have been the most extensively studied.

Figure 1: Distribution of hypochlorous acid and hypochlorite ion in water at different pH values and temperatures (Morris, 1951)

Since Rook's discovery of the formation of haloforms during chlorination of Rotterdam water supply (Rook, 1974), numerous halogenated compounds have been identified in chlorinated drinking-water and their toxicity assessed. Precursors of these halogenated compounds include natural humic and fulvic compounds and algal material. The most commonly found chlorine disinfection by-products are the trihalomethanes (THM), halogenated acetic acids, halogenated acetonitriles, chloral hydrate and the chlorinated phenols. Others include chlorinated furanone MX, halopicrins, cyanogen halides, haloketones and haloaldehydes. The halogenated disinfection by-products identified account for only about half of the total formed.

Based on animal toxicological studies, Guideline Values (GVs) have been recommended for a number of these compounds. Undoubtedly, the third edition of the Guidelines, planned for the year 2002, will include additional chlorination by-products.

The following chemicals resulting from chlorination of water supplies have been evaluated in the Guidelines:

· free chlorine (HOCl + OCl^-)
· trihalomethanes
· chlorinated acetic acids
· halogenated acetonitriles
· chloral hydrate (trichloroacetaldehyde)
· chlorophenols
· MX (3-chloro-4-dichloromethyl-5-hydroxy-2(5H)-furanone)

For countries wishing to control DBP, it may not be necessary to set standards for all of the DBP for which guideline values have been proposed. The trihalomethanes, of which chloroform is the major component, are likely to be the main DBP, together with the chlorinated acetic acids in some instances. In many cases, control of chloroform levels and, where appropriate, trichloroacetic acid will also provide an adequate measure of control over other chlorination by-products.

(a) Chlorine

Free chlorine in drinking-water is not particularly toxic to humans. The major source of exposure to chlorine is drinking-water. Therefore, 100% of the TDI was allocated to drinking-water giving a health-based GV of 5 mg/litre for the sum of hypochlorous acid and hypochlorite ion. Based on the taste and odour threshold of free chlorine, it is doubtful however that consumers would tolerate such a high level of chlorine. Most individuals are able to taste chlorine at concentrations below 5mg/litre, and some at levels as low as 0.3 mg/litre. The health-based GV for chlorine should not be interpreted as a desirable level of chlorination.

(b) Trihalomethanes

The predominant chlorine disinfection by-products are the THMs. Nevertheless, they account for only about 10% of the total organic halogen compounds formed by water chlorination.

THMs are formed by the aqueous chlorination of humic substances, of soluble compounds secreted from algae and of naturally occurring nitrogenous compounds (Morris, 1982). THMs consist primarily of chloroform, bromodichloromethane, dibromochloromethane and bromoform.

When bromide is present in drinking-water, it is oxidized to hypobromous acid by chlorine:

HOCl + Br^- = HOBr + Cl^-

HOBr reacts with natural organic compounds to form brominated halomethanes. Similarly, the presence of iodide may lead to the formation of mixed chlorobromoiodo-methanes.

Some generalized statements can be made with regard to THMs in chlorinated drinking-water (IARC, 1991; Morris, 1982; Canada, 1993):

· Concentration of THMs in drinking-water varies widely and ranges from not detectable to 1 mg/litre or more;

· THM levels are higher in chlorinated surface water than in chlorinated groundwater;

· Concentrations of THMs tend to increase with increasing temperature, pH and chlorine dosage;

· Concentrations of THMs increase upon storage even after exhaustion of residual chlorine or after dechlorination. This indicates the formation of intermediates products leading to the slow production of THMs;

· Chloroform is usually the most abundant THM often accounting for greater than 90% of the total THM concentration;

· If there is a significant amount of bromide in the raw water, the brominated THMs, including bromoform, may be dominant;

· Formation of THMs can be minimized by avoiding pre-chlorination and by effective coagulation, sedimentation and filtration to remove organic precursors prior to final disinfection;

· Removal of THMs after their formation is difficult and involves resource-intensive processes such as activated carbon adsorption or air stripping.

Because trihalomethanes usually occur together, it has been the practice to consider total trihalomethanes as a group, and a number of countries have set guidelines or standards on this basis, ranging from 0.025 to 0.25 mg/litre.

In the 1993 WHO Guidelines, individual GV have been recommended for the four trihalomethanes. With an underlying assumption that the THMs may exert potential toxic effects through similar biological mechanisms, authorities may want to establish standards for total THMs that would account for possible additive effects and not simply add up the guideline values for the individual compounds in order to arrive at a standard. Instead, the following approach is recommended:

where

C = concentration, and
GV = guideline value

Epidemiological studies of carcinogenicity of chlorine and DBP

In 1991, WHO International Agency for Research on Cancer (IARC) published an evaluation of the carcinogenic risks to humans of chlorinated drinking-water based on a number of animal toxicological and epidemiological studies. IARC concluded that because of one or more methodological weaknesses, the epidemiological studies reviewed cannot constitute the basis of valid risk assessment.

The epidemiological investigation of the relation between exposure to chlorinated drinking-water and cancer occurrence was considered problematic because any increase in relative risk over that in people drinking unchlorinated water is likely to be small and therefore difficult to detect in epidemiological studies. In all of the studies evaluated, estimates of exposure were imprecise and surrogates (e.g surface versus groundwater) do not reflect exposure during the relevant time periods for the etiology of the cancers in question. Many variables, such as smoking habits, dietary practices, use of alcohol, socio-economic status, and ethnicity are known to affect cancer incidence and were not taken into account in most of the studies (IARC, 1991).

In its overall evaluation, IARC concluded that there is inadequate evidence for the carcinogenicity of chlorinated drinking-water in humans as well as in experimental animals (IARC, 1991).

Chloramine and its by-products

Chloramine generally produces by-products similar to those observed with chlorine but at much lower concentrations. An exception to this is the formation of cyanogen chloride, CNCl (Bull and Kopfler, 1991). The use of chloramine as a disinfectant has increased in recent years because of limited formation of THMs, however, little is known about the nature of other byproducts.

Monochloramine is about 2000 and 100 000 times less effective than free chlorine for the inactivation of E. coli and rotaviruses, respectively. Monochloramine cannot therefore be relied upon as primary disinfectant. It is useful for maintaining a residual disinfectant in distribution systems. The shift to monochloramine to control THM formation may thus compromise disinfection and the Guidelines caution against such procedure. Organic chloramines are even less effective disinfectants than monochloramine.

Chlorine dioxide and its by-products

Because of its explosive hazard, chlorine dioxide is manufactured at the point of use. CLO₂ is generated through the reaction of sodium chlorite and chlorine. Chlorine dioxide reactions with humic substances do not form significant levels of THMs. In addition, it does not react with ammonia to form chloramines. The main disinfection by-products of chlorine dioxide are chloride, chlorate and chlorite.

Chlorine dioxide is more effective towards inactivation of Giardia cysts than free chlorine, but less effective towards rotavirus and E. coli. Unlike chlorine, the disinfection efficiency of chlorine dioxide is independent of pH and the presence of ammonia.

A provisional GV was recommended for chlorite while no adequate data were available to recommend a GV for chlorate. No GV has been recommended for chlorine dioxide per se because of its rapid breakdown in aqueous solutions and the chlorite GV is adequately protective for potential toxicity from chlorine dioxide. Furthermore, the taste and odour threshold for chlorine dioxide in water is 0.4 mg/litre which constitutes a limiting factor and a signal for its presence at higher concentrations in drinking-water.

Other reaction by-products of chlorine dioxide with organics in drinking-water have not been well characterized but include aldehydes, carboxylic acids, haloacids, chlorophenols, quinones and benzoquinone (Bull and Kopfler, 1991). In a recent article, more than 40 organic disinfection by-products were identified in a pilot plant in Indiana which uses chlorine dioxide as a primary disinfectant. The toxicity of these by-products is largely unknown (Richardson et al. 1994).

Ozone and its by-products

Ozone decomposes rapidly following application, and for this reason no GV has been proposed for ozone.

By products of ozonation that have been identified include formaldehyde and other aldehydes, carboxylic acids, hydrogen peroxide, bromate, bromomethanes, brominated acetic acids, brominated acetonitriles and ketones. Guideline values have been recommended for bromate and formaldehyde.

Ozone is the most efficient disinfectant for all types of microorganisms. Disadvantages include lack of disinfectant residual, biological regrowth problems in distribution systems, high cost, and limited information on the nature and toxicity of its by-products.

Balancing Chemical and Microbial Risks

Quantitative assessments of risks associated with the microbial contamination of drinking-water are scarce. Although there are gaps in our knowledge, we cannot afford to postpone action until rigorous quantitative assessment of chemical versus microbial risks are available and every answer is known.

A semi-quantitative presentation of risks associated with disinfection was first attempted by Morris (1978) and is given in Figure 2. The following is more or less a quote of his work: The risk of waterborne infectious disease is very high when no chlorination is used, and drops sharply to a low value when even minimal levels of chlorination are maintained. We know this on the basis of a century's experience, Morris stated. As the level of chlorination is increased the risk continues to drop slightly, but never quite reaches zero, for no system is perfect. At very high levels of chlorine the microbial risk increases as taste and odour may cause the use of unsafe supplies.

Figure 2: Risks and benefits of water chlorination (Morris, 1978)

The chemical risk does not start at zero for there is some hazard connected with the organic matter before chlorination. The chemical risk decreases initially because destruction of chemicals by oxidation more than compensate for the formation of new chemicals at low levels of chlorination. Because of the formation of by-products, the chemical risk increases with increasing level of chlorination. Intuitively, he depicted the chemical risk from chlorination as being considerably lower than the microbial risk from a non-disinfected supply.

In developed countries, since filtration and chlorination became common for community water supplies, morbidity and mortality due to waterborne intestinal diseases, particularly typhoid fever and cholera, have declined to negligible levels. Almost all of the waterborne outbreaks that still occur are associated with the use of untreated water or water from systems in which chlorination was inadequate.

Other health impact studies concern the beneficial effects on health of safe and sufficient water supplies and adequate sanitation, three factors that are so intertwined that it is often not feasible to draw definite lines of demarcations between them. Together, they constitute the pillars of public health protection. Projected reduction in morbidity achievable through the provision of safe and sufficient water supplies and adequate sanitation are estimated to be (WHO, 1992):

When applying these percentage reductions to the global morbidity and mortality rates for these diseases, the benefits of saving millions of lives through these interventions are immediately apparent.

As shown in Figure 3 overleaf, provision of safe drinking-water can result in a 20% reduction in infant mortality (Regli et al., 1993).

In their pioneering work on comparison of estimated risk from known pathogens in untreated surface water and chlorination by products in drinking-water, Regli et al. (1993) concluded that:

· the risk of death from pathogens is at least 100 to 1000 times greater than the risk of cancer from disinfection by-products (DBPs);

· the risk of illness from pathogens is at least 10 000 to 1 million times greater than the risk of cancer from DBPs;

· morbidity and mortality rates from pathogens compared with those from DBPs, may be considerably higher in developing countries where the sanitary and health status is not as good;

· in societies where infant mortality and life expectancy is low, many people would not be expected to live long enough to incur cancer, which also causes much higher differences in risk resulting from exposure to pathogens versus DBPs cited above.

While this last statement seems cynical, it does reflect the true situation in many developing countries.

Figure 3: Infant mortality versus access to safe water (Regli et. al., 1993)

Conclusion

Adequate disinfection of drinking-water is the most important priority to assure a safe water supply. Recent cholera outbreaks in Latin America and Rwanda provide dramatic evidence of the importance of adequate water disinfection. There is some limited evidence of possible health effects from disinfectant by-products, particularly possible cancer risks from chloroform and the other trihalomethanes and by-products. This evidence is based on high-dose animal studies.

Epidemiological studies conducted to date do not provide any evidence that disinfectants and their by-products affect human health at the concentrations found in drinking-water. The International Agency for Research on Cancer has concluded that there is inadequate evidence for the carcinogenicity of chlorinated drinking-water in humans and experimental animals.

Although stated in qualitative way, the message of the Guidelines is clear:

The estimated risks to health from disinfectants and their by-products are extremely small in comparison to the real risks associated with inadequate disinfection, and it is important that disinfection should not be compromised in attempting to control such by-products. The destruction of microbial pathogens through the use of disinfectants is essential for the protection of public health.

All disinfectants by necessity are reactive substances and produce by-products. Little is known about the nature and toxicity of the by-products of ozone, chlorine dioxide or chloramines. The by-products of chlorination are the ones that have been most extensively identified and their toxicity assessed. Disinfection with chlorine should not be penalized for this reason. In addition, in many countries, if disinfection can be practised at all, it will be through the use of chlorine.

There are now more and more indication that the estimated risks to health from disinfectants and their by-products are several order of magnitude lower than the real risks associated with inadequate disinfection. So while there is great scientific certainty that inadequately disinfected water results in devastating microbial disease epidemics, there is relatively great uncertainty regarding the possible health risks from DDBPs. In establishing standards for disinfectants by products, it is emphasized that "Where local circumstances require that a choice must be made between meeting either microbiological guidelines or guidelines for disinfectants or disinfectant by-products, the microbiological quality must always take precedence, and where necessary, a chemical guideline value can be adopted at a higher level of risk. Efficient disinfection must never be compromised." (1993 Guidelines)

References

Bull R.J., Kopfler F.C. (1991). Health Effects of Disinfectants and Disinfection By-products, American Water Works Association, Denver.

Canada Health and Welfare (1993). Water treatment principles and applications. Canadian water and wastewater association, Ottawa.

Clark RM, Hurst CJ, Regli S (1993). Costs and benefits of pathogen control in drinking-water. In: Safety of water disinfection: balancing chemical and microbial risks. Craun G.F. ed. ILSI Press, Washington, D.C.

International Agency for Research on Cancer (1991). IARC monographs on the evaluation of carcinogenic risks to humans, Volume 52 Chlorinated drinking-water; chlorination by-products; some other halogenated compounds; cobalt and cobalt compounds, Lyon.

Morris J.C. (1951), unpublished research, Harvard University, 1951

Morris J.C. (1978). Conference summary. In: Water chlorination environmental impact and health effects. Volume 2. Jolley RL, Gorchev H., Hamilton D.H. eds., Ann Arbor Science, Ann Arbor, Michigan.

Morris J.C. (1982). Health perspective in the oxidative treatment of water for potable supply. Part 2 Health assessment of current oxidant-disinfectants. National Institute for Water Supply. Leidschendam, Netherlands.

Regli S., Berger P., Macler B., Haas C. (1993). Proposed decision tree for management of risks in drinking-water: consideration for health and socioeconomic factors. In: Safety of water disinfection: balancing chemical and microbial risks. Craun G.F. ed. ILSI Press, Washington, D.C.

Richardson S.D., Thruston A.D., Collette T.W. (1994). Multispectral identification of chlorine dioxide disinfection byproducts in drinking-water. Environ. Sci. Technol., 28: 592-599.

Rook J.J. (1974). Formation of haloforms during chlorination of natural waters. J. Soc. for water treatment and examination 23: 234-243.

World Health Organization (1992). Our planet, our health. Report of the WHO Commission on health and environment. Geneva.

Annex 1
BACTERIOLOGICAL QUALITY OF DRINKING-WATER

Table 1: Summary of C.t values (mg/L. min) for 99% inactivation at 5°C (Clark et al, 1993)

^a Values for 99.9% inactivation at pH 6-9.
^b 99% inactivation at pH 7 and 25°C.
^c 90% inactivation at pH 7 and 25°C.

Presentation Plan

Disinfectants Evaluated

· Chlorine
· Chloramine
· Chlorine dioxide
· Ozone
· Iodine

Disinfectants and Disinfectant by-products

· Overall ozone is the most effective disinfectant, although chlorine is effective and efficient

· All disinfectants have advantages and disadvantages and all produce by-products

· A number of disinfectant by-products were evaluated in the Guidelines

Microbiological quality of water should never be compromised by concerns about disinfection by-products

Distribution of Hypochlorous Acid and Hypochlorite Ion in Water at Different pH Values and Temperatures

Relationship between Measured Free Residual Available Chlorine (HOCl⁺, OCl^-) and Bactericidally Active (HOCl)

Chlorine

· Chlorine is the most common disinfectant

· Chlorine by-products

» Free chlorine
» Trihalomethanes (THMs)
» Chlorinated acetics acids
» Halogenated acetonitriles
» Chloral hydrate (trichloroacetaldehyde)
» Chlorophenols
» MX

(3-chloro-dichlormethyl-5-hydroxy-2(5H)-furanone)

May not need to set standards for all by-products included in the Guidelines, it is better to concentrate on the major groups (e.g. THMs)

Trihalomethanes

· The principal by-product of chlorination

· Formed by the aqueous chlorination of humic substances

· More likely to occur in chlorinated surface water than groundwater

· Concentrations of THMs tend to increase with increaseing temperature, pH and chlroine dosage

· THMs consist primarily of:

» Chlroform
» Bromodichloromethane
» Dibromochloromethane
» Bromoform

· Formation of THMs can be minimised by avoiding prechlorination and optimising treatment

Chloramine and its By-products

· Chloramines formed by reaction of chlorine and ammonia or organic amines

· Mono-, di- and trichloramines may be formed depending upon pH and temperature

· Chloramine by-products similar to free chlorine with the exception of cyanogen chloride

· Mono-chloramine is a less effective disinfectant than free chlorine and cannot be relied upon as a primary disinfectant; though useful for maintaining a residual.

Chlorine dioxide and its By-products

· Chlorine dioxide made at point of use because of its explosive hazard

· Reactions with humic substances do not form significant levels of THMs or chloramines

· Main by-products are:

» chlorite
» chlorate
» chloride

· More effective than free chlorine in inactivation of Giardia cysts but less effective against E.coli and rotaviruses

· No GV for chlorine dioxide in water as it dissociates rapidly. GVs set for chlorite but not chlorate

Ozone and its By-products

· Most efficient disinfectant for all types of micro-organisims

· Decomposes rapidly following application thus no GV has been proposed for ozone

· By-products include:

» formaldehyde
» aldehydes
» hydrogen peroxide
» bromomethanes

· Disdavantges include:

» lack of residual
» biological regrowth in distribution systems
» high cost
» limited information on toxicity of its by-products

Balancing chemical and microbiological risks

Inorganic Constituents and Aesthetic Parameters

Session Objectives

· To describe the process of setting Guideline Values for inorganic parameters and describe the narrow divide between toxic and essential elements.

· To describe some basic physico-chemical characteristics of water.

· To provide some examples of inorganic chemicals to illustrate the uses of GVs for inorganics and highlight priority substances.

· To describe the basis of monitoring of physical and chemical parameters.

All substances are poisons; there is none which is not a poison. The right dose differentiates a poison and a remedy.

PARACELSUS (1493-1541)

Introduction

Many of the inorganic and aesthetic constituents evaluated in the Guidelines are known to be essential for life. Chromium, copper, fluoride, iodine, manganese, molybdenum, and selenium are essential elements in human nutrition; arsenic and nickel are considered by some researchers as essential elements. Of the aesthetic constituents, iron, chloride, calcium and magnesium (hardness), sodium and zinc are essential elements.

A classification into "essential" and "toxic" elements is fraught with difficulties since as science advances, there is a constant shift of the elements from one group to the other. Toxicity is inherent in all elements, and is a function of the concentration to which humans are exposed. Paracelsus' statement remains valid, "The right dose differentiates a poison and a remedy". Ordinary salt, calcium, magnesium and iron, are all toxic above certain doses. This is illustrated in Figure 1 below.

The plateau of "safe and adequate intake of essential elements" are mainly a matter for nutritionists to decide.

Figure 1. Dose-response curve (Source: Galal-Gorchev, 1995)

Notes: NOAEL: no observed adverse effect limit; LOAEL: lowest observed adverse effect limit; Threshold chemicals: only toxic above certain dose, therefore have a NOAEL; Non-threshold chemical: toxic at any dose, therefore do not have a NOAEL

No attempt has been made in the Guidelines to define a minimum desirable concentration of essential elements in drinking-water. The Guidelines are concerned with the quantification of toxic effects.

A few of the inorganic chemicals and aesthetic constituents of common interest to many countries will be discussed here to illustrate the approach taken in deriving, or not deriving, guideline values (GV). A complete list of inorganic substances for which GVs have been derived is given in Table A2.2 (A) in Annex 2 of Volume 1 of the Guidelines.

Inorganic Constituents

Asbestos in drinking-water: no health hazard

Because of numerous inquiries from governments, industry and academia on the potential adverse health effects from asbestos in drinking-water, WHO issued the attached Press Release.

Asbestos is used in a large number of applications, particularly construction materials, such as asbestos cement (A/C) sheet and pipe, electrical and thermal insulation, and friction products, such as brake linings.

Asbestos is introduced into water by the natural dissolution of asbestos-containing minerals as well as from industrial effluents, atmospheric pollution, and A/C pipes in water distribution systems. High levels of asbestos have been found in drinking-water from corrosion of A/C pipes.

The asbestos content of food has not been well studied because of the lack of a simple, reliable analytical method. Based on crude estimates, intake of asbestos in food may be significant compared with that in drinking-water. Concentrations of 7 million fibres per litre (MFL) in beer and 12 MFL in soft drinks have been reported.

Asbestos is a known human carcinogen by the inhalation route. Based on the inhalation route, IARC has assigned it to Group 1 (the agent is carcinogenic to humans), while recognising that asbestos behaved differently by the oral route.

Asbestos was not found to be carcinogenic in several animal feeding studies. Epidemiological studies of population exposed to high levels of asbestos in drinking-water (200 MFL) did not reveal any excess cancer risk. It was therefore concluded that ingested asbestos is not hazardous to health and there was no need to establish a GV for asbestos in drinking-water.

Another question that needs to be answered is: Can high concentration of asbestos fibres in drinking-water become airborne and create a health hazard?

In a study in New York State, asbestos contamination in excess of 10 billion fibres per litre was detected in a community's drinking-water. Mean airborne asbestos concentrations were significantly higher in a small number of homes with water containing this elevated concentration of asbestos than in three control homes; however, the difference in concentrations was primarily due to increased numbers of short (<1 µm) fibres, which are considered to contribute little to health risk. Moreover, all fibre concentrations determined in this limited study were within the range of those measured in indoor and outdoor air in other investigations.

In another study, using a conventional drum-type humidifier, testing showed that release of asbestos fibres to air from water containing 40 ± 10 MFL was negligible.

The final question - does corrosive water transported in A/C pipe pose any specific or unique health risk? Corrosive water does not create a specific health risk as it relates to A/C pipe since asbestos fibres in drinking-water do not pose a health risk and are not transferred into the air. However, corrosive water is an important problem that must be addressed by all water utilities no matter what type of water pipe material is used in the distribution system or homes. Proper selection of the quality of A/C pipes is important and some national institutions have issued standards for A/C pipes suitable for water with different degrees of aggressiveness.

Fluoride and dental health

Fluoride levels between 0.5 and 1 mg/litre provide substantial protection against dental caries. However, for fluoride, the margin between beneficial and toxic effects is rather small (see Figure 1). Excessive exposure may lead to adverse health effects varying from mottling of teeth to crippling skeletal fluorosis.

The Guidelines recommend a GV of 1.5 mg/litre on the assumption that the daily per capita consumption of drinking-water is about 2 litres. At this level, dental fluorosis may occur in a certain proportion of the population. In setting national standards for fluoride, it is particularly important to consider climatic conditions, volumes of water intake, and intake of fluoride from other sources (e.g. food, air).

Nitrate and nitrite

Nitrate and nitrite in drinking-water may be of natural origin, or can be leached from septic tanks, pig farms and feed lots. Use of fertilizers (too much and at the wrong time of the year) can also result in nitrate pollution.

High concentration of nitrate, and especially nitrite in drinking-water may cause methaemoglobinaemia. Groups especially susceptible to methaemoglobin formation are young infants, children and pregnant women.

Epidemiological studies indicate that at levels of nitrates less than 50 mg/litre (as nitrate), there does not seem to be any problem with methaemoglobinaemia. There are considerable uncertainties as to the level of nitrite which may cause such clinical effects. On the assumption that nitrite was ten times more potent than nitrate (on a molar basis) with respect to methaemoglobin formation, the Guidelines recommend a provisional GV of 3 mg/litre (as nitrite). In addition, since nitrite and nitrate exert similar toxicological effect and may occur simultaneously, the following condition was also specified:

where

C = concentration in drinking-water
GV = guideline value

Lead and IQ

In 1986 the Joint FAO/WHO Expert Committee on food Additives (JEFCA) established a provisional tolerable weekly intake (PTWI) of lead from all sources of 25 µg lead/kg body weight (equivalent to 3.5 µg/kg body weight per day) for infants and children on the basis that lead is a cumulative poison and that there should be no accumulation of body burden of lead. In 1993, the Committee reconfirmed this PTWI and extended it to all age groups.

Assuming a 50% allocation of the PTWI to drinking-water for a 5-kg bottle-fed infant consuming 0.75 litres of drinking-water per day, the health-based guideline value is 0.01 mg/litre (rounded figure).

The most significant health effect from lead is the association of lead exposure with reduced cognitive development and intellectual performance in children. Results of studies on children with blood lead concentrations below 25 µg/dl indicate that, on average, the intelligence quotient (IQ) is reduced by 1-3 points for each 10 µg/dl increment in the blood lead concentration. Existing epidemiological studies do not provide evidence of a threshold.

Steps are now being taken to reduce all sources of lead exposure of children with some apparent success in various countries. In countries where lead has been removed from petrol, and where there is no specific source of excess lead exposure, blood lead concentrations in children are decreasing and are now approximately 4-6 µg/dl. The almost complete elimination of lead-soldered side-seams in canned foods in a number of countries has also contributed to the reduction in lead exposure. Corrosion control measures are being implemented to reduce the lead content of drinking-water and new plumbing and fittings now seldom contain lead.

Aesthetic Aspects

Contrary to the 1984 Guidelines, the 1993 Guidelines do not propose guideline values for substances and parameters that affect the acceptability of drinking-water to consumers. The Review Groups were of the opinion that guideline values should be recommended only for those substances that are directly relevant to health. A list of substances which may give rise to consumer complaints is given in Table A2.5 in Volume 1 of the Guidelines.

In the case of characteristics based on human sensory evaluation, judgement is often subjective. Aesthetic/organoleptic characteristics are very much subject to social, economic and cultural considerations, and the establishment of standards for the aesthetic quality of drinking-water should take into consideration implementation possibilities, and the existing socio-economic and environmental constraints. When resources are severely limited, establishment of priorities becomes even more important, and such priorities should be set in relation to their direct impact on health. Some countries have elected to set enforceable standards for constituents of health significance, whereas recommendations only are made for aesthetic and organoleptic characteristics.

Total dissolved solids

Total dissolved solids (TDS) in drinking-water consist mainly of chloride, sulphate, carbonates, sodium, magnesium and calcium. Excessive dissolved solids in drinking-water may lead to objectionable taste, and corrosion or encrustation in water distribution system. At concentrations greater than approximately 1000 mg/litre, the taste of water becomes increasingly unpalatable.

As far as health aspects are concerned, there is no evidence of adverse physiological reactions at TDS levels greater than 1000 mg/litre. On the contrary, there are vague indications from epidemiological studies that high levels of certain salts (calcium and magnesium) may have beneficial health effects.

It should be emphasized that the factor of acclimatization to TDS is particularly important. Many people enjoy highly mineralized waters containing more than 2000 mg/l of TDS.

Removal of TDS from drinking-water is an expensive proposition, and if a national standard for TDS is being considered, it should take into account the feasibility of implementation.

Turbidity

Particles in drinking-water are aesthetically objectionable, and can serve as shields for pathogenic microorganisms. Moreover, many toxic chemicals such as pesticides and heavy metals are selectively adsorbed on suspended particulate matter. The efficiency of disinfection may be reduced in the presence of turbidity: the disinfectant is unable to reach the target organism because of a physical barrier and/or chemical reactions with turbidity particles may occur thus decreasing the available disinfectant concentration. Where disinfection is practised, the turbidity should preferably be less than 1 Nephelometric Turbidity Unit and always below 5.

The effect of turbidity depends on its physico-chemical characteristics. Certain water supplies, such as groundwater, may contain non-organic turbidity, which may not affect disinfection. The complex factors involved in the potential health risk from the presence of turbidity precluded the derivation of a health-based GV.

References

WHO (1994) Fluorides and oral health. WHO Technical Report Series 846. Geneva.

WHO (1993) Evaluation of certain food additives and contaminants (lead). WHO Technical Report Series 837. Geneva.

Presentation Plan

Introduction

"All substances are poisons; there is none which is not a poison. The right dose differentiates a poison and a remedy".

Paracelsus (1493-1541)

Dose-response of Chemicals

Source: Galal-Gorchev, 1995

Physical Characteristics of Water

· Temperature
· Taste and odour
· Colour
· Turbidity
· Suspended solids
· Conductivity
· Total dissolved solids

Chemical Characteristics of Water

· pH
· Alkalinity
· Acidity
· Hardness
· Dissolved oxygen
· Oxygen demand
· Nitrogen
· Chloride

Sources of Fluoride and Impact on Health

Guideline Value for Nitrate and Nitrite in Drinking-water

where:

C = concentration
GV = Guideline Value

Lead and Health

Organic Chemicals

Session Objectives

· To demonstrate the range of organic chemicals that may be found in drinking-water and describe the sources of such chemicals.

· To describe the health risks associated with the consumption of drinking-water containing organic contaminants.

· To highlight the taste and odour problems associated with organic contamination.

Introduction

Specific chlorinated alkanes, chlorinated ethylenes, aromatic hydrocarbons, chlorinated benzenes, and miscellaneous organic chemicals were evaluated in the 1993 WHO Guidelines for Drinking-Water Quality.

Many of these organic chemicals are widely used as solvents, in chemical synthesis, in petroleum products, and in the production of plastics and resins. Some 30 organic chemicals were evaluated and guideline values (GV), or provisional guideline values, recommended for 27 of these.

Some organic substances do not have a 'no observed adverse effect limit' (NOAEL) - the highest concentration of a substance which causes no detectable adverse health effect. Therefore for some substances, a provisional guideline has been set on the basis of the 'lowest adverse effect limit' (LOAEL) - the lowest concentration or dose of a substance where there is a detectable adverse effect. Where a LOAEL is used, an additional uncertainty factor (UF) - a measure of the uncertainty regarding the information about a substance - is used. For further information, please refer to the background paper for Session II of this teaching pack or to the Guidelines Volume 1 and 2.

Adequate toxicological data were not available to derive guideline values for a number of chemicals, including 1,1-dichloroethane, 1,3-dichlorobenzene, dialkyltins, and a number of polynuclear aromatic hydrocarbons, with the exception of benzo[a]pyrene.

For most of the aromatic hydrocarbons and chlorinated benzenes, the taste and/or odour thresholds of these chemicals are well below the health-based guideline values, thus constituting an assurance that the GVs would not be exceeded or even approached. Consumers complaints would constitute a safety net against such a situation.

A number of organic chemicals were considered to be genotoxic and carcinogenic and the linearized multistage extrapolation model was used to derive guideline values corresponding to an upper bound estimate of an excess cancer risk of 10^-5.

For most organic chemicals that show a threshold for toxic effects, little was known of the magnitude of exposure from drinking-water relative to other sources. Consequently, the default value of 10% of the TDI was allocated to drinking-water.

Nitrilotriacetic acid (NTA) is a compound used in detergents, and tributyltin oxide (TBTA) is used in boat paints. Substantial exposure may occur from drinking-water, and consequently 50% and 20% of the TDI was allocated to drinking-water for NTA and TBTA respectively.

Di(2-ethylhexyl)adipate and di(2-ethylhexyl)phtalate are used in food-contact materials and therefore exposure from food is expected to be high. Therefore, only 1% of their respective TDI was allocated to drinking-water.

Because of their general persistence in the environment and concern over their potential toxicity, a large number of the organic chemicals evaluated are chlorine-containing chemicals.

For further information please see the Guidelines for Drinking Water Quality Volumes 1 and 2.

Presentation Plan

Example of Organic Pollutants by Class and Typical Use 1

· Halogenated aliphatic (chain) hydrocarbons

» Trihalomethanes (THMs) may be formed during disinfection of drinking-water; other compounds used as solvents for decaffeinating coffee, general solvents and in products such as propellant, degreasers, spot removers and dyes

· Aromatic (ring) hydrocarbons

» Many products derived from fossil fuels, also as additives in petrol, moth balls, adhesives and cigarette smoke

· Chloro-and nitro-aromatic hydrocarbons

» Fungicides and explosives

· Phthalates

» Phthalates are added to plastic to make them flexible; found in rain wear, footwear, shower curtains, childrens toys

Example of Organic Pollutants by Class and Typical Use 2

· Halogenated ethers

» Halogenated ethers are used in production of plastics and resins and in research laboratories

· Phenols

» Fungicide; wood preservative; Chloro-dichloro-, trichloro-phenols are by-products in the production of pentachlorophenol

· Organochlorines

» DDT, lindane, aldrin and chlorodane are examples of the extremely persistent organochlorine pesticides widely used in the 1950s and 1960s.

The Breakdown of Organic Compounds

· Enter water from a variety of sources including:

» Human/animal wastes
» Plants
» Soil erosion
» Industrial wastes

· Generally unstable, may be oxidised to stable and relatively inert end products e.g. CO₂, NO₃, H₂O

· Oxidation: loss of electrons to an oxidising agent such as oxygen or chlorine which accepts electrons

· Oxidation of organic compounds occurs in aerobic or anaerobic conditions

· Quantity of organic material and the quantity of oxygen needed to oxidise it estimated by using:» Biochemical oxygen demand (BOD)

» Permanganate value (PV)
» Chemical oxygen demand (COD)

Organics reviewed in the Guidelines

· A total of 30 organic substances reviewed in the following groups:

» Chlorinated alkanes (5 substances)
» Chlorinated ethenes (5 substances)
» Aromatic hydrocarbons (6 substances)
» Chlorinated benzenes (5 substances)
» Miscellaneous organic constituents (9 substances)

Pesticides in Drinking-Water

Session Objectives

· To demonstrate the conflict in pesticide use between agricultural and public health needs.

· To describe the two principal methods of pesticide classification.

· To describe the GVs set for pesticides and their by-products.

Pesticides are used for agricultural as well as public health purposes. Often a choice has to be made between their detrimental effects on the environment and their use for disease vector control, as for example, for malaria or schistosomiasis control. The adverse environmental effects of pesticides used in public health can often be mitigated through proper selection and application procedures. Equally, many pesticides have both beneficial and harmful health effects - their use may reduce the presence of particular vectors, although they may be toxic is consumed through water. In these circumstances, the relative benefits and dis-benefits should be evaluated.

With all pesticides, whether they have harmful health effects or not, the application should be well focused both in terms of application technique, quantity used and timing of application. As a general rule, the minimum of pesticide should be applied by the most efficient method at the most suitable time to achive the required goal. Over-application and/or application at times when the action is less likely to be effective should be avoided.

Pesticides can be classified according to chemical class (e.g. organochlorine, carbamate, organophosphorus, chlorophenoxy compounds) or according to their intended use (e.g. fungicide, herbicide, fumigant). It is important to know both since the chemical structure of the pesticide and its use often determine its behaviour in the environment, occurrence in drinking-water and toxicity to humans. Table 1 indicates the chemical class and use of the pesticides evaluated in the Guidelines.

Of the 36 pesticides evaluated, 28 contain chlorine. Organophosphorus pesticides were not evaluated although their use has increased as replacement for organochlorine pesticides. However, the organophosphorus pesticides are readily hydrolysed in water, adsorbed on sediments, or readily degraded in soil. As a result, they are seldom if ever found in drinking-water.

Many of the pesticides evaluated are herbicides. Because of their frequent use near waterbodies they have often been found in surface water. Furthermore many of these herbicides are fairly mobile in soil and readily migrate into groundwater.

While the use of organochlorine pesticides has declined in industrialized countries, their use continues in developing countries for public health as well as for agricultural purposes. For this reason, several organochlorine pesticides were evaluated in the Guidelines.

The toxicological basis of the guideline values and exposure assumptions made, as reflected in the percentage allocation of the TDI to drinking-water, are summarized in Tables 2 and 3.

For organochlorine pesticides such as aldrin/dieldrin, chlordane, DDT, heptachlor, and hexachlorobenzene only 1% of the TDI was allocated to drinking-water since it is known that these pesticides are highly persistent, have a high bioaccumulation potential, and are often found in food (Table 2).

In the majority of cases limited information was available on the contribution of drinking-water to the total exposure. Therefore a default value of 10% of the TDI was used (Table 3).

While considerable information is available on the toxicity of metabolites of pesticides formed in mammalian systems, the nature and toxicity of the environmental degradation products of pesticides are largely unknown and have not been taken into consideration in the Guidelines.

Alachlor, 1,2-dibromo-3-chloropropane, 1,3-dichloropropene and hexachlorobenzene were considered to be carcinogenic. The linearized multistage extrapolation model was therefore used to derive guideline values corresponding to an upper-bound estimate of an excess lifetime cancer risk of 1 per 100,000 of the population exposed.

Because limited information was available on the toxicity of 1,3-dichloropropane, ethylene dibromide and MCPB, no guideline values were derived for these pesticides.

Not all pesticides that have been found in water have been evaluated in the Guidelines. However, over 240 pesticides have been evaluated by the Joint FAO/WHO Meeting on Pesticide Residues (JMPR). Such evaluations could be used by countries wishing to establish standards or guidelines for pesticides of national concern.

In many circumatances, it may not be the principla component of the pesticide which is of concern, but impurities and by-products. It may be more effective to control the release of toxic substances into the aqautic environment through proper product quality control that by establishing standards for drinking-water. It may be more appropriate therefore, to ensure that product quality standards and their enforcement are in place than drinking-water quality standards.

References

International Programme on Chemical Safety (IPCS). Summary of Toxicological Evaluations Performed by the Joint FAO/WHO Meeting on Pesticide Residues, 1996.

Table 1: Chemical family and use of pesticides evaluated in the Guidelines

Key for Table 1:

Codes for chemical use

Codes for use

Table 2: Risk assessment of pesticides where substantial exposure from food is expected

Key:

Table 3: Risk assessment of pesticides where knowledge of exposure from different media is limited

Key:

Presentation Plan

Pesticides in the Guidelines

· Of the 35 pesticides evaluated, 28 contain chlorine

· Many of the pesticides evaluated are herbicides and readily migrate into groundwater

· Organochlorine pesticides have been included since they still have public health uses in developing countries

· Setting GVs is difficult due to uncertainty of health impacts

· 10% of the TDI allocated to drinking-water

· Nature and toxicity of the environmental degration products of pesticides are largely unknown

Monitoring and Assessment of Microbiological Quality

Session Objectives

· To describe the process of planning monitoring and surveillance activities and the need for progression through a number of stages starting with a pilot phase.

· To describe the development of analytical ranges in water quality monitoring.

· To introduce the critical parameters concept and emphasise the need for monitoring to focus on health related parameters of water quality.

· To describe the design of sampling networks and frequencies of sampling in routine monitoring programmes.

· To discuss the linkage of monitoring to water supply improvement.

Introduction

The routine monitoring and assessment of the microbiological quality of water is the key priority for both water suppliers and surveillance agencies. Microbiological quality is of principal concern because of the acute risk to health posed by viruses, bacteria and helminths in drinking-water. Therefore, monitoring and assessment of drinking-water is primarily a health-based activity which emphasises the protection of public health through ensuring that the water supplied is of a good quality.

Because monitoring is a health-based activity, other parameters of water supply should also be assessed: quantity, continuity, coverage and cost. All these parameters will affect public health. The water supplier should aim to monitor all aspects of water supply within their area of responsibility and aim for a continuous water supply which is of sufficient quantity and quality at an affordable cost to be available to all the population connected to the supply. The surveillance body should monitor the entire population and identify unserved groups and actively promote universal access to adequate water supplies.

Strategies for monitoring of microbiological quality of safe water should also include hazard identification and risk assessment, processes commonly incorporated within sanitary inspection. It is important that these are systematic and quantifiable and can be used to facilitate decision making at local, regional and national levels on preventative and remedial actions. In addition, minimum treatment requirements and source protection should be emphasised as essential complementary activities to monitoring and assessment.

Planning monitoring and assessment

Monitoring water supply quality will only be effective and efficient if it is properly planned and implemented. In many countries where there has not been routine surveillance and surveillance programmes being developed there may be uncertainty as to what standards should be adopted, the number of water supplies that should be covered, how many samples should be taken, what should be analysed, frequency of inspection etc. These may vary with time and it is important that the surveillance programme remains flexible and open to modification in response to evolving water quality priorities.

In many cases, WHO guideline values are adopted initially as the national standards for drinking-water quality. However, with time these may be superseded by national or regional standards depending on water quality priorities.

When designing a surveillance programme and planning its implementation, it is important that achievable aims and objectives are set. It is useful to clarify these terms: aims are general expression of targets, the 'end' result which is desired, are not generally time constrained. Objectives are indicators of the rate of success in achieving the aim and are specific goals set with definite time scales with indicators of achievement to provide a means of measuring success.

Extensive use of indicator bacteria will be required to monitor microbiological quality and the relative merits of these are discussed in the background paper on the microbiological aspects of water quality and in the Guidelines Volume 1. Thermotolerant (faecal) coliforms are the indicator most widely used for routine monitoring of water quality, although extensive use is also made of total coliforms. In addition to indicator monitoring, routine monitoring of turbidity and chlorine residual (in chlorinated supplies) is recommended in order to ensure that any deterioration of water quality post treatment is rapidly identified.

The rest of this paper will discuss the development of strategies for water quality monitoring and assessment with specific reference to microbiological risks.

Aims and objectives

The aim of surveillance has been defined in Guidelines for drinking-water quality. Volume III (2^nd edition) as follows: "Surveillance is an investigative activity which is undertaken in order to identify and evaluate factors associated with drinking-water quality which could pose a risk to health. Surveillance contributes to the protection of public health by promoting the improvement of water supply with respect to quality, quantity, coverage, cost and continuity". Guidelines Volume III also defines the aim of quality control in the water supply sector (which may be seen as an integral part of surveillance) as being: "to ensure that water services meet national standards and institutional targets."

In order to achieve these aims, a number of objectives may be identified, for example:

· the formulation of working methodologies for information gathering, decision making and communication;

· the review of existing quality standards and modifications of these as appropriate;

· the identification of appropriate analytical techniques;

· the identification of appropriate equipment and facilities (including evaluation of the use of on-site equipment) required to conduct a surveillance programme;

· identification of analytical quality control procedures for laboratories and on-site techniques and the identification of national (and possibly regional) analytical reference centres;

· establishing staff requirements and assessing skills of current employees, identification of training needs for staff, recruitment needs for the sector;

· to establish a protocol for approval of water sources as fit for drinking;

· to establish whether there are any particular problems in terms of sources, treatment technology, designs, operation and maintenance regimes etc. which are leading to persistent contamination problems.

The time in which it is expected to achieve these objectives must be outlined and indicators selected to measure progress, for instance definition of appropriate chemical assay equipment or type of on-site equipment required.

Routine monitoring of water supplies

In many circumstances, there is a desire to attempt to apply all aspects of surveillance to all the water supplies immediately. Whilst this is a laudable ideal, it is rarely possible to achieve successfully and may lead to resources being over-stretched and the failure of surveillance to provide the expected improvements in water supply quality. This may in turn lead to disappointment and increasing apathy towards surveillance.

Surveillance should be introduced progressively and at each stage objectives set which are achievable and which positively promote the continued development of surveillance. The experience from earlier stages should be used to improve surveillance and lead to a progressively more efficient and comprehensive surveillance programme. Thus, in the initial stages of surveillance, activities may be restricted to sanitary inspections and critical parameter analysis on a restricted number of water supplies. As the programme develops, the number of water supplies covered will be increased, frequency of sampling and inspections increased and the analytical range increased.

Sanitary inspections can be carried out relatively cheaply and easily and can be implemented on all water supplies from the start of surveillance. Sanitary inspection is as much a tool for the supplier or community as the surveillance agency for determining state of the water supply infrastructure and the identification of actual or potential faults and should be carried out on a regular basis by the supplier. The surveillance agency should conduct some independent inspections to verify the reliability of the supplier's information but this does not need to be as frequent.

In the long-term it is desirable that all water supplies should be included in a surveillance programme. However, it is important to be realistic in planning initial surveillance activities taking into account infrastructure, available trained personnel, the number of water supplies in the country and the ability to fund on-going surveillance activities.

Initial surveillance activities may be limited. However, it is important that short, medium and long tern achievable aims and objectives are included in the planning stage of surveillance. There should also be a set of clearly defined indicators that can be used to assess whether targets have been met. The establishment of credible indicators of a monitoring and evaluation programme at the start of the planning process is central to good planning of implementation. Time scales should be attached to each objective and aim and a proposed strategy for achieving these outlined.

Planned surveillance activities will only be possible if there are funds to pay for them and budgets for surveillance require careful preparation. In most cases, there are limited funds available and this inevitably will affect how many supplies can be included, how often they can be visited and how many samples can be analysed. It is therefore imperative that the following are identified and the cost calculated: the number and location of water supplies to be included in each stage of surveillance; staff time; consumable requirements; equipment purchase and maintenance costs; fuel costs; the cost of reporting results to suppliers and communities; and the cost of follow-up activities. It is vital that all these elements are accurately budgeted for and cost-effectiveness achieved.

Pre-surveillance activities

Prior to the start of a surveillance programme, there are a number of activities which should be undertaken to ensure that the planners have access to all baseline data to design the surveillance programme. The pre-surveillance activities should provide the information concerning current status of water supply and surveillance in the country, and will include the following:

Current surveillance activities: scope; analytes; reliability; who is responsible; geographical spread.

Inventory of supplies: type (borehole/spring/gallery/surface etc); treatment technologies; age; population served; existing quality data; source approval;

Staff assessment: numbers available; skills available; recruitment requirements; training requirements;

Surveillance infrastructure: available laboratories; laboratory equipment; on-site equipment; consumables; transport; geographical spread; computer availability; database availability.

Once this information is available the programme can be designed and will include recommendations for improving on all the above and to test appropriate methods. It is usual to run a pilot project to evaluate the approach to be adopted and to identify any parts of the programme which require improvement.

Pilot project

There are essentially two approaches to the establishment of a pilot phase that can be adopted:

1. the use of a pilot project concentrated within one geographical area;

2. establishing surveillance on small scale national basis, with a small number of supplies included from each region.

Selected supplies may be restricted to those with large populations (for instance over 10,000 people or provincial capitals) or may include supplies that serve all types of population centre.

The first of these approaches allows a more intensive allocation of resources and it may be easier to measure the effectiveness of the approach adopted. However, there is a risk that the area selected may not be representative or that successful approaches may not be replicable in other parts of the country. This may be due to different types of source or treatment used, different staffing structure or resource base. It may be difficult to develop and sustain the level of support available for a small regional, pilot-scale project on a national scale in the short term. Different regional water suppliers may have different priorities based on the principal threats to water quality in each region, the surveillance infrastructure available, the expertise available and the number of people served.

A national pilot project may be more expensive due to increased travel costs and may be difficult to manage. However, there are many advantages in this approach. It is much easier to establish a large-scale national surveillance programme if in the pilot phase a national approach was adopted. In this way, difference in priorities between regions will have been identified at an early stage and can be incorporated within the national plans. It will highlight any logistical, staffing or infrastructural problems that exist in a region and these can be planned against in the full programme. This approach is also likely to permit all types of water source and supply found in the country to be represented in the pilot phase, something that may be difficult to achieve in a single region.

Once the pilot project is complete and modifications made as required, the surveillance programme proper can start. The implementation of surveillance is likely to be staged over a number of years and short, medium and long term plans will have to be drafted.

Short-term plans

The short-term aim for a water surveillance programme should be to establish it as a perceived key priority of water supply and water resource management and to create an environment which actively promotes surveillance.

The short-term objectives of national surveillance programmes should be to achieve routine analysis of critical parameters covering a representative sample of all water supplies. There are no hard and fast rules determining this, but a figure of around 30 per cent of all water supplies has been adopted in some circumstances. If only a proportion of supplies is to be included initially, the supplies should be spread geographically, encompass examples of all (or at least the principal) source types and treatment technologies and should be concentrated on communities with larger populations. Given that protected groundwater sources tends to have less bacteriological contamination, it is common to include a greater proportion of surface supplies in the early stages of surveillance. Supplies where there are known problems with water quality should be included at this stage to try to establish the causes, rectify these and prevent their recurrence.

Where chlorinated water supplies are surveyed, from whatever source, turbidity and chlorine residual within the network should be tested regularly. As the equipment and consumables required are very cheap and testing is field based, it is feasible to test frequently and this may help reduce the number of microbiological samples required.

The number of microbiological samples taken and their frequency will vary depending on resources and population size, but as far as possible samples should be taken at least quarterly by the supplier and at least annually by the surveillance agency. The number of samples taken is largely dependent on population supplied, time available, analytical resources and type of distribution network. However, the more samples that are taken, the more representative the results. Below is a very crude guide for the minimum number of samples that should be taken:

If water supplies are from a point source, for instance a borehole or well, not connected to a pipe network, analysis need not be as regular. However, there should be a minimum of two analyses per year - one wet season and one dry season - to take into account water level fluctuations and to assess whether quality varies seasonally.

Sanitary inspections should be carried out regularly by the supplier or the community on all water supplies and not merely those where analysis is being carried out. Sanitary inspections may be undertaken by staff such as systems operators or by trained community members.

Where there is a supply agency responsible for the provision of drinking-water, the results of the sanitary inspection and any recommendations for action should be noted and shared with the supply agency. An annual summary of inspection results should be passed to the surveillance agency which highlights any actions which have been recommended and the outcome of these.

Where the water supply is community managed, the results of the sanitary inspections should be used by them to plan improvements to their supply and an annual summary of the results should be passed to the surveillance agency. An annual independent sanitary inspection by the surveillance agency should also be carried out.

Training programmes should be initiated to ensure that staff are able to carry out surveillance activities and can pass on skills in sanitary inspection to communities. Staff from both the suppliers and the surveillance agency are likely to require training, as are any community members involved in surveillance.

Medium-term plans

The medium-term aim of surveillance should be to review and consolidate the programme and expand it to cover a greater proportion of water supplies and to ensure that adequate standards are established.

One of the medium-term objectives of water surveillance should be to increase the coverage of the surveillance programme and to make modifications to the programme as appropriate. The proportion of supplies that have regular analysis of critical parameters should be increased, for example to 80 per cent. The additional water supplies included in the surveillance programme should be distributed to reflect population distribution and the number of groundwater supplies covered increased.

Chlorine residual and turbidity testing in the distribution network should continue to include all chlorinated water supplies and the frequency of testing increased. The number of samples taken from the distribution network for microbiological analysis should be increased and the supplier should aim to carry out microbiological analysis of samples at least quarterly and preferably monthly on large supplies. An independent analysis by the surveillance agency should be carried out at least annually or even more frequently on large supplies. The analytical range may also be extended to include other parameters such as total coliforms or other faecal indicator bacteria.

Lines of communication should be established between supplier, consumer and surveillance agency and the population should be kept aware of water quality problems that arise and precautionary actions that they should adopt.

Quality standards may be revised or if in the initial phase employed the use of guideline figures for water quality, the second phase may well include adoption of legally binding water quality standards. Standards or guidelines for other substances of health importance, for instance nitrate, should be drafted and analysed for as frequently as feasible.

Sanitary inspections should continue to be carried out monthly by the supplier at all supplies and independent inspections carried out at least annually. Larger supplies and a significant number of other supplies should be inspected quarterly by the surveillance agency.

In addition, codes of practice and construction standards for plumbers and builders should also be established and supported within the legal framework. A system for licensing approved craftsmen should be established with them regularly assessed and if consistently blow acceptable standards, a mechanism established to revoke their licenses.

An on-going training programme for staff involved in surveillance should be established and cover analytical techniques, sanitary inspection techniques and community education. This should include ensuring that appropriate courses are offered and made accessible for staff with extensive experience but limited formal qualifications. In-service training in appropriate topics should be provided and taught through short-courses and 'on-the-job' training.

The national laboratory network should be increased and appropriate AQC procedures established to ensure analytical quality is maintained.

Long-term plans

The long-term expansion of the surveillance programme should be to include all water supplies in the country in the surveillance programme and to ensure that both the supplier and the surveillance agency undertake regular sampling and sanitary inspection. The long-term surveillance plan should include the assessment and revision if necessary of drinking water quality standards and these should be expanded to cover all substances of health and environmental importance.

Samples should be taken monthly for microbiological analysis and in larger supplies this should be expanded to weekly or daily. the surveillance agency should aim to undertake regular independent analysis. The number of samples taken from the distribution network for microbiological analysis should be increased and should be representative of the entire network.

Sanitary inspections should be carried out by the suppliers on a monthly basis on larger supplies and at least quarterly on smaller supplies with independent inspections carried out quarterly or bi-annually. In chlorinated supplies, the supplier should sample for turbidity and chlorine residual at least weekly and in large supplies daily.

As in each phase of the surveillance programme, there must be clear lines of communication between supplier, consumer and surveillance agency. Legislation should be drafted which gives a framework of the steps to be taken by the supplier to inform the consumers and the surveillance agency when water quality is sub-standard. This should include time limits within which contamination must be reported and advice given about precautionary actions that should be taken by the consumer (for instance boiling). There should also be time limits imposed within which the surveillance agency should be informed of any failure to meet standards and proposed action.

Full analytical quality control and assurance procedures should be established and all laboratories where analyses are undertaken should be part of analytical quality control and analytical quality assurance programmes. A national reference centre should be established, possibly supported by a network of regional centres. There should also be clear guidelines for ensuring the analytical quality and reliability of results obtained from on-site equipment.

A human resource development strategy should be drafted which identifies sector training needs and how best training should be provided. This should include in-service training and establishment if appropriate of further and higher education courses which will produce appropriately qualified staff.

Conclusion

Monitoring and assessment of microbiological water quality is a key priority in the water sector which involves water suppliers, surveillance agencies and communities. It is a health-based activity and should include elements of hazard identification and risk assessment, through the use of systematic sanitary inspection, as a means of improving water supply quality.

Analysis of indicator bacteria should be supported by turbidity and chlorine residual testing and these elements, combined with sanitary inspection, should be used to define the sanitary status of the water supply.

Monitoring is best implemented through a series of stages to ensure that any problems in implementation are identified and rectified during the early stages of programme development. Initial pilot projects should test the methodology to be used and this should then be progressively implemented on a nation-wide basis.

References

Anon. Guidelines for Drinking Water Quality. Volume 3 (2^nd Edition). WHO, Geneva 1997

Bartram, J. (1996) Optimising the Monitoring and Assessment of Rural Water Supplies. PhD Thesis, University of Surrey.

Bartram, J. and Balance, R. (1996) Water Quality Monitoring. Chapman and Hall, London

Howard, G. Developing Drinking Water Quality Monitoring in Zimbabwe, ODA Project Reports 2-7, National Water Quality Analysis Laboratory

Howard, G. (1997) Water Quality Monitoring: Key Issues and Approaches. Waterlines (in press)

Lloyd, B. and Helmer, R. (1991) Surveillance of Drinking Water Quality in Rural Areas. Longmans' London

Lloyd, B., Bartram, J., Rojas, R., Pardon, M., Wheeler, D. and Wedgwood, K. (1991) Surveillance and Improvement of Peruvian Drinking Water Supplies. ODA. Guildford.

Presentation Plan

Why Monitor?

· Protection of human health
· Compliance with standards and guidelines
· Situation analysis/impact assessment
· Environmental change and trends
· Rapid detection of faults and failure
· Prioritisation of remedial actions
· Adequate quality of service

Five Key Elements for a Water Supply

· Quantity: Enough water for everyone to drink, cook and bath, e.g. 30-100 litres/person/day

· Quality: The water will not cause disease in those drinking or using it

· Cost: The cost of sufficient water for basic needs is within everyone's reach

· Coverage: Water is available to everyone in the community

· Continuity: Water is available all day, every day

All five elements are vital if health is to be improved and maintained

Ensuring Microbiological Quality

· Source protection
· Minimum treatment requirements
· Sanitary inspection
· Water quality analysis

Objectives of Water Quality Monitoring

· Evaluate risks to the population
· Improve the situation
· Determine long-term trends
· Prioritise interventions

Developing Water Quality Monitoring

Water Quality Surveillance

'...the keeping of a careful watch at all times, from the public health point of view, over the safety and acceptability of drinking-water supplies'

(WHO, 1985)

· Source
· Treatment
· Distribution

Coliform Analysis + Sanitary Inspection = Surveillance

Selection of Parameters

· First Stage

» Critical parameters (WHO)
» Organoleptic parameters (taste, odours, colour)
» Known problem of public health concern

· Expansion of Analytical Range

» Should be the objective only once critical parameters are being used and improvements are being made

· Selection of New Parameters

» Those parameters which affect health
» Those parameters which lead to rejection of supply by consumers

The expansion of the range of parameters must be progressive and allow priorities to be identified

Critical Parameters

Samples must be analysed within 6 hours of taking the sample form a water supply. In areas where transport or roads are poor and this is not possible, portable water testing kits can be used

Design of Surveillance Networks

· Site selection representative of:

» Water source
» Treatment plant
» Storage tank
» Household connection
» Point of use

· Look at where problems are likely to occur:

» Main lines
» Remote branches
» Dead ends

· Use 'zoning' for large systems with several sources.

Sampling Network Design (Small supply)

Sample sites in a small water supply with ring main

Sampling Network Design (Large supply)

Sample sites on a large supply with an open network

Classification of Sample Sites

· Fixed - Agreed in consultation with supply agency

» To help surveillance agency and water supply agency to compare results
» Allows legal action to be used to ensure improvement

· Fixed - No consultation with water supply agency

» Used with other fixed sites to determine changes in water quality with time

· Variable - Samples taken at random by surveillance agency

» Good for identification of local problems e.g. complaints, leaks

» Could include 'points of use' sampling

Minimum Sampling Frequency for Water Supply Systems

Using monitoring to improve water supplies