Environmental Health Risk Assessment Guidelines for assessing human health risks from environmental hazards
Appendix 4 Environmental Health Risk Assessment for Water
- Objectives
- Audience
- Acknowledgments
- List of participants
- Summary
- Abbreviations
- Glossary
1. Background to Risk Assessment- 2. An Australian Framework for Risk Assessment
- 3. Issue Identification
- 4. Hazard Assessment— Part 1: Hazard Identification—Toxicology
- 5. Hazard Assessment—Part 2:Hazard Identification—Epidemiology
- 6. Hazard Assessment—Part 3:Dose–Response Assessment
- 7. Hazard Assessment—Part 4:Hazard Assessment Reports
- 8. Exposure Assessment
- 9. Risk Characterisation
- 10. Appraisal of Assessments
- 11. Setting Environmental Health Criteria
Appendices- Appendix 1 Environmental Health Risk Assessment for Contaminated Sites
- Appendix 2 Environmental Health Risk Assessment for Air Pollutants
- Appendix 3 Environmental Health Risk Assessment for Food
- Appendix 4 Environmental Health Risk Assessment for Water
- Appendix 5 Australian Models of Risk Assessment
- Appendix 6 International Models of Risk Assessment
- Appendix 7 WHO/IPCS Conceptual Framework for Cancer Risk Assessment
- Appendix 8 Microbiological Risk Assessment
- References
- enHealth Council Membership and Terms of Reference
4.1 Introduction
There is a wide range of water types, water uses and possible routes of transmission of waterborne hazards to humans. In undertaking health risk assessments the characteristics and potential uses of water bodies need to be determined. Water sources include fresh, estuarine, marine and waste waters. Water uses can include supply of potable water for drinking and bathing, recreation, aquaculture or irrigation of crops. Human exposure to waterborne contaminants can include:- direct exposure through ingestion, dermal contact, inhalation of aerosols or sprays; or
- indirect exposure through foods contaminated by irrigation water or water used for aquaculture and seafoods contaminated by waste water discharges. Health risk associated with food contamination via a waterborne route is within the scope of Appendix 3 addressing the risk assessment of food.
4.2 Identifying the Issues
Health risks associated with various types of water are considered in a range of guidelines detailed in the following text.Drinking water is generally the highest and most direct source of human exposure to waterborne contaminants and accordingly it usually receives the most attention in water-related health risk assessment. Drinking water quality issues are addressed in detail within the Australian Drinking Water Guidelines (NHMRC and ARMCANZ, 1996). These guidelines (ADWG) provide an excellent description of water quality management needs from source water to tap. They also provide detailed fact sheets describing the rationale and health risk evidence for setting of guideline numbers for all individual quality parameters currently covered.
The ADWG are undergoing rolling revisions and their current status and draft guidelines can be found at www.health.gov.au/nhmrc/advice/water.htm
Health risks associated with recreational water are considered in published guidelines for recreational water quality (US EPA, 1999) and in guidelines being prepared by WHO and NHMRC. Health risks associated with use or reclaimed water are considered in draft guidelines for use of reclaimed water from sewerage systems (NHMRC, ARMCANZ and ANZECC) which should be published in early 2000. In addition annual reviews of health effects associated with waste water disposal and reuse are published in Water Environment Research (e.g. Froese and Bodo 1999)
4.2.1 Drinking water
Drinking water quality is usually categorised in terms of physical, chemical or biological parameters. For drinking water, the health-related physical parameters are primarily radiological (NHMRC and ARMCANZ, 1996). Turbidity, as a physical parameter in its own right is an aesthetic concern. However, in drinking water produced by filtration plants, turbidity is also used as an indicator of treatment efficiency for the removal of pathogenic microorganisms (e.g. see US EPA, 1998). The chemical parameters of concern in drinking water are categorised into inorganic and organic chemicals. The latter are sub-categorised into disinfection by-products, pesticides and other organic compounds (NHMRC and ARMCANZ, 1996). The biological parameters are focused on pathogenic microorganisms which are categorised into bacteria, protozoa, toxic algae (cyanobacteria) and viruses. The water quality parameters currently covered by the ADWG are listed in Table 1 A4.The overall relationship among common water quality parameters can be summarised in a variety of ways, with one approach presented in Table 2 A4 (Hrudey, 1999). This perspective shows how the physical characteristics of water contaminants (suspended vs. dissolved, volatile vs. non-volatile) relate to their chemical character as well as their classification as biological or chemical contaminants.
Table 1 A4: Parameters covered by the Australian Drinking Water Guidelines (NHMRC and ARMCANZ, 1996)
Micro-organisms |
Physical characteristics |
Inorganic chemicals |
Organic chemicals |
|---|---|---|---|
Bacteria |
Radionuclides |
• Aluminium |
Disinfection by-products Other organics Pesticides |
Table 2 A4: Classification of water quality parameters
(reprinted by permission of S Hrudey)
Comments:
- Exchange of contaminants between phases is governed by a dynamic equilibrium that is dependant on temperature and the relative concentration of contaminant in each phase
Operational definitions:
- dissolved—passes micro filtration but not reverse osmosis
- colloidal—not removed by sedimentation or direct granular filtration
- volatile—air strippable
- semi-volatile—steam strippable
- viable—can replicate under favourable conditions
- infectious—can infect a susceptible mammal representative of humans
Technology specification examples:
- chlorination/ozonation—converts organic matter into new organic compounds with some oxidised to inorganic products and disinfects by making viable organisms non-viable
- coagulation—converts colloidal and suspended matter so that sedimentation and granular filtration can remove suspended and colloidal matter
- granular filtration—removes some fraction of suspended and coagulated colloidal matter
(Hrudey, 1999)
The management of water quality requires attention ranging from sound management of land use and human activities in catchments through to the application of treatment technology to achieve safe drinking water. A variety of measures may be considered for the protection and enhancement of source water quality including (Reinert and Hroncich, 1990):
- assessment of safe yield in terms of water quantity;
- catchment land ownership;
- land use controls or management agreements;
- in situ treatment (mixing, aeration, algae control);
- wildlife control;
- forest or agricultural management practices;
- emergency response measures;
- routine sanitary surveys and catchment inspection;
- access control (e.g. fencing); and
- public education.
Water treatment has evolved from classical technology developed in the early 1900s that was originally based on coagulation, filtration and disinfection. These basic approaches have been refined and improved through the use of a number of technological alternatives so that a modern water treatment process scheme can be developed that will satisfy specified finished water quality requirements given identified source water quality challenges. The menu of drinking water treatment process types that are available for consideration now generally includes (AWWA, 1990; Dezuane, 1997):
- air stripping / aeration;
- coagulation processes (destabilisation, mixing and flocculation);
- sedimentation and flotation;
- filtration;
- ion exchange and inorganic adsorption;
- chemical precipitation;
- membrane processes;
- chemical oxidation;
- adsorption of organic compounds; and
- disinfection.
4.2.2 Recreational water
Recreational water quality is usually categorised in terms of microbiological parameters although physical features that could represent a hazard to bathers are also considered in some guidelines (NHMRC, 1990) and will be included and probably expanded in those being prepared by WHO and NHMRC.As for drinking water the management of water quality requires attention to sound management of potential sources of pathogens from adjoining drainage areas and catchments. Pathogens can be transported through point sources such as waste water outfalls or diffuse sources where contamination is related to rainfall events.
A practical guide to the design and implementation of assessments and monitoring programmes associated with recreational water is:
- Bartram J. and Rees G. (eds) (2000).Monitoring Bathing Waters. E&FN Spon, London.
4.2.3 Reclaimed water
Like drinking water, reclaimed water quality is usually categorised in terms of physical, chemical and biological parameters. The management of risk is usually based on a combination of treatment, controlled use and controlled exposure.Treatment processes have not changed a great deal and are based on a combination of primary, secondary and tertiary treatment and disinfection. Detention in lagoons is a low technology but robust method of treating sewage and can be used to provide primary and/or secondary treatment. Tertiary treatment generally involves filtration and is required for uses with potential for higher exposures such as residential non-potable use.
4.3 Hazard Identification
Health risk assessment for chemical parameters was first documented in some detail in a series of publications of the National Research Council of the U.S. National Academy of Sciences (NAS, 1977; 1986; 1987; 1989). These expert panel reports, published as separate volumes in the series Drinking Water and Health addressed the emerging key issues underlying risk assessment of chemical contaminants in drinking water and provided a framework which has guided the evolution of health risk assessment of drinking water contaminants.The most common and likely source of human health effects associated with water is through exposure to microbiological pathogens. The ability of a pathogen to cause illness is usually well established but in assessing water quality there is a tendency to ignore species variability. For example, while Cryptosporidium as a generic group has been identified as a cause of waterborne illness only one species C. parvum is regarded as causing human infections. In addition it is likely that only sub types of C. parvum are infectious for humans.
Water has been documented as a source of large disease outbreaks such as the 1993 Milwaukee outbreak of Cryptosporidiosis that was estimated to have infected over 400,000 people (Mackenzie et al, 1994). However, quantitative risk assessment for microorganisms has only recently developed in comparison with chemical risk assessment (Haas et al, 1999). Models for microbiological pathogens have been developed for a few organisms including Cryptosporidium, Giardia and some types of viruses but the models are limited. A range of factors that are generally not yet adequately considered include: human variability in the form of immune status and partial or total immunity through prior exposure, variations in virulence and variations in seriousness of illness outcomes. With a few exceptions (e.g. Legionella, Naegleria fowleri) water borne pathogens tend to be transient contaminants and not free-living organisms.
Limited quantitative microbiological risk assessment has also been undertaken for use of reclaimed water. As for drinking water, models have been developed for Cryptosporidium, Giardia and some types of viruses.
Drinking water health risk assessment provides a compelling example of risk tradeoffs because acute microbiological disease is almost certain to arise with surface water supplies that are not subjected to disinfection (Singer, 1999). However, since the discovery in 1974 that chloroform and other trihalomethanes are produced as by-products of chlorine disinfection, there has been a phenomenal growth in the identification of disinfection by-products (DBPs), as summarised in the organic DBP section of Table 1 A4. As a result there have been numerous epidemiological studies, some of which suggest possible links between DBPs and adverse health effects ranging from bladder cancer to adverse reproductive outcomes. Maintaining a sensible balance between the known infectious disease risks that can be controlled by disinfection and the hypothesised health effects associated with DBPs has presented the drinking water industry with a substantial challenge in the assessment and trade- off of competing risks (Craun, 1993). Issues associated with potential conflicts between compliance with requirements for microbiological and DBP control are discussed in a US EPA Guidance Manual (1999). (www.epa.gov/safewater/mdbp/mdbptg.html)
Drinking water sources or recreational water are most likely to be polluted with human, animal, agricultural and industrial wastes. The hazards associated with human and animal wastes are predominantly microbiological while those associated with industrial wastes are generally chemical. Agricultural wastes can be microbiological (animal) or chemical (fertilisers, pesticides). In addition pollution can exert secondary effects through the support of increased growth of naturally occurring organisms such as cyanobacteria which may pose health risks from both drinking water and recreational water use perspectives (Chorus and Bartram, 1999). This recent World Health Organization monograph, prepared with substantial input from Australian experts, provides an excellent illustration of a practical approach to hazard identification and preliminary risk assessment for cyanobacterial hazards to drinking water supplies. This is summarised as a generic approach in Figure 1 A4.
A similar approach is likely to be included in guidelines being prepared by NHMRC and WHO for recreational water.
Figure 1 A4: A generic rationale for hazard identification and preliminary risk assessment for drinking water health risks
(adapted from Bartram et al, 1999)
4.4 Multiple Barrier Approach to Reduce Contamination and Health Risks
The provision of barriers to the transmission of pathogens and contaminants is important in reducing health risks associated with water. The multiple barrier approach relies on the concept of using more than one type of protection or treatment in a series in a water treatment process to control contamination and provide overall process reliability, redundancy and performance.An example of the multiple barrier approach to protect drinking water occurs in normal catchment-to-tap management. The barriers include the following:
- Protection of source water from contamination with an active catchment protection program;
- Long detention times within reservoirs (weeks to months);
- Water treatment e.g. coagulation, settling and filtration;
- Finished water to be disinfected before it enters the distribution system;
- Maintenance of an adequate disinfection residual throughout the distribution system; and
- Maintenance of the integrity of the distribution system i.e. no breaks in the pipes, roofs on water tanks etc.
- Monitoring for microbiological quality should be regarded as a check that the barriers are maintained.
4.5 Monitoring Methodologies
The most widely accessible comprehensive reference for techniques of water analysis and sampling is the publication, ‘Standard Methods for the Examination of Water and Waste water’ (Clesceri et al, 1998). Explicit guidance on the frequency of monitoring for parameters that are covered by the ADWG has been provided in the ADWG document (NHMRC and ARMCANZ, 1996). Water quality monitoring has long been based on indicator or surrogate parameters to represent agents of health concern. For example, the presence of microbiological pathogens, which have been impractical to monitor directly, has been inferred by indicator bacteria such as the total or thermotolerant coliform bacteria. The presence of indicator organisms has been taken as a sign that water quality may have been compromised with the possible presence of enteric pathogens. Confidence in the reliability of these indicator organisms has been undermined by the finding of protozoan pathogens like Cryptosporidium and Giardia species that are substantially more resistant to chemical disinfection than the indicator organisms. This reality means that inactivating the indicator organisms by disinfection does not assure inactivation of the pathogens. These circumstances are further complicated by the lack of reliable methods to monitor for viable and infective strains of the resistant pathogens so that the relevance of non-specific monitoring data remains a challenge.In the case of chemical contaminants, some parameters like the trihalomethanes (THMs) may be only indicator or surrogate measures for other chemical agents that may or may not pose health risks. There are also issues about the chemical species that are measured in water quality monitoring. For example, arsenic toxicity varies over a thousand fold depending on the chemical form of the arsenic. The more common inorganic forms of arsenate or arsenite that are most likely to be present in drinking water are believed to pose the greatest health concerns.
Guidance on sampling water has also been provided by Keith (1988; 1991; 1992). Water bodies are often not homogeneous mixtures and a number of issues need to be addressed in designing sampling programs including differences between stream flows and embayments, water depth, stratification and the impacts of silts and sediments as sources of microbiological and chemical contaminants.
4.6 Assessment of Summary Statistics and Presentation of Data
The appropriate format for presenting water quality data depends on the nature of the hazard. Microbiological hazards generally pose an acute risk so that short term monitoring is needed and transient excursions above guideline levels can pose the danger of waterborne disease transmission. Because real time (instantaneous) monitoring of microbiological parameters is currently not possible, factors which can be monitored frequently or continuously, like disinfectant residual and turbidity, are often used to document treatment performance and thereby infer acceptable microbiological control. Chemical parameters that pose a chronic risk, such as suspected carcinogens, are usually judged in relation to standards based on lifetime exposure. For these parameters, long-term average exposure is generally considered for dose, although it is usually expressed on a daily basis (i.e. as an average daily dose for a lifetime). The more recent interest in the possibility of adverse reproductive outcomes associated with various water quality parameters has changed this perspective making short-term (possibly even peak) exposures more relevant.4.7 Censored Data and Levels of Reporting
As a general guide, reporting to a sensitivity of one tenth of the guideline level is preferred but may not be practicable for some substances, such as pesticides, where the guidelines have been set at a level of detection. Reporting levels need to be set sufficiently low so as to be able to distinguish parameter trends from background levels. Like most environmental data, water quality data are often highly skewed because sub-detection values cannot exist so that data sets are truncated at the detection limit. Often, log normal distributions may fit the data, unless a few extreme values skew the data more than a log normal distribution will readily accommodate.4.8 Dose–Response Assessment
There are difficulties associated with dose–response assessments for both chemical and microbiological contaminants. Microbiological dose–response assessments are faced with the difficulty of considering human and microbiological variation. The likelihood of contracting an infection is influenced by factors such as immune status, immunity imparted by previous exposure and virulence of the specific type or strain of microorganism.Measurement of chemical parameters in water is generally well developed. However, evaluating causal linkages and dose–response relationships between estimated doses and disease generally involves a great deal of uncertainty. Predictions are often based on extrapolation of high dose animal toxicology data to chronic low level exposures in humans. Thus, uncertainty arises from both interspecies extrapolation and high to low dose extrapolation. For chronic diseases like cancer, the causal linkages must be inferred from observational epidemiology studies for which drinking water contaminant exposures must usually be reconstructed from limited historical data. These realities raise considerable uncertainty about actual exposure levels, as well as the uncertainty arising from bias and confounding on the estimation of relative risk.
4.9 Exposure Assessment
There is a tendency to simplify exposure assessment by focusing on ingestion of standardised volumes of water. However, exposure to drinking water and recreational water contaminants can occur through ingestion, inhalation and dermal contact. For example, with volatile and lipophilic organic contaminants (e.g. THMs) doses from showering and bathing may be higher than via ingestion (Weisel and Jo, 1996). The scope and complexity of drinking water contaminant exposure assessment has recently been addressed in some detail (Olin, 1999) and is summarised in Figure 2 A4.Figure 2A4 removed due to copyright restrictions
In determining guideline values for drinking water a standard daily consumption is used which in Australia has been judged to be 2 litres per day for an adult. However, there is variation in consumption and individuals may derive drinking water from a number of sources including tap water at home and work, bottled water and rainwater. More detailed exposure evaluations must take into account the full range of contaminant exposure routes.
In recreational guidelines an estimated maximum ingestion of 100 mL per recreational session is used for both marine and fresh waters (NHMRC, 1990).
In regard to reclaimed water maximum exposures have been calculated for a number of uses including irrigation of edible crops (10mL), irrigation of public areas, golf courses etc (1mL) (Asano et al, 1992).
Determination of potential doses in terms of the concentration of contaminants in water is better developed for chemicals than for pathogens. For some chemicals (e.g. some pesticides), there may be health concerns at concentrations near routine detection limits. It is possible that these limits are higher than for other chemicals which can be detected in parts per billion or even lower concentrations. However, for many microorganisms there are no reliable or sensitive quantification methods and in some cases methods have not been developed to identify species or strains that can cause human infections.
4.10 Risk Characterisation
The characterisation of risk that is inherent in setting drinking water guideline levels can be generally summarised according to:| Guideline Level = | RL x BM x AF x ED IR x UF x AT |
| Where: | |
| Guideline Level = the guideline concentration of contaminant in water | |
| RL = the reference toxicity level (often a no effect level) BM = the body mass, often 70 kg for an adult AF = what proportion of total exposure can be attributed to drinking water ED =exposure duration (if exposure is less than continuous) IR = an ingestion rate, often taken as 2L per day UF = uncertainty factors applied to reduce the RL AT =an averaging time of exposure, will equal ED if exposure is continuous |
|
Health-related criteria have been established for a wide range of chemical contaminants in water and initial comparison of estimated exposure levels should be with the ADWG (NHMRC and ARMCANZ, 1996). An additional source of information is provided by the US EPA Drinking Water Health Advisories within the Integrated Risk Information System (IRIS) which can be found at: www.epa.gov/ngispgm3/iris/dwater.html These Health Advisories provide data for drinking water exposures for up to one day, 14 days, 7 years and lifetime exposures.