Document LKOm0keKgxVaJ1erk88GXjYM3

A Division of The Society of The Plastics Industry, Inc. October 4, 1988 TO: The VI Health, Safety & Environment Committee Enclosed is a copy of the Vinyl Institute statement filed on September 28th with the California Department of Health Services on the proposed MCL for vinyl chloride. Also enclosed are copies of statements by the Uni-Bell PVC Pipe Association and the National Sanitation Foundation. MNS/pmb enclosures JJjLA^CUXlU Meredith N. Scheck Assistant Director CTL019787 Wayne Interchange Plaza II 155 Route 46 West Wayne, NJ 07470 (201) 890-9299 LAW OFFICES Keller and Heckman uobcam c CLLC* N MCCH"*N Ch*v(S m mcCman WIllUM H 0*4"CSANI, JA MALCOCM 0 MCAMTHga WAYNlV lkAC rtcNcce joncs MAftTlN W *CCOVtCI JOHN S CLO*CO CaOcC C ha**i$ miCmaCL r homoni MAAK FOX CVCNS jon s ouaccx CTC* L ( k C*uz CMAIfTlNC A MCAOMCA SmialCy 9 ryjiMOTO LAWMCNCC A HAkMIN aalam a Simmons ACTCA a sussca c oouolas jaaactt IMCILA A Ml LL AA JOHN AlCHAAOS MCLVIM 9 OAOICM FACOCAICA J DAr. V ** MAACI C OACCNSTClN* SUSAN J AlSNCA AATAlCK U MUAO** S CAAlO TAUTFCST OAVIO M JCTT MAUACCN A O'CONNtLt** NINA M SlNSTCIN MAAK A SICVCAS*** MAAr CHAMICA) OAANOT C FAAMKMN KOONTZ** CATHCAINC N AOCNC*** AOtCAT jcffact **" AlCHAAO A. JAF FI* KAI9 AN N C MONTCITN** SUSAN A SAiCACAO AOMITTCO IN ACNNSTLVANIA ONLY `AOMITYIO IN VIAOlNIA ONLY AOMITTCO IN MAAYLANO ONLY AOMITTCO IN COLOAAOO ONLY 1150 17" STREET. N.W. SUITE lOOO WASHINGTON. D.C. 20000 (202) 050-5000 September 27, 1988 Office of Regulations Department of Health Services 714 P Street, Room 1000 Post Office Box 942732 Sacramento, California 94234-7320 SC'CNTirtc STAFF OANICL S OIXLCA OUAWAAO f dooocn CNAALCS V BACOCA BOBCAT A MATNCWS TCLCCOmmuniCATIOnS CNOINCCB CNAALCS F TUBNCB TCLCX AS BSSSl TCLCCOAiCB IZOII 2B TB2 CablCaooacss-aclman- WBITCB S OlBtCT OUL NUMBfB (202) 956-5641 Re: Proposed Rulemaking on Maximum Contaminant Levels for Drinking Water Dear Sir or Madam: Enclosed are comments submitted on behalf of the Vinyl Institute, a division of the Society of the Plastics Industry, Inc. (SPI), on the California Department of Health Services' proposed rulemaking on Maximum Contaminant Levels (MCLs) in water for various organic chemicals, including vinyl chloride. The Vinyl Institute opposes DHS's proposed MCL for vinyl chlo ride as unnecessarily stringent and analytically problematic. We urge that California adopt a vinyl chloride MCL consistent with the U.S. Environmental Protection Agency's level of 0.002 mg/1. We thank you for your consideration of our comments. Should you have any questions concerning the comments, please do not hesitate to contact us. Sincerely, Peter L. de la Cruz Enclosures cc: W. C. Holbrook Robert D. Luss, Esq. Meredith M. Scheck Larry Thomas (w/o enc.) Lewis R. Freeman, Jr. (w/o enc.) Robert W. Sherman CTL019788 Before the CALIFORNIA DEPARTMENT OF HEALTH SERVICES Maximum Contaminant Levels Proposed Rulemaking R-38-88 (June 23, 1988) _________ ) ) ) ) ) ) COMMENTS OF THE VINYL INSTITUTE, A Division of The Society of the Plastics Industry, Inc. Vinyl Institute Wayne Interchange Plaza II 155 Route 46 West Wayne, New Jersey 07470 September 28, 1988 CTL019789 VINYL INSTITUTE COMMENTS I. INTRODUCTION AND STATEMENT OF POSITION The Vinyl Institute appreciates the opportunity to comment on the California Department of Health Services' (CDHS) proposed maximum contaminant levels rulemaking. Cal. R-38-88 (June 23, 1988). The Vinyl Institute, a Division of The Society of the Plastics Industry, Inc. (SPI) ^ represents the major producers of vinyl chloride and polyvinyl chloride (PVC). Members of the Vinyl Institute account for approximately 82% of the domestic production of vinyl chloride and 85% of the domestic production of polyvinyl chloride. While the Vinyl Institute does not oppose CDHS's proposed rulemaking in its entirety, we do have serious concerns with the maximum contaminant level (MCL) proposed for vinyl chloride. For the reasons set forth below, the vinyl chloride MCL should not be finalized as proposed. California has assumed primary enforcement responsi bility for its public water systems under the Safe Drinking Water Act.--^ On June 23, 1988, the California Department of 1/ The Society of the Plastics Industry, Inc. is a trade or ganization of more than 2,000 members representing all segments of the plastics industry in the United States. SPI's operating units and committees are composed of resin manufacturers, distributors, machinery manufacturers, plastic processors, mold makers, and other industry-related companies and individuals. 2/ 42 U.S.C. 300g-2 (1982, as amended, 1986). CTL019790 2 Health Services proposed maximum contaminant levels in drinking water for nine compounds, including vinyl chloride. The Department has proposed a level of 0.0005 mg/1, which it contends is the current limit of detection, as the maximum permissible level of vinyl chloride in drinking water. The California Department of Health Services' proposed rulemaking followed the U.S. Environmental Protection Agency's (EPA) July 1987 promulgation of final maximum contaminant levels for eight volatile organic compounds by nearly a year.--7 Among the eight compounds regulated by EPA in that 1987 rulemaking is vinyl chloride, for which EPA set an MCL of 0.002 mg/1. CDHS's proposed vinyl chloride MCL of 0.0005 mg/1 is far more stringent than that recommended by EPA. CDHS's rationale in lowering the maximum contaminant level to equal the limit of detection is based in CDHS's unfounded criticism that EPA's risk assessment underestimates the health risks of vinyl chloride. CDHS has asserted that because EPA did not consider saturable pharmacokinetics in its risk assessment of vinyl chloride, EPA's analysis is "limited in its usefulness." However, the data actually relied upon by California in developing its maximum contaminant level for vinyl chloride evidences a one-to-one ratio of dose to incidence of cancer in the low dose region, i .e., the dose-response relationship is 3/ See 52 Fed. Reg. 25690 (1987) for federal standards. CTL019791 3 linear at low dose. Consequently, at levels which are important for low dose extrapolation, there is no evidence of saturation effects. California's Department of Health Services appears to have drawn conclusions concerning vinyl chloride which have no basis in fact. We question whether California's standard is supported by available scientific evidence on the potential health effects associated with vinyl chloride. We also question whether technological issues were adequately considered by the Department of Health Services. Since EPA's administrative record supporting its promulgation of maximum contaminant levels is based upon conservative risk assessment methodology and feasible analytical capabilities, there is no rational basis for the Department of Health Services to establish maximum contaminant levels below EPA's safe recommendations. To do so because it is occasionally possible using procedures that have not been shown to be consistently reproducible, is likewise unjustified. We recommend that California's Department of Health Services adopt a vinyl chloride MCL consistent with EPA's 0.002 mg/1 level. II. EPA'S REGULATION OF VINYL CHLORIDE EPA's final rule promulgating a maximum contaminant level for vinyl chloride of 0.002 mg/1 requires analysis of CTL019792 4 vinyl chloride in groundwater systems only where one or more of the following two-carbon organic compounds are also detected: trichloroethylene, tetrachloroethylene, 1,2-dichloroethane, 1,1,1-trichloroethane , cis-1,2-dichloroethylene, trans-1,2-dichloroethylene, or 1,1-dichloroethylene.--^ If one of the above compounds is present, vinyl chloride analysis must occur at each distribution or entry point;--^ if no vinyl chloride is detected at a 0.002 mg/1 level, a state need only monitor the location for vinyl chloride once every three years.--^ Vinyl chloride analysis of surface water systems is left to the discretion of each state.--^ The former requirement, applicable to ground water systems, is based on the rationale that any detectable quantity of vinyl chloride in drinking water likely results from the degradation of trichloroethylene and similar solvents, rather than the actual disposal of vinyl chloride or polyvinyl chloride. As EPA stated in its proposed rulemaking in Novem ber, 1985, "EPA has concluded that the most likely explanation for vinyl chloride detection in groundwaters is from iji situ 4/ 52 Fed. Reg. 25690, 25703. 5/ Id^ at 25713. 6/ 52 Fed. Reg. at 25713. 7/ Id. CTL019793 5 transformation."--^ As to a state's discretion in analyzing surface water for vinyl chloride, EPA stated, "[N]o require ments would be set for vinyl chloride monitoring in surface water systems because vinyl chloride is not expected to be found due to its high volatility."--9 / While only one community water system well in Cali fornia currently contains vinyl chloride at a level higher than CDHS's proposed MCL,--/ the ramifications of California's vinyl chloride standard are far-reaching. Under California's proposed rulemaking, all community water systems will be required to monitor for vinyl chloride, regardless of the tech nical feasibility of analyzing for the chemical at such a low level, and regardless of the presence of halocarbon solvents indicating the possible presence of vinyl chloride. Community water systems in California are required to test for vinyl chloride, and other organic chemicals, pursuant to the following schedule. All systems are considered by the state to be initially vulnerable. Following quarterly monitor- 8/ 50 Fed. Reg. 46902, 46919-46920. 9/ Id. 10/ According to CDHS, the well is used only for industrial purposes; there is currently "no known exposure [in drinking water] to vinyl chloride," California Department of Health Services, Proposed Rulemaking for Maximum Contaminant Levels, Initial Statement of Reasons pp. 1 and 8. CTL019794 6 ing during the first year, a system must monitor thereafter every other year. This monitoring schedule is far more onerous than that required by EPA, and is simply unnecessary given the unlikelihood of vinyl chloride's presence, and given the negli gible risk factors presented by the compound at the extra ordinarily low maximum contaminant level designated by EPA. III. BPA DETERMINATION OF MAXIMUM CONTAMINANT LEVELS In its initial statement of reasons for lowering vinyl chloride's MCL, CDHS identified what it perceived as inadequa cies in EPA's risk assessment for the organic compound. Never theless, we submit that EPA's risk assessment procedures are sufficiently "conservative" to protect the public safety even if one were to assume that the CDHS criticism is justified. Carcinogenic risk assessment is a complicated science, at best. Any number of factors enter into the analysis of a particular chemical's human cancer effect; whether one relies on epidemiological or long- or short-term animal studies, or a combination of the three, may significantly alter one's con clusions. Equally competent scientists differ in their inter pretation of pathology slides. They must also choose from among several alternative dose response models for projecting low-dose risks because of a fundamental absence of definite CTL019795 7 data on cancer induction and development mechanisms. And a risk assessment may well be hampered by a simple lack of or gap in sound available data. In response to the inherent uncertainty found in any study of potential carcinogens, carcinogenic risk assessment procedures adopt "conservatisms" to provide ample margins of safety. For example, the initial stage of EPA's MCL develop ment, the recommendation that maximum contaminant level goals be set at zero, is rooted in the notion that a so-called "threshold" for carcinogens does not exist, or if it does, cannot be demonstrated by current science.--^ Proponents of such a view hold that the reaction of a single molecule of a carcinogen with a sensitive cell can initiate a cancer response -- that there is no threshold. Since science is unable to demonstrate unequivocally such a threshold effect for all potential carcinogens, they argue that any exposure to a carcinogen presents a finite level of risk. However, scientific evidence does exist that, at certain exposure levels, no carcinogenic effect can be observed. Despite this, EPA assumed the "worst" scenario in drawing the conclusion that a "no effect" level does not exist in order to ensure human health is properly safeguarded by the final MCLs. 11/ 50 Fed. Reg. 46880, 46884 (1985). CTL019796 8 EPA builds into its risk assessments other safety fac tors. EPA not only factors into its analysis the increased incidence of malignant tumors initiated by the dosage, but also considers increased occurrence of benign tumors.--12/ According to EPA, such evidence is an indication, although limited, of potential carcinogenicity, and therefore should be considered.--13'/ To develop its MCLs, EPA used the most conservative of extrapolation models, the linearized multi-stage model.--14/ In its general guidelines to assess human cancer risk, EPA notes that the choice of a particular low-dose extrapolation model can have a "pronounced influence on the estimated low-dose risk."--^ The multi-stage model recognizes that a developing tumor goes through several stages, all of which can be affected by the particular carcinogen before it is clinically detect able.--^ Because the multi-stage model is linear at the low- dose region, its risk estimates are generally regarded as very conservative.--17'/ One further safety factor is EPA's use of a 12/ 51 Fed. Reg. at 34002 13/ 50 Fed. Reg. at 46884 14/ 50 Fed. Reg. at 10438 15/ Id. at 10441 16/ Id. at 10438 17/ Id. CTL019797 9 95% confidence level for resulting carcinogenic risk. This very cautious statistical measure results in a significantly higher calculated risk of cancer incidence than observed in the animals and thus errs on the side of safety. Finally, EPA relates the experimental animal exposures to equivalent human exposure by utilizing body surface areas instead of weight -- another approximately ten-fold safety factor. The extreme conservatism created by these risk modeling assumptions may be demonstrated in the following example. Assume fifty laboratory animals (rats) are exposed to a puta tive carcinogen at a dosage of 1,000 parts per million (ppm); suppose further that no tumors of any kind are observed result ing from that dosage. A 95% confidence level will assume that 5.8% of those rats actually contracted cancer from that expo sure to the 1,000 ppm dosage. The 5.8% value is then used to calculate the one in a million risk of cancer, resulting in a safe dosage of approximately 2.9 parts per billion (ppb). In other words, the observation that none of the fifty rats showed any incidence of cancer at a dose of 1000 ppm (0.1%) leads to a "safe dose" of only 2.9 ppb, a safety factor of more than 340,000. Risk assessment values were never intended to measure actual cancers (e ,g. , one in one million) produced by exposure CTL019798 10 to a substance. Rather, at the calculated dose, the risk may well be zero but is virtually certain not to exceed 1 in one million. Despite EPA's extremely cautious approach, California rejected EPA's methodology and established a vinyl chloride MCL at an unnecessarily low level. IV. CALIFORNIA RISK ASSESSMENT As discussed above, CDHS dismissed EPA's risk assess ment of vinyl chloride for EPA's alleged failure to consider saturable pharmacokinetics in assessing its testing results. Pharmacokinetics is generally an appropriate factor to consider for certain substances at certain exposure levels. However, given the test results of the low dose experiments relied on by CDHS in its risk assessment of vinyl chloride, consideration of pharmacokinetics is unjustified here. In the studies referenced by CDHS in its risk assessment -- including the 1984 Maltoni study, the 1981 Feron study, and the 1983 Til study -- incidence of cancer is directly proportional to the low doses of vinyl chloride. CDHS did not recognize this proportional relationship. CTL019799 11 V. EPA'S STANDARDS PROVIDE FOR A CONSISTENT LEVEL OF DATA QUALITY FROM A DIVERSE GROUP OF LABORATORIES In light of EPA's extreme conservatisms in setting health effects levels, it is appropriate to recognize that practical considerations can enter into the setting of enforce ment levels without adverse health consequences. In the present case, it is necessary to recognize the inherent vari ability experienced by good laboratories conducting routine, periodic tests for substances present at very low concentra tions. In this light, the MCLs set by EPA are those that can regularly be achieved by good laboratories working under good procedures; the California limit requires extreme analytical care that cannot be achieved regularly -- and that is not necessary to protect the public health. In November 1985, EPA introduced the concept of prac tical quantitation level (PQL) , the lowe-st level that can "reliably be achieved" within specified parameters of precision during "routine" laboratory conditions.--' PQL is a measure ment concentration independent of time and independent of specific laboratory,--19/ and is itself based on another estimated measurement, the method detection limit (MDL).--20/ An 18/ 50 Fed. Reg. 46902, 46906 (1985). 19/ 52 Fed..Reg. at 25699. 20/ 52 Fed. Reg. at 46906. CTL019800 12 MDL is the minimum concentration of a substance that can be measured with 99% assurance that the true value is greater than zero.--^ While laboratories can achieve MDLs of 0.0005 mg/1 or lower,--^ these detection levels "are not necessarily reproduc ible over time in a given laboratory even when the same analytical procedure, instrumentation, and sample matrix are used."--' Because MDLs result from laboratory performances under ideal conditions, PQLs are better described as models of what normal good laboratory practices can achieve. Usually set at five to ten times the MDL, EPA characterizes the PQL as a 'relatively stringent target" for everyday performance. 24/ While EPA recognizes that a number of laboratories have reported data at levels below EPA's PQLs, EPA has explicitly found that the data is insufficient upon which to base regulatory standards.--^ The PQLs developed by the EPA for vinyl chloride equal 0.002 mg/1 (2 ug/1) ^ California not only has established a maximum contaminant level far below this level and far below 21/ Id. 22/ Id. 23/ 50 Fed. Reg. at 46906 24/ 52 Fed. Reg. at 25700 25/ Id. 26/ Id. CTL019801 13 what most laboratories can identify, it has set an MCL equal to the method detection limit for vinyl chloride -- 0.0005 mg/1. In setting this level, CDHS has ignored the fact that even good laboratories are incapable of consistently performing at such a level. In its proposed maximum contaminant level rulemaking in November 1985, EPA stated that the "effectiveness of the pro posed regulations is dependent upon the ability of analytical laboratories to produce reliable data at relatively low levels for these contaminants."--27/ EPA thus established performance requirements for laboratories analyzing compliance samples for the various compounds; the final acceptance limits for labora tories for vinyl chloride analysis were based on a i 40% of the true value of vinyl chloride levels found in a sample. EPA found that, for vinyl chloride, only 8 out of 18 laboratories (or 44% of the laboratories) were able to analyze three varying levels within the i 40% acceptance limits. When the lowest concentration (0.0015 mg/1, far greater than California's pro posed MCL of 0.0005 mg/1) was not considered, the laboratories found within the acceptance limits increased to 13 of 18 (or 72% of the laboratories). In other words, analyzing samples at the 0.0015 mg/1 level posed significant problems in accuracy for the laboratories. A second study by EPA confirmed these 27/ 50 Fed. Reg. at 46970. CTL019802 14 initial results. Twenty-nine out of 44 laboratories (or 66% of the laboratories) analyzed vinyl chloride within the 40% limits, when the lowest level of 0.0015 mg/1 was not considered. In light of these figures, California's proposed maximum contaminant level for vinyl chloride stands out as an unachievable regulatory standard for even highly competent laboratories. The proposed MCL will inevitably lead to arbitrary and confusing test results rendering the regulation itself invalid as it applies to vinyl chloride. The rationale offered by the EPA in rejecting lower standards is equally applicable to California's standards; unlike California's MCL for vinyl chloride, the federal standard provides for a consistent level of data quality from a diverse group of laboratories and hence, an enforceable standard. VI. THE RATIONAL SANITATION FOUNDATION Finally, it is appropriate to note that the plastics pipe industry--28'/ has long participated in a voluntary standards program in which its products are evaluated by an independent 28/ Plastics materials used for potable water systems include polyethylene (PE), polyvinyl chloride (PVC) and polybutylene (PB). Plastic pipe systems are used for water mains, service lines, and interior plumbing applications. CTL019803 15 organization, the National Sanitation Foundation (NSF), to assure product integrity for purposes of transporting drinking water. As innovative products, plastic materials used for pot able water systems have been subject to intense scrutiny since first placed on the market. Industry determined early that health and safety-related standards would be appropriate to gain public acceptance of these new materials. Consequently, an extensive and thorough program was undertaken, first with Battelle Institute in Columbus, Ohio, and then with the National Sanitation Foundation, both nationally recognized and respected safety organizations, to establish health and safety criteria for plastic pipe and to provide a means of enforcing such criteria. NSF Standard No. 14 on plastic pipe was originally adopted in 1965. The Standard covers not only plastic pipe itself, but also fittings, valves, tanks, jointing materials and appurtenances used for potable water. Thus, the NSF program represents a comprehensive pre-screening of plastic pipe systems used for potable water applications. Although NSF Standards are voluntary, most manufactur ers' products are certified by NSF to assure that their plastic pipe will meet the requirements of various model codes which are adopted by most states. As a practical matter, NSF approval has' become a market-driven requirement. CTL019804 16 To obtain NSF approval for a given type of plastic pipe, a manufacturer is required to file complete chemical formulations for all products and submit compounds to NSF for testing. New testing must be conducted any time the formula tion is modified in any way. If a basic constituent of a new ingredient is not on FDA's list of materials cleared for food contact use, an animal feeding study and Ames test (a test for mutagenicity and, indirectly, carcinogenicity) may be required to determine if the substance poses a potential health problem. The National Sanitation Foundation's evaluation program assures the safety and quality of plastic pipes, including those constructed from polyvinyl chloride. NSF's certification process, together with EPA's maximum contaminant level for vinyl chloride, helps assure the safety of California's drinking water supply. VII. COHCLOSIOH In summary, the Vinyl Institute strongly urges the California Department of Health Services to adopt a vinyl chloride MCL consistent with the U.S. Environmental Protection Agency's maximum contaminant level, issued July 7, 1987. EPA has carefully considered all currently available scientific data on vinyl chloride, and has carefully weighed all pertinent CTL019805 17 factors in making its recommendations. Any lower levels will only unjustifiably burden California's communities with no concomitant gain in public safety. Respectfully submitted, eter L. de la Cruz Counsel OF COUNSEL: PETER L. DE LA CRUZ CATHERINE N. ROEMER Keller and Heckman 1150 17th Street, N.W. Suite 1000 Washington, D.C. 20036 (202) 956-5600 CTL0198Q6