REGULATORY

TOXICOLOGY

AND

PHARMACOLOGY

13, 18-35 (1991)

A Survey and Analysis of States’ Methodologies for Deriving Drinking Water Guidelines for Chemical Contaminants JEFFREY M. PAULL,*‘t”

LOIS M. JOELLENBECK,? AND KIRPAL S. SIDHU*$

ROGER C. COCHRAN,?

*Federal-State Toxicology and Regulatory Alliance Committee,. TMaryland Department of the Environment, Environmental Toxicology and Risk Assessment Division. 2500 Broening Highway, Baltimore, Maryland 21224; and #Center for Environmental Health Sciences, Michigan Department of Public Health. P.O. Box 30195, Lansing, Michigan 48909

Received May 18. 1990

Data from a national survey questionnaire regarding the development of guidelines for chemical contaminants in drinking water were collected from all SO states. Twenty-three states develop at least some of their own guidelines; the other 27 states rely on guidelines previously developed by the U.S. Environmental Protection Agency (USEPA) or by other states.States which derive guidelines generally employ toxicological criteria and risk assessment methodologies developed by the USEPA. Fourteen of the twenty-three states that develop their own guidelines depend on cancer potency factors derived by the USEPA to establish risk-based concentrations for carcinogens. Most of the statesdevelop guidelines based on preventing possible excesscancer risk greater than one in one million. Seventeen of these twenty-three states rely on oral reference doses (RtDs) to derive guidelines for noncarcinogens. Examination and clarification of the states’ approaches to guideline derivation reveal that although similar risk assessmenttechniques are generally employed. differences in assumptions, chemical classifications, and uncertainty factors may lead to variation in resultant guidance levels. Improved communication and coordination between states and the federal government may help reduce the variations and inconsistencies among the states in establishing drinking water guidelines for chemical contaminants. o 1991 Academic PESS, IIIC.

INTRODUCTION Contamination of groundwater by organic and inorganic chemicals, radionuclides, and microorganisms has occurred in every state, and is being detected with increasing frequency (OTA, 1984). As approximately 20% of the nation’s population may rely on private wells for drinking water (OTA, 1984) contamination of drinking water supplies is becoming an increasingly serious problem. ’ To whom all correspondence should be addressed at present address: ChemRisk McLaren/Hart, Atlantic Avenue, Alameda, CA 9450 I. 18 0273-2300191 $3.00 copyrisht0 1991 by Academic All rights of reproduction

Press, Inc. in any form reserved.

1 I35

STATES

METHODOLOGIES

FOR

DRINKING

WATER

GUIDELINES

19

Federal and state programs are designed primarily for the protection of public drinking water supplies. The Safe Drinking Water Act as amended in 1986 grants the U.S. Environmental Protection Agency (USEPA) the authority to set standards for drinking water contaminants that pose a health risk. Except for Indiana and Wyoming, all other states are responsible for enforcing the drinking water standards for public water supplies. A state may adopt the USEPA standard or choose a more stringent standard for a contaminant. Also, a state may regulate additional contaminants for which EPA has not developed a standard. The drinking water standards are not enforceable for contaminants present in the water from private wells. The EPA develops maximum contaminant level goals (MCLGs) and maximum contaminant levels (MCLs) to regulate contaminants in public water supplies. The MCLGs are nonenforceable health goals developed by the USEPA, defined as “the maximum level of a contaminant in the drinking water at which no known or anticipated adverse effect on the health of persons would occur and that allows an adequate margin of safety” (USEPA, 1987). For chemicals classified by USEPA as carcinogens or probable human carcinogens, the MCLG is set at zero (USEPA, 1987). MCLs are enforceable standards developed by the USEPA to regulate contaminants in public water supplies. Though human health is one of the primary factors considered in establishing MCLs, economic impact and technological feasibility considerations are also factored in. Under federal law, MCLs are enforceable only for systems serving at least 25 of the same persons over 6 months of the year or with 15 or more service connections used by year-round residents (USEPA, 1987). An Office of Technology Assessment (OTA) study reported the presence of over 200 substances in groundwater in the United States (OTA, 1984). These substances included about 175 organic chemicals, approximately 50 inorganic chemicals (metals, nonmetals, and acids), biological organisms, and radionuclides. At the time of the study, OTA estimated that standards or guidelines of some type (state or federal) had been promulgated for less than one-half the substances that were detected in groundwater (OTA, 1984). At present, the USEPA has established 30 drinking water standards or MCLs for environmental contaminants. On May 22, 1989, the Agency proposed drinking water standards for an additional 30 organic and 8 inorganic substances (USEPA, 1989). In addition to standards, the USEPA has developed Health Advisories for approximately 110 contaminants. Health Advisories are purely health-based guidelines developed by the USEPA Office of the Drinking Water. These guidelines are set at contamination levels for which adverse noncarcinogenic health effects would not be anticipated. Adequate margins of safety for 1-day, 1O-day chronic, and lifetime exposure periods are factored in (Ware, 1988). Faced with necessary decisions concerning the potability of drinking water supplies for an increasing array of chemical contaminants, many states have recently become involved in the process of establishing drinking water guidelines and standards. These guidelines and standards provide a framework for detection, correction, and prevention activities. As in the federal system, guidelines usually refer to nonenforceable advisory levels, while standards are enforceable regulations. When a numerical limit is necessary states may prefer to develop guidelines rather than standards to avoid the cumbersome time- and resource-intensive process that generally accompanies standard setting. Moreover, guidelines are primarily health-

20

PAULL

ET AL.

based limits, while standards must take account of other factors, such as feasibility and cost. Increasingly, states are employing methodologies involving risk assessment techniques for the derivation of drinking water guidelines and criteria. A survey performed in 1987 by the Chemical Communications Subcommittee of the Federal-State Toxicology and Regulatory Alliance Committee (FSTRAC) revealed major differences in the drinking water guidelines developed by various states for chemical contaminants. For example, the drinking water guideline for the pesticide alachlor varied between 0.15 and 200 ppb in different states, while the guideline for the chlorinated hydrocarbon 1,l -dichloroethane varied from 1 to 850 ppb. A followup survey was suggested by the Subcommittee to identify and summarize the specific processes used by different states to establish standards or guidelines for drinking water contaminants. This national follow-up survey was initiated to clarify the guideline-development process among the states. Central to the purpose of this survey was elucidating the various types of approaches the states use for the derivation of drinking water guidelines in the absence of USEPA-developed criteria. METHODS A total of 102 survey questionnaires were sent to state agencies or EPA regional representatives in all 50 states in February 1989. The FSTRAC membership list was used to identify contact individuals. Survey questionnaires were also sent to the directors of appropriate state agencies for forwarding to those responsible for drinking water programs. The survey questionnaire included 20 questions designed to identify the agency involved in developing drinking water guidelines and establish the methodologies used. Responders were asked to rank their reliance on different sources of USEPA carcinogen data, and how they utilized various types of information in the absence of USEPApublished criteria. States were asked how they derive risk-based concentrations, how varying exposure periods are handled, which models for cancer potency factor calculations are used, how group C carcinogens are treated, whether USEPA weight-ofevidence classifications are used, and whether mechanism of carcinogenicity is considered in guideline formulation. Similar questions were asked about the development of guidelines for noncarcinogens, including the ranking of USEPA toxicity data and other information sources, in the absence of USEPA standards or guidelines. The approaches states use to derive guidance criteria were examined with regard to the uncertainty factors used in calculations, information sources used to locate toxicity data, and whether the guidelines they develop are peer or otherwise reviewed. Survey responses from all 50 states were clarified and verified by telephone during the summer of 1989, with effort made to ensure that the most appropriate representatives of the states’ programs responded. Information gathered during the telephone discussions with representatives in all 50 states supplemented the questionnaire responses. Results are reported in simple percentages reflecting a ratio of the number of states using a particular method to the number of states analyzed. A list of the state agencies that responded, and a copy of the survey questionnaire are presented in Appendices 1 and 2, respectively.

STATES’

METHODOLOGIES

FOR DRINKING

WATER GUIDELINES

21

RESULTS Based on the survey responses, 23 of the 50 states were found to develop drinking water guidelines for chemical contaminants. As the survey questionnaire was designed primarily to clarify state methodologies for deriving guidelines in the absence of USEPA-developed criteria, only the data submitted for the states that independently derive guidelines could be compiled and analyzed. Therefore, the results of the survey and inferences drawn from these results will apply to these 23 states, identified in Fig. 1. Except as otherwise noted, reported percentages are based upon these 23 states. State responses to the survey may reflect the views of state personnel involved in groundwater quality programs. The extent to which the response of a single individual accurately represents state activities is uncertain. It may depend, in part, on the relative role of their agency in dealing with drinking water contamination and the role of that individual within that agency. Since many states are actively developing or revising their development programs for drinking water guidelines and standards, the responses reflect program status only as of the date of the questionnaire. Guidelines for Carcinogens When a guideline is needed for a carcinogenic contaminant of a water supply where use of a MCL is not mandatory, the survey indicated that MCLs, if available, are heavily relied upon as surrogate guidelines for carcinogenic substances (Fig. 2). Twentyone of the twenty-three states (91%) indicated that they make some use of MCLs in their guideline development process and 57% indicated the MCL as their first choice. Excess cancer risk levels from Health Advisories developed by the USEPA Office of Drinking Water are used in guideline development by 96% of these 23 states, with 6 1% indicating the Health Advisory risk levels as a first or second preference. Ninetysix percent of the states indicated use of best available toxicological data but only 39% indicated use of toxicological data as a first or second preference in deriving guidelines. Most of the 23 states (57%) rely on toxicity data as a third or last choice following the use of USEPA criteria. Maximum contaminant level goals for carcinogens are not used at all by 35% of these states and an additional 30% indicated their use only as a last choice.

FIG. 1. States’ development of drinking water guidelines map. (a) Independently develop drinking water guidelines. (0) Do not independently develop drinking water guidelines.

22

PAULL

ET AL.

60 L ;

50

F4 T

40

6 E 0 F

30

MCL

MCLG

HLTH

ADVSY

-

FIRST

FEZ4

SECOND

0

THIRD

m

FOURTH UNRANKED

TOX DATA

TYPE OF DATA 2. States’ ranking of data for development of guidelines for carcinogens.

FIG.

When a USEPA MCL, MCLG, or Health Advisory is not available for a carcinogenic contaminant, all 23 states reported developing drinking water guidelines using riskbased concentrations. Eighty-seven percent of these states indicated the risk-based concentration as their first preference and only 13% rely primarily upon the analytical detection limit (Fig. 3). Feasibility/cost is not considered at all by 74% of these states. In general, these states follow USEPA risk assessment guidelines and procedures to derive their guidelines for carcinogens. Most of them use cancer potency factors developed by USEPA in conjunction with exposure data to estimate possible cancer risks. Although most derive guidelines based on preventing possible excess

-7 P

,

i

!'

80

El -

FIRST

@%

SECOND

E

0

THIRD

0 F

m

FOURTH

40 m

UNRANKED

s A

20

'G

60

I s 0 HEALTH

RISK

DETECT

TYPE FIG.

LIMIT

FEAS/COST

OF CRITERIA

3. States’ ranking of criteria for development of drinking water guidelines.

STATES

METHODOLOGIES

FOR

DRINKING

WATER

GUIDELINES

23

cancer risk greater than 1 in one million (lo-‘j), four states employ a risk level of 1 in 100,000 ( 1O-5). All 23 states use or develop risk-based concentrations for lifetime exposures. Three of the states establish guidelines for acute exposures, and five also establish guidelines for chronic and less-than-lifetime exposures. The survey did not determine whether less-than-lifetime guidelines were based on time-weighted cancer risk levels or noncancer threshold effects. Thirty-nine percent of the twenty-three states develop cancer potency factors and derive risk-based concentrations directly from the primary toxicological and epidemiological data, usually using the linear multistage model. Two states, however, also use the Weibull time-to-tumor model when the appropriate data are available. The remaining 6 1% of the states, which also use toxicology data, do not derive cancer potency factors. Instead, they depend on potency factors derived by USEPA to establish risk-based concentrations. Eighty-three percent of the 23 states follow the EPA weight-of-evidence classification scheme describing carcinogens (USEPA, 1985). In the case of substances for which the weight of evidence for carcinogenicity is equivocal (group C carcinogens), 48% of the states indicated that they follow a procedure similar to that of the EPA Office of Drinking Water, of dividing the drinking water equivalent level (DWEL) by an additional uncertainty factor of 1 to 10 to represent possible carcinogenicity (Ware, 1988). Thirty-five percent of the 23 states indicated that their treatment of these compounds varies depending on the evidence or information available, and the remaining 17% treat group C carcinogens exactly as group A and B carcinogens. Only two states indicated that they consider the contaminant’s specific mechanism of carcinogenicity in developing their guidelines. Guidelines for Noncarcinogens The development of guidelines for noncarcinogens is based on a threshold concept. The USEPA Office of Drinking Water usually derives MCLGs and Health Advisories for contaminants from toxicity studies performed on laboratory animals. Doses at the no-observed-adverse-effect level (NOAEL) or the lowest-observed-adverse-effect level (LOAEL) are adjusted with uncertainty factors to compensate for limited data on human health effects and differences in response to toxicants within and between species to arrive at a reference dose (RlD). This value is then multiplied by dose scaling and exposure factors to account for typical human body size, drinking water consumption, and contributions from other sources to arrive at a Drinking Water Equivalent Level (DWEL). In deriving guidelines for noncarcinogens, 96% of the 23 states take the MCL into account where it exists, and 57% rely on the MCL as a first choice in deriving guidelines for noncarcinogens (Fig. 4). In contrast with the limited use of MCLGs for carcinogens, 96% of the 23 states use the MCLG in establishing guidelines for noncarcinogens. It is selected as a first or second priority by 6 1%. Five of the states prefer to derive guidelines from primary toxicological data for noncarcinogens, even where an MCL, MCLG, or Health Advisory is available. When MCLs, MCLGs, and Health Advisories are not available, 74% of the 23 states indicated that they rely on the oral reference dose (RID) to derive guidelines for a

24

PAULL

ET AL.

C F4 50 L G E

z

40

-

FIRST

m

SECOND

0

THIRD

30

FOURTH a

;

20

8E

10

UNRANKED

0 MCL

MCLG

TYPE FIG. 4. States’

HLTH

OF

ADVSY

TOX

DATA

DATA

ranking of data for development of guidelines for noncarcinogens.

contaminant. A NOAEL, from which an RfD may be derived, was the most frequent second choice (44%) of toxicity data. The acceptable daily intake (ADI), the predecessor to the RID, was most often selected as a second (30%) or third choice (26%). The lowest-observed-adverse-effect level (LOAEL) was ranked low or last by 65% of the states (Fig. 5). Thirty percent of the states indicated use of other sources of information in the absence of USEPA-developed criteria. These sources included the toxicology data on which the NOAELs are based, acute toxicity data such as LD+, suggested no-adverseeffect levels (the predecessor to the Health Advisory), acceptable-intake chronic levels, and other states’ guidelines.

60

40

:Ii/ I -

FIRST

=

SECOND

0

THIRD FOURTH

20

0 RID

ADI

NOAEL

TYPE

OF

LOAEL

DATA

OTHER

m

,Z,FT,,

17

UNRANKED

FIG. 5. States’ ranking of toxicological data for noncarcinogen guideline development.

STATES

METHODOLOGIES

FOR

DRINKING

WATER

GUIDELINES

25

To derive an RlD from NOAELs or LOAELs, the USEPA utilizes standard uncertainty factors to represent various categories of uncertainty involved in the risk assessment process for intraspecies variation, interspecies variation, less-than-lifetime data, and adequacy of the database (USEPA, 1985). Table 1 lists the percentage of the 19 responding states which indicated use of these uncertainty factors. Data Sources and Guideline Review There are a variety of sources which states can access for toxicity data. Of the 22 states who ranked their use of these alternate sources of data for the purpose of developing their own guidelines, 59% selected the Integrated Risk Information System (IRIS) database as their primary data resource. The IRIS computer database contains risk assessment information on about 400 chemicals. It includes RfDs for noncarcinogens, drinking water contaminant health advisories, cancer potency factors, riskbased concentrations, appropriate EPA contact numbers, and a variety of other information helpful for risk assessment (Sidhu, 1989). The MEDLARS TOXNET system of databases, of the National Library of Medicine, Bethesda, Maryland, which includes Chemical Carcinogenesis Research Information System (CCRIS), Registry of Toxic Effects of Chemical Substances (RTECS), and Hazardous Substances Data Bank (HSDB), was selected second by 46% of the 22 states. In March 1990, the IRIS database was included in the MEDLARS TOXNET system (Wexler and Vasta, 1990). Reference texts and journals were chosen equally as third choices by 27% and the Federal Register along with other unspecified sources of information were indicated most frequently as lowest priority resources (Fig. 6). Most of the states (6 1%) indicated the existence of some form of review process for their guidelines, ranging from informal review within their agency to a more formalized peer review and public comment process. DISCUSSION The results of this survey showed that 23 states currently engage in efforts to derive drinking water guidelines when USEPA-developed guidelines are not available. This TABLE PERCENTAGE OF I9 STATES USING IN DERIVING GUIDELINES Uncertainty

factor

applied

Intraspecies variation Interspecies variation Dose-route extrapolation Less-than-lifetime data Other-than-NOAEL data Mutagenicity Teratogenicity Carcinogenicity Adequacy of database Other factors

I VARIOUS UNCERTAINTY FOR NONCARCINOGENS

FACTORS

Percentage

of states 95 95 14 95 95 26 3-l 63 95 5

26

PAULL ET AL. 60

; 50 E

E

40

Y

m

FIRST

=

SECOND

m

THIRD

m

F,FT,,

0

SIXTH

FOURTH

UNRANKED

IRIS

TOXNET

REF

SOURCE

TEXTJOURNAL

FED

REG

OF DATA

FIG. 6. States’ ranking of data sources.

result is largely consistent with a previous FSTRAC survey on state guidelines and standards (Krietzman et al., 1988). Whenever USEPA guidelines such as MCLs and Health Advisories are available there is little deviation from the federally established norms. However, when EPA guidelines for drinking water contaminants are not available, these states develop their own guidelines by using USEPA-approved methodologies with EPA-derived criteria such as cancer potency factors and RID values. In some cases these states develop their own toxicological criteria, such as cancer potency values, to derive their guidelines. The survey data indicate that most states prefer not to take the lead in developing water quality criteria. Only a few states rely upon primary toxicological data to derive their guidelines when there are existing EPA guidelines for the chemical contaminants of interest. This appears to be true of guidelines for both carcinogens and noncarcinogens. Although the states will turn to toxicological data to derive their own guidelines when necessary, it is apparent that there is a reluctance to do so, and most rely on USEPA standards or guidance levels when they are available. The data in Fig. 2 suggest that for the many carcinogenic contaminants for which no MCL has been established, most of the states rely, in practice, upon Health Advisories developed by the USEPA Office of Drinking Water. Interestingly, the MCLG for carcinogens was ranked last as an information source and is not used at all by 65% of the 23 states. The USEPA generally sets MCLGs for carcinogens at zero, reflecting its policy that no safe level can exist for carcinogens (USEPA, 1987). When USEPA criteria for carcinogens are not available, all 23 states indicated that they develop guidelines using health risk-based concentrations, indicating that the states’ primary concern is with protecting public health. The survey did not determine how the states that consider nonhealth factors such as cost, feasibility of treatment, and detection techniques use these factors in setting guidelines for carcinogens. The survey results for noncarcinogens also indicated heavy reliance on USEPA criteria in developing guidelines. As in the case of carcinogens, MCLs for noncarcin-

STATES

METHODOLOGIES

FOR

DRINKING

WATER

GUIDELINES

27

ogens, where available, are generally given first preference (Fig. 4). Since MCLs have been established for only a limited number of chemicals, however, states turn frequently to MCLGs. MCLGs for noncarcinogens, unlike those for carcinogens, assume a threshold effect and are therefore set at nonzero values. The states apparently find these values more useful in their guideline derivations than the zero MCLG values recommended for carcinogens. When guidelines for noncarcinogens must be developed in the absence of USEPA guidance levels, states rely upon other forms of toxicological information (Fig. 5). Derived criteria, such as the RfD, appear to be preferred over primary toxicity data, possibly because the use of primary data is more time and resource intensive. When using primary data, 18 of the 19 states reporting use of uncertainty factors employ the standard USEPA uncertainty factors to adjust for intra- and interspecies variability; less-than-lifetime and other-than-NOAEL data, and for the adequacy of the database. A number of states, however, use additional factors to represent other sources of uncertainty, such as dose-route extrapolation, and possible mutagenicity, teratogenicity, and carcinogenicity (Table 1). For both carcinogens and noncarcinogens, the heavy reliance placed upon the MCL suggests some degree of inconsistency in the states’ policies toward developing their drinking water guidelines. Guidelines are generally distinguished from standards in that they are principally health-based levels. As state guidelines are usually developed to make decisions for small or individual water supply systems, they do not need to factor in cost or feasibility considerations, as costs are generally minimal for small systems. Yet in establishing guidelines for contaminants found in small water systems or private wells, states are placing heavy reliance on the MCLs, which are influenced by the feasibility and cost considerations involved with large public water supply systems, as well as by health considerations. The tendency to use MCLs and Health Advisory guidance levels instead of other sources of toxicological data reflects the states’ preference for previously established drinking water guidelines which are readily available. This preference is understandable in view of the limited technical and financial resources of most of the states. State decisions concerning drinking water contamination are often made in response to the discovery of contamination of a residential or small community water supply. In these circumstances, decisions concerning potability must be prompt and unambiguous, and the use of a USEPA-developed guidance level has the inherent advantages of being less time consuming to derive and easier to defend, both legally and politically, than values independently developed by the states. In addition, for chemicals where an MCL exists, states cannot apply criteria which are less stringent (US Congress, 1986). States’ prioritization of sources of data may reflect the time pressure under which guidelines must be developed. Of the 22 states who ranked their preference of data sources, the first choice of 16 were databases that provide concise summations of toxicity studies. Since the Federal Register contains only USEPA’s brief synopsis of the toxicological and epidemiological literature for contaminants, and this source is less accessible and convenient than the toxicological on-line databases, its utility to the states appears to be limited. The survey results indicated a reliance on EPA-developed drinking water criteria and in their absence, development of guidelines by routes and methods which are

28

PAULL

ET AL.

maximally expedient. Indeed, many states indicated that time and resource constraints make this almost inevitable. Most of the states which develop their own drinking water guidelines do so on a case-by-case basis. The methods used depend a great deal on the professional judgment of the staff toxicologists and the policies of the individual state, and therefore, the resultant guidelines may be either more or less stringent than those subsequently developed by the EPA. Although the methodologies employed are similar, there is presently no single uniform approach to the development of drinking water guidelines. As a result, different guidelines are often developed by different state agencies for the same drinking water contaminants. This finding supports OTA’s observation that people in different states currently do not receive a uniform level of health protection against groundwater contaminants (OTA, 1984). Only a small number of states have adequate resources to repeat, parallel, or go beyond the work EPA does in developing drinking water guidelines. These states develop their guidelines from toxicity data using methods which sometimes differ from those of the USEPA and examine the documentation and data supporting USEPA-developed guidelines to assesstheir adequacy. The work of these states provides a valuable source of information for certain drinking water contaminants. Although only the 23 states which currently engage in guideline development for drinking water contaminants could respond in detail to the survey questionnaire, information received from the other 27 states regarding their programs also proved enlightening. Many of these states do establish drinking water guidelines and standards, but their responses indicated that their programs incorporate standards or advisory levels developed by USEPA or by other states, and that they do not currently have methodologies in place for deriving their guidelines directly from toxicological data. Just as the states that independently develop guidelines utilize a variety of risk assessment practices and techniques, these states also employ a spectrum of practices for establishing guidelines, ranging from very informal case-by-case decisions to formal procedures for evaluating and adopting guidance levels. The distinction between the two groups of states should neither be construed as a value judgment concerning the relative merit of state programs for establishing and implementing drinking water guidelines, nor be interpreted as an indication of overall program effectiveness. Some states with strong regulatory programs incorporating enforceable numerical limits for drinking water contaminants may not develop their own guidelines, while other states which do develop guidelines may have few enforceable state standards. This survey has identified a fundamental difference in approaches to guideline development, but further policy research concerning drinking water standards in state regulatory programs needs to be conducted to evaluate questions concerning the effectiveness of these programs. The wide variations among the states in their approach to the development of drinking water guidelines for chemical contaminants suggest several explanations. States vary in their capabilities, perception of contamination problems, priorities of sources and users, stages of program development and implementation, and institutional arrangements (OTA, 1984). The frequency of contamination incidents and the availability of resources to support qualified staff are also likely to be important factors. The nature of the toxicological database is continually changing, and some states are able to research and update their guidance levels more readily than others. Differences

STATES

METHODOLOGIES

FOR

DRINKING

WATER

GUIDELINES

29

in economic and political priorities, as well as the level of public awareness and concern, may also exert significant influences. Supplemental survey data and additional comments were related primarily to having sufficient resources and support to fully establish or implement drinking water criteria development. Several states also noted a number of problems with their institutional framework, including lack of authority to deal with drinking water contamination and inability to develop and implement a coordinated strategy. Resolution of these problems may be complicated by the lack of support of various interest groups, policy conflicts or coordination problems among state agencies and between state and federal programs, and the low priority of state research programs relative to enforcement programs. Some organizations and individuals have suggested that inconsistencies in the relative stringency of state regulatory programs could affect states’ abilities to attract industry. However, a review conducted by the U.S. General Accounting Office showed that at least with regard to air toxics, environmental regulation does not generally affect industry location decisions to a significant extent (USGAO, 1987). Several states are involved in organizations designed to increase the consistency and validity of drinking water guidelines. Regional groups like the Northeast Regional Environmental Public Health Center and the Risk Assessment Council of Great Lakes Governors provide mechanisms for review and sharing of work. Recently, the National Governors Association has proposed the creation of a State Environmental Health Studies Clearinghouse. This would allow states to compare risk assessment protocols and policies, evaluate approaches to environmental health problems, and develop a national consensus on issues (NGA, 1990). Organizations such as the Federal-State Toxicology and Regulatory Alliance Committee provide important forums in which the federal government can facilitate the exchange of information between the states. Information exchange includes not only the details of specific case studies, but also programmatic information about state approaches to public health protection. This enables greater coordination between states which fosters greater consistency in the development of health-based drinking water guidelines. For all 50 states to individually recreate the USEPA’s work in deriving drinking water guidelines would be enormously redundant. Much duplication of effort has already occurred on the state level, as the USEPA has yet to address many of the chemicals making their way into the public water supplies. In the interim, it is suggested that resources, such as those described above, be developed on a regional or national level for quickly responding to states’ needs for guidelines for specific contaminants and for facilitating the exchange of information. Once such mechanisms are developed for accessing and sharing information, greater consistency will be achieved in the process of establishing safe levels of contaminants in drinking water.

30

PAULL

ET AL.

APPENDIX State Agencies/Individuals

Responding to the FSTRAC Survey Questionnaire

Alabama Department of Environmental Management Montgomery, Alabama J. Power Alaska Department Conservation Juneau, Alaska C. Denys

of Environmental

Arizona Department of Health Services Division of Disease Prevention Office of Risk Assessment and Investigation Phoenix, Arizona N. Peterson Arkansas Department of Health Division of Engineering Little Rock, Arkansas H. Seifert California Department Services Berkeley, California D. Spath

of Health

Colorado Department Denver, Colorado J. Biberstine

of Health

Connecticut Department Services Toxic Hazard Section Hartford, Connecticut C. J. Dupuy

1

of Health

Delaware Division of Public Health Office of Sanitary Engineering Dover, Delaware R. Howell Florida Department of Health and Rehabilitation Tallahassee, Florida T. Atkeson

Georgia Department of Natural Resources Environmental Protection Division Drinking Water Program Atlanta, Georgia F. Lehmann Hawaii State Department of Health Hazard Evaluation and Emergency Response Program Honolulu, Hawaii J. Ingoglia Idaho Department of Health and Welfare Division of the Environment, Water Quality Bureau Boise, Idaho R. Rogers Illinois Environmental Protection Agency Office of Chemical Safety Springfield, Illinois T. Homshaw Indiana Department of Environmental Management PWS Section Indianapolis, Indiana A. Viere Iowa Department of Natural Resources Water Supply Section Des Moines, Iowa D. Alt Kansas Department Environment Topeka, Kansas V. Robbins

of Health and

Kentucky Department for Environmental Protection Division of Water Frankfort, Kentucky J. Smithers

STATES

METHODOLOGIES

FOR

Louisiana Department of Health and Hospitals Drinking Water Program New Orleans, Louisiana J. Ray Maine Department Bureau of Health Augusta, Maine R. Frakes

of Human

Services

Maryland Department of the Environment Environmental Toxicology and Risk Assessment Division Baltimore, Maryland J. Paul1 Massachusetts Department of Environmental Quality Engineering Office of Research and Standards Boston, Massachusetts K. Martin Michigan Department of Public Health Center for Environmental Health Sciences Lansing, Michigan K. Sidhu Minnesota Department of Health Minneapolis, Minnesota L. Gust Mississippi Department of Health Division of Public Health Laboratories Jackson, Mississippi R. Andrews Missouri Department of Health Jefferson City, Missouri J. Crellin Montana Department of Health and Environmental Services Billings, Montana K. Miller Nebraska Department of Health Public Drinking Water Section Lincoln, Nebraska B. Lee

DRINKING

WATER

31

GUIDELINES

Nevada Health Division Carson City, Nevada L. Rountree New Hampshire Division of Public Health Services Concord, New Hampshire J. Dreisig New Jersey Department of Environmental Protection Division of Science and Research Trenton, New Jersey L. Jowa New Mexico Health and Environment Department Environmental Improvement Division Santa Fe, New Mexico R. Gallegos State of New York Department of Health Bureau of Public Water Supply Protection Albany, New York R. Entringer North Carolina Division of Health Services Environmental Epidemiology Raleigh, North Carolina T. Taylor North Dakota Department Bismarck, North Dakota F. Schwindt Ohio Environmental Columbus, Ohio K. Leifheit Oklahoma Oklahoma F. Walker

of Health

Protection

State Department City, Oklahoma

Agency

of Health

Oregon Department of Human Resources Health Division, Drinking Water Program Portland, Oregon J. Boydston

32

PAULL

Pennsylvania Department of Environmental Resources Bureau of Community Environmental Control Harrisburg, Pennsylvania J. Wroblewski

ET AL.

Vermont Department Burlington, Vermont B. Bress

Rhode Island Department of Health Providence, Rhode Island B. Matyas South Carolina Department of Health and Environmental Control Division of Health Hazard Evaluation Columbia, South Carolina R. Marino South Dakota Department Natural Resources Pierre, South Dakota D. Busch

of Water and

Tennessee Department of Health and Environment Division of Water Supply North Nashville, Tennessee W. Draughon Texas Department of Health Division of Water Hygiene Austin, Texas A. Bennett

of Health

Virginia Department of Health Division of Water Supply Engineering Richmond, Virginia A. Hammer Washington Department of Social and Health Services Office of Environmental Health Programs, Toxic Substances Section Olympia, Washington D. Nash West Virginia Department of Health Office of Environmental Health Services Charleston, West Virginia D. Kuntz Wisconsin Department Social Services Division of Health Madison, Wisconsin D. Belluck

of Health and

Wyoming Division of Health and Medical Services Environmental Health Cheyenne, Wyoming H. Hutchings

Utah Department of Health Bureau of Drinking Water Sanitation Salt Lake City, Utah K. Bousfield APPENDIX

Methodologies

2:

FSTRAC SURVEY-QUESTIONNAIRE for Deriving Drinking Water Guidelines/Standards

1. Please state your name, agency, address, and telephone: 2. Which Agency in your state is responsible for developing drinking water guidelines/ standards? Agency Department As legal standards 3. How are they used: As guidelines

STATES

METHODOLOGIES

FOR

DRINKING

WATER

33

GUIDELINES

4. If used as legal standards, which agency in your state is responsible for promulgating them as regulations? 5. For carcinogens, guidelines/standards are based on which ofthe following? (Please rank in order of priority, from 1 to 4.) (c) EPA Health Advisory ___ (a) MCL (b) MCLG (d) Best Available Toxicity Data 6. Please comment briefly on your rationale for Question 5: 7. In the absence of MCLs, MCLGs, or EPA Health Advisories, guidelines/standards for carcinogens are based on which of the following? (Please rank from 1 to 4.) (a) Risk-based concentration (b) Detection limit (c) Feasibility/cost (d) Other (specify) 8. If risk-based concentrations are used, how are they derived? (Explain briefly.) 9. Please provide the risk-based concentrations, or risk levels corresponding to various exposure periods (short-term, chronic, lifetime), if applicable. IO. If cancer potency factors are calculated by your office, which extrapolation models are used? 11. Does your office follow EPA weight-of-evidence classification scheme for establishing guidelines/standards for carcinogens? (If not, explain briefly.) 12. How are class C carcinogens treated? (Explain briefly.) 13. Is any distinction made for mechanism of carcinogenicity (example: epigenetic, genotoxic, site specific)? 14. For noncarcinogens. guidelines/standards are based on which of the following? (Please rank from 1 to 4.) (a) MCL (b) MCLG (c) EPA Health Advisory ___ (d) Best available toxicity data 15. In the absence of MCLs, MCLGs, or EPA Health Advisories, guidelines/standards for noncarcinogens are based on which of the following? (Please rank from 1 to 5.) (b) AD1 (c) NOAEL (d) LOAEL (4 RfD ~~~

(e) Other (please specify) 16. In deriving guidelines/standards for noncarcinogens, which of the following uncertainty factors are considered? (Please specify applicable factors.) (a) Intraspecies variability ___ (f) Mutagenicity (b) Interspecies variability ___ (g) Teratogenicity (c) Dose-route extrapolation ___ (h) Carcinogenicity ~ (d) Less-than-lifetime data ___ (i) Adequacy of database ___ (e) Other-than-NOAEL data ___ (j ) Other factors 17. If you rely principally on the best available toxicity data for deriving guidelines/ standards, what is the source of this data? (Please rank from I to 6.) (a) Reference texts ~ (b) Journals ~ (c> Federal Register ___ (d) Toxnet database (e) IRIS database (f) Other

34

PAULL ET AL.

18. Is there a peer review process for the guidelines/standards office? (If yes, explain briefly.)

developed by your

19. Please indicate the name, address, and telephone of the key contact individual in your state for providing additional information regarding the derivation of drinking water guidelines/standards. 20. Please provide any additional comments you may have on your state’s methodology for deriving drinking water guidelines/standards.

Please forward your completed questionnaire

to:

Jeffrey M. Paull, Program Manager Environmental Toxicology and Risk Assessment Division Maryland Department of the Environment 2500 Broening Highway Baltimore, MD 2 1224 ACKNOWLEDGMENTS The authors acknowledge the members of the Toxicology/Risk Assessment Subcommittee of the FederalState Toxicology and Regulatory Alliance Committee listed below for their assistance in the design and distribution of the survey questionnaire and for their helpful comments and suggestions. We also express our thanks to the staff at the USEPA Office of Drinking Water for their assistance regarding EPA risk assessment methodologies and regulatory procedures, and for their thoughtful comments on the manuscript.

Federal-State

Toxicology and Regulatory Alliance Committee

Toxicology/Risk Bill Bress Vinh Cam Robert Cantilli Clara Chow Millicent Edison Lubow Jowa Bela Matyas Karen Martin David Nash Edward Ohanian Jeffrey Paul1 Kit-pal Sidhu

(FSTRAC)

Assessment Subcommittee

Vermont Department of Health USEPA, Region 2 FSTRA Coordinator, USEPA Office of Drinking Water USEPA, Region 1 New Mexico Health and Environmental Department New Jersey Department of Environmental Protection Rhode Island Department of Health Massachusetts Department of Environmental Protection Washington Department of Health USEPA Office of Drinking Water Maryland Department of the Environment Michigan Department of Public Health

STATES’ METHODOLOGIES

FOR DRINKING

WATER GUIDELINES

35

REFERENCES

Federal-State Toxicology and Regulatory Alliance Committee (FSTRAC), Chemical Communications Subcommittee (1990). Federal Drinking Water Standards and Guidelines. American Waterworks Assoc., Washington, DC. KRIETZMAN, S. J., DUPUY, C. J., MCGEORGE, L., AND MINTZ, B. (1988). National Survey of State Drinking Wafer Standards for Organic Contaminants. Federal-State Toxicology and Regulatory Alliance Committee, Chemical Communications Subcommittee. National Governors Association (January 19, 1990). Assessment Q Establishing a State Environmental Health Studies Clearinghouse. National Governors Association Center for Policy Research, Washington. DC. Office of Technology Assessment ( 1984). Protecting the Nation’s Groundwater from Contamination, Vol. I, OTA-0-233, pp. 5-42. U.S. Congress, Office of Technology Assessment, Washington, DC. SIDHLJ,K. S. (I 989). Computer databases for carcinogenicity and risk assessment. Environ. Toxicol. Chem. 8, 1217-1221. U.S. Congress. (I 986). Safe Drinking Wafer Act. Public Law 99-339. U.S. Govt. Printing Office, Washington, DC. U.S. General Accounting Office. (1987). Air Pollution: States Assigned a Major Role in EPA’s Air Toxic Sfrategy, p. 3. USGAO Resources, Community, and Economic Development Division, Washington, DC. U.S. Environmental Protection Agency ( 1987). Federal-State Toxicology and Regulatory Alliance Committee: FSTRAC Manual. USEPA, Criteria and Standards Division, Office of Drinking Water, Washington, DC. U.S. Environmental Protection Agency (Nov. 13, 1985). National Primary Drinking Water Regulations; Synthetic Organic Chemicals Inorganic Chemicals and Microorganisms. Proposed Rule. Fed. Regist. 50(2 l9), 46936-4702 1. U.S. Environmental Protection Agency (July 8, 1987). National Primary and Secondary Drinking Water Regulations; Synthetic Organic Chemicals. Final Rule. Fed. Regisf. 52( 130), 25690-257 17. U.S. Environmental Protection Agency (May 22, 1989). National Primary and Secondary Drinking Water Regulations. Proposed Rule. Fed. Regist. 54(97), 22062-22 160. WARE, G. (1988). Introduction. Rev. Environ. Contam. Toxicol. 104, l-8. WEXLER. P.. AND VASTA, B. (1990). IRIS to Join TOXNET. NLM Tech. Bull. No. 249. p. 15.