Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
l Approach to the Study INTRODUCTION In this chapter the general approach, principles, and criteria adopted in the study are discussed in outline. Considerations that entered into evaluations of the ejects on health of the various contaminants of drinking water are described, together with the reasons for selecting the subjects that were studied. The findings of the study are not summarized comprehensively in this section; each succeeding chapter includes a summary of the relevant conclusions and recommendations. A short summary of the principal conclusions of the study is given in Appendix C. The study was undertaken by the NAS-NRC to meet the needs expressed in the Safe Drinking Water Act (PL 93-523), which requires the Environmental Protection Agency to promulgate national drinking water standards and, for the first time, regulations for enforcing them. The Act also directs the Administrator of the Environmental Protection Agency to arrange with the National Academy of Sciences, or other appropriate organization, to study the adverse ejects on health attributable to contaminants in drinking water. Although the high quality of drinking water in the United States is recognized throughout the world, the law is an expression by the Congress of the concern of many citizens about maintaining the quality of public water supplies in this country. The reader should not equate the size of this report or that of any of its chapters with the Committee's assessment of the magnitude of the challenge to public health that may be due to the presence of particular 9
10 DRINKING WATER AND HEALTH constituents in drinking water in the United States. Several factors have contributed to the length of this report: The Safe Drinking Water Act defined the scope of the study in encyclopedic terms and consequently the length of some of the chapters reflects the large number of topics and substances that it was necessary to consider. Other chapters deal with subjects that are complex and about which there are uncertainties, conflicting opinions, and inconclusive or incomplete data. The relevant studies, assumptions, methodologies, health elects, and research recom- mendations for each group of constituents required detailed consider- ation from several points of view before balanced judgments could be achieved. In some cases brevity had to be sacrificed to reach this objective within a reasonable time. The primary purpose of the study was to assess the significance of the adverse elects that the constituents of drinking water may have on public health. The economic or technological feasibility of controlling the concentration of these constituents was outside the scope of the study. The health elects associated with some methods of disinfection were noted, but the relative effectiveness and potential hazards associated with the various methods of water disinfection were not evaluated. Application of analytical methods of great sensitivity has, in recent years, expanded our knowledge of the occurrence and diversity of impurities in drinking water. However, information about the biological results of chronic ingestion, at low dose rates, of most of these substances is acquired slowly because the bioassays that are usually required may take two or more years to complete. Although new approaches to the problem of estimating chronic adverse health elects may, in the future, ease this difficulty, the current knowledge on which this study is based is insufficient to assess all the contaminants of drinking water. The results reported here must therefore be considered as a contribution to an effort that should be continued. Besides the known constituents of drinking water, some were also considered that it would be plausible to expect to be present, even though they have not yet been detected in water. (Certain pesticides used in large quantities fall into this category.) In our review of water constituents, we have attempted to take into account not only their identities, concentrations, and toxicities, but also to consider other questions, such as: 1. What reason is there for concern about the material? What risks are associated with its presence in water? 2. How does the material get into water? 3. What sources are there other than water?
Approach to the Study 11 4. What contaminants need to be controlled? 5. Are there special places or persons at higher than average risk? 6. Are there essential nutritional requirements for this material? 7. In view of the data at hand, can one say that this is a material that causes temporary ill effects? Permanent ill-e~ects? Reversible ejects? 8. In view of these effects and their reversibility (or lack of it) is it possible to set "no-observed-adverse-health-effects" levels? 9. For materials with special health benefits, what concentrations will maximize these benefits, while keeping the health risk associated with them at an acceptably low level? 10. What additional information is required to resolve the outstanding problems? Many of the constituents of drinking water are natural materials, and enter water from the rocks and the soil and the air. Some are the natural waste products of men or animals. Others are artificial or synthetic materials, made and used for special purposes, that inadvertently find their way into water. Yet others occur naturally, but have become more widely distributed in populated areas as a result of industrial and agricultural activity. WATER CONSUMPTION In this study, a quantity of 2 liters per day has been taken to be the average amount of water consumed per person. This is also the amount used by EPA to calculate the current interim standards. Daily consump- tion of water is a function of temperature, humidity, physical activity, and other factors that vary widely. The average per capita water (liquid) consumption per day as calculated from a survey of nine different literature sources was 1.63 liters (NAS, 1974; McNall and Schlegal, 1968; Wolf, 1958; Guyton, 1968; Evans, 1941; Bourne and Kidder, 1953; Walker et al., 1957; Randall, 1973; Pike and Brown, 1975~. However, the larger volume of 2 liters/day was adopted as representing the intake of the majority of water consumers. We estimate that most-of those who consume more than 2 liters per day still are afforded adequate protection, because the margin of safety estimated for the contaminants is sufficient to offset excess water consumption. Nevertheless, consideratioin should be given to establishing some standards on a regional or occupational basis, to take extremes of water consumption into account. ~, . J
12 DRINKING WATER AND HEALTH RISK AND SAFETY The hazards of ingesting chemical pollutants in drinking water have been assessed in two general ways: with laboratory toxicity studies and epidemiological studies. The aim of studies of both types is to provide information about the risk to man. Risk constitutes only half of the equation; the other half is benefit to the exposed population. It is not possible to guarantee a risk-free society. The scientific methods and criteria we have used for evaluating long-term ejects and risks in man are described in Chapter II, "Chemical Contaminants: Safety and Risk Assessment" and in the chapters concerning each group of contaminants. Most of the experimental results on which the current knowledge of toxicity rests are based on observed effects on man and animals of doses and dose rates that are much larger than those that correspond to the usual concentrations of harmful materials in drinking water. There is, consequently, great uncertainty in estimating the magnitude of the risk to health that ingestion of contaminants in water ~ may produce. An Lou; iiQllal p1 Ill in tO LAKE ill LO account ine combined ejects of two or more contaminants. The theoretical and experimental bases for extrapolating estimations of risk to low levels of dose have been reviewed, and some principles are proposed to guide the conduct of this and similar studies. ~ _ . . ~ ~ . . . . ~ . . .. MICROBIOLOGICAL CONTAMINANTS Outbreaks of waterborne disease are reported to the National Center for Disease Control (CDC) by state health departments. In addition, EPA obtains information about additional outbreaks from state water-supply agencies. Both CDC and EPA are aware that data on waterborne outbreaks have limitations and must be interpreted with caution. The data collected represent only a small part of a larger public health problem. The number and kind of reported outbreaks and of some etiologies may depend upon the interest or capabilities of a particular state health department or individual. They do not reflect the actual number of outbreaks, cases, or etiologies of disease associated with drinking water. Many small outbreaks are not reported to state health departments. There is no law or regulation requiring state authorities to report all gastroenteritis cases to CDC. In 1975, CDC reported 24 waterborne disease outbreaks involving 10,879 cases. No etiologic agent was found for the two largest outbreaks (Sewickley, Pa., 5,000 cases and Sellersburg,
Approach to the Study 13 Ind., 1,400 cases). In fact, no etiologic agent was identified in 17 of the 24 outbreaks. These 17 outbreaks accounted for 9,760 cases, or 89%, of the total reported in 1975. Conclusions in the microbiology chapter, based on epidemiological data, are subject to the limitations of the reporting system and to our limited ability to identify etiologic agents in outbreaks known to be associated with drinking water. The microbiological contaminants selected for consideration in this report are those for which there is epidemiological or clinical evidence of transmission by drinking water. They include a variety of bacteria, viruses, and protozoa. Methods of detecting these contaminants of drinking water were reviewed, and the quantitative relationships between dose levels and infectivity were examined. Because current drinking water standards place major emphasis on detection of microbiological contaminants, attention was devoted to the validity and health sig- nificance of microbiological standards. PARTICULATE CONTAMINANTS Finely divided solid particles of mineral and organic composition are commonly found suspended in some drinking water, particularly those supplies that do not practice coagulation and filtration. To discover whether or not the long-term ingestion of these materials in water is likely to produce adverse ejects on human health, their occurrence, composi- tion, and properties were reviewed. This review indicated that many kinds of particulate matter may indirectly, through adsorption, facilitate the transport of toxic substances and pathogenic organisms and affect the efficiency of disinfection. Particles of organic composition also may indirectly give rise to chlorinated compounds by reaction with chlorine in water treatment. Only in the case of particles derived from asbestos minerals, however, are there grounds for suspecting that direct ejects on human health could be involved. Fibrous particles of asbestos minerals are known to be associated with increased incidence of cancer, including gastrointestinal cancer, among workers who inhale asbestos-laden air. Experiments on the inhalation of asbestos mineral fibers by animals have-also demon- strated a carcinogenic eject. The particulate matter in drinking water often includes similar particles. Although epidemiological studies have not indicated an increase with time in cancer death rates that can be ascribed to fibrous contamination of the drinking water, these negative findings do not exclude the
14 DRINKING WATER AND HEALTH possibility that such an increase may be detected in the future, because many cancers have long induction periods. For these and other reasons, detailed elsewhere, it is believed to be important that research on the analysis of fibrous mineral particles in water, and on the toxicity of these materials when ingested, should be strongly pursued. INORGANIC SOLUTES The Interim Primary Drinking Water Regulations list maximum allow- able concentrations for six metallic elements barium, cadmium, chro- mium, lead, mercury, and silver. Ten additional metals were reviewed in this study beryllium, cobalt, copper, magnesium, manganese, molybde- num, nickel, tin, vanadium, and zinc. Sodium, which is also a metal, was considered separately, because the problems it poses are quite distinct from those associated with the other metallic substances. In addition, the ejects on health of several other inorganic constituents of drinking water were studied. These include arsenic, selenium, fluoride, nitrate, and sulfate. The relationship between water hardness and health also received attention. The sources of inorganic ions in groundwater, surface water, water- treatment chemicals, and from the storage and distribution system are considered along with the health ejects resulting from the total intake from food, air, and water. ORGANIC SOLUTES Of the 298 volatile organic compounds so far identified in drinking water, 74 were selected for detailed study along with 55 pesticides. A compound was selected for consideration if any of the following criteria applied: Experimental evidence of toxicity in man or animals, including carcinogenicity, mutagenicity, and teratogenicity. 2. Identified in drinking water at relatively high concentration. 3. Molecular structure closely related to that of another compound of known toxicity. 4. Pesticide in heavy use; potential contaminant of drinking water supplies. 5. Listed in the Safe Drinking Water Act or National Interim Primary Drinking Water Regulations.
Approach to the Study 15 The evaluation of toxicity was based on a critical review of the scientific literature. The available data were of variable quality and quantity and, in some instances, inadequate for proper assessment of toxicity. In those cases where sufficient data were available, professional judgment was used to determine which compounds are carcinogenic, mutagenic, teratogenic, and noncarcinogenic. The limitations that are inherent in the extrapolation of high-dose animal bioassay data to low-dose human exposure and the difficulty of making predictions for species that may have different metabolic rates and pathways for handling carcinogens, or different target-organ responses, are well known. Such risk assessment and extrapolation procedures are further compromised by the limited information that is available concerning the mechanisms by which these agents act (such as initiators, promoters, and modifiers) and the almost total lack of data regarding the potential synergistic and antagonistic interactions of these agents with each other and with other environmental agents. The risk of ingesting known or suspected carcinogens was estimated by the methods described in Chapter II. These methods are based on an assumption that there is no threshold in the dose-response relationship. The risk-estimate approach may provide unique advantages for other areas of toxicologic evaluation. The more traditional approach of combining the maximum no- observed-adverse-effect level with an uncertainty (safety) factor to calculate an acceptable daily intake (ADI) was used for agents that were not considered to be known or suspected carcinogens and for which there was adequate toxicity data from prolonged ingestion studies in man or animals. Several alternative terms other than ADI were considered, but it was concluded that the introduction of new terms might lead to confusion and that the use of a widely recognized and generally acceptable term would be preferable for this report. Although the ADI has been used previously as an internationally established standard for the toxicological evaluation of food additives and contaminants, the concept is applicable to other cases of exposure by ingestion. The ADI is an empirically derived value that reflects a particular combination of knowledge and uncertainty concerning the relative safety of a chemical. The uncertainty factors used to calculate ADI values in this report represent the level of confidence that can be justified on the basis of the animal and human toxicity data. ADI values were not calculated for agents where the data were considered to be inadequate. Since the calculation of the ADI values is based on the total amount of
16 DRINKING WATER AND HEALTH a chemical ingested, the ADI values calculated in this report do not represent the safe level for drinking water. Little or no data are available on the toxicity of many organic compounds identified in drinking water. There is a need to determine which of these compounds should be subjected to extensive toxicity testing. Some of the criteria used for developing the order in which compounds should be tested are: 1. The relative concentrations of the compounds and the number of people likely to be exposed. 2. The number of supplies in which they occur. 3. Positive responses in in vitro mutagen screening systems. 4. Positive responses in in vitro prescreening systems for potential carcinogens (mammalian cell transformations). 5. Similarity of the chemical structure of the test compound with that of other compounds having defined toxic properties (i.e., structure- activity relationships). 6. Relationships of dose from water to total body burden. RADIOACTIVE CONTAMINANTS Because the presence of ionizing radiation is one of the standard features of the earth's surface, the adverse effects on health that may be ascribed to radioactive contaminants of drinking water were assessed in relation to the average background radiation dose, from all sources, of 100 mrem per year. Previous estimations of the biological effects of the background radiation on human health were reviewed in the light of more recent scientific knowledge and used to calculate the magnitude of three kinds of adverse health ejects that radiation can produce; namely, developmental and teratogenic ejects on the fetus, genetic disease, and somatic (principally carcinogenic) ejects. When these estimates are related to the concentrations of radionuclides that are commonly found in drinking water, it is seen that consumption of 2 liters of water per day contributes such a small fraction to the total radiation background that the incidence of developmental, teratogenic, and genetic disorders is not increased enough for the change to be detectable. Where somatic ejects are concerned, it is estimated that the radionu- clides in drinking water typically account for less than 1% of the incidence of cancers that may be attributed to the natural background of
Approach to the Study 17 radiation. Only certain bone-seeking radionuclides (chiefly radium), in a few regions, reach concentrations in drinking water that are high enough to cause a significant increase in the incidence of bone cancer. SUSCEPTIBLE SUBGROUPS AND OTHER CONSIDERATIONS Groups that are more susceptible than the normal population are considered in the chapters on various classes of contaminants. This report is concerned only with water used for drinking. Although all contaminants may cause problems when present in water used in health care facilities, the health hazards associated with such diverse uses of water as in humidifiers, kidney dialysis units, laundries, heating and cooling equipment, or many special uses that require further treatment of tap water, have not been considered. References and summaries of the scientific literature in this field have been published by DeRoos et al. (19741. REFERENCES Bourne, G.H., and G.W. Kidder, eds. 1953. Biochemistry and Physiology of Nutrition, vol. 1. Academic Press, New York. DeRoos, R.L., V.R. Oviatt, A.G. DuChene, and N.J. Vick. 1974. Water use in biomedical research and health care failities-A presentation of article summaries. National Institutes of Health, Department of Health, Education, and Welfare, Contract no. NIH- ORS-72-2 1 1 1. Evans, C.L. ed. Starling's Principles of Human Physiology, 8th ed. Lea and Febiger, Philadelphia. Federal Register, Wednesday, December 24, 1975, vol. 40, no. 248. Guyton, A.C. 1968. Textbook of Medical Physiology, ad ed. W.B. Saunders Co., Philadelphia. McNall, P.E., and J.C. Schlegel. 1968. Practical thermal environmental limits for young adult males working in hot, humid environments. ASlIRAE Transactions 74:225-235. National Academy of Sciences-National Research Council. 1974. Recommended Dietary Allowances, 8th ed. Washington, D.C. Pike, R.L., and M. Brown. 1975. Minerals and Water in Nutrition-An Integrated Approach. 2d Ed. John Wiley, New York. Randall, H.T. 1973. Water, electrolytes and acid base balance. In R.S. Goodhart and M.E. Shils, eds. Modern Nutrition in Health and Disease. Lea and Febiger, Philadelphia. Walker, B.S., W.C. Boyd, and I. Asimov. 1957. Biochemistry and Human Metabolism, 2d ed. Williams & Wilkins Co., Baltimore. Wolf, A.V. 1958. Body water. Sci. Am. 99:12S.