1 2 3 4 5 6 7 8 9 Standing Operating Procedures 10 11 12 13 14 ofthe 15 16 17 18 19 20 National Advisory Committee 21 22 on 23 24 Acute Exposure Guideline Levels 2s for Hazardous Substances 26 27 28 29 30 31 32 33 34 35 36 37 Version 08-02 38 June 30, 2000 ------- 2 PREFACE 3 4 The National Advisory Committee for Acute Exposure Guideline Levels for Hazardous 5 Substances (NAC/AEGL Committee) was established to develop scientifically credible short-term 6 exposure limits for approximately 400 to 500 acutely toxic substances. These short-term exposure limits, 7 referred to as Acute Exposure Guideline Levels, or AEGLs, are essential for emergency planning, 8 response, and prevention of accidental releases of chemical substances. Further, it is important that the 9 values developed be scientifically credible so that effective planning, response, and prevention can be 10 accomplished. 11 12 To insure scientific credibility, five major elements have been integrated into the AEGL 13 development process. These include adherence to the 1993a National Resource Council with changes or 14 additions as set forth in the Standing Operating Procedures Manual (SOP Manual), U. S. National 15 Academy of Sciences (NRC-NAS) guidelines for developing short-term exposure limits, a 16 comprehensive search and review of relevant data and information from both published and unpublished 17 sources, the extensive evaluation of the data and the development of AEGLs by a committee of scientific 18 and technical experts from both the public and private sectors, a multi-tiered peer review process 19 culminating with final review and concurrence by the U. S. National Academy of Sciences (NAS), and, 20 the use of scientifically acceptable processes and methodologies to insure consistent and scientifically 21 credible AEGL values. 22 23 With the recent participation of certain member-countries of the Organization for Economic and 24 Cooperation Development (OECD), it is anticipated that the AEGL program will be expanded to include 25 the international community. This should result in increased scientific and technical support, a broader 26 scope of the review process, and an even greater assurance of scientifically credible AEGL values. 27 28 This Standing Operating Procedures Manual (SOP Manual) represents the documentation by the 29 NAC/AEGL Committee's SOP Workgroup of those procedures, methodologies, criteria and other 30 guidelines employed by the NAC/AEGL Committee in the development of the AEGL values. The 31 information contained herein is based on the guidance provided by the NAS in its 1993 publication 32 Guidelines for Developing Community Emergency Exposure Levels for Hazardous Substances (NRC, 33 1993a) This manual contains additions and further details and clarification of specific procedures, 34 methodologies, criteria, and guidelines interpreted from the NAS guidelines that have been determined 35 by the NAC/AEGL Committee to be a necessary supplement to the 1993a NAS guidelines Procedures 36 and methodologies included in this manual have been reviewed by the NAC/AEGL Committee, 37 numerous OECD member countries, and have received a review and concurrence by the U S. National 38 Academy of Sciences. New or modified procedures and methodologies that are developed and adopted 39 by the NAC/AEGL Committee are classified as "Proposed." Such procedures and methodologies will, 40 from time to time, be submitted to the NAS for review and concurrence. Upon concurrence by the NAS, 41 they will be considered final and will serve as a supplement to the 1993 NRC-NAS guidelines and to the 42 2000 SOP guidance manual 43 44 It is believed that adherence to a rigorous AEGL development process in general, and the use of 45 scientifically sound procedures and methodologies in particular, will provide the most scientifically 46 credible exposure levels that are reasonably possible to achieve. This document is considered a "living 47 document" and the various procedures and methodologies, including those classified as "Final", are ------- 1 subject to change as deemed necessary by the NAC/AEGL Committee and the U. S. National Academy 2 of Sciences (NAS Subcommittee on Acute Exposure Guideline Levels, Committee on Toxicology, 3 National Research Council). As new data become available and new scientific procedures and 4 methodologies become accepted by a majority of the relevant scientific community, the NAC/AEGL 5 Committee and the National Academy of Sciences, they will be integrated into the AEGL development 6 process and the SOP Manual. With this approach, both the scientific credibility of the AEGL values and 7 the reduction in nsk to the general population will be insured 11 ------- 1 TABLE OF CONTENTS 2 3 PREFACE i 4 5 1. OVERVIEW OF CURRENT AEGL PROGRAM AND NAC/AEGL COMMITTEE 1 6 HISTORY 1 7 PURPOSE AND OBJECTIVES OF THE AEGL PROGRAM AND THE NAC/AEGL 8 COMMITTEE 3 9 COMMITTEE MEMBERSHIP AND ORGANIZATIONAL STRUCTURE 4 10 SELECTION OF CHEMICALS FOR AEGL DEVELOPMENT 5 11 SCIENTIFIC CREDIBILITY OF AEGLS 6 12 THE AEGL DEVELOPMENT AND PEER REVIEW PROCESS 7 13 OPERATION OF THE COMMITTEE 9 14 VALUE OF A COLLABORATIVE EFFORT IN THE AEGL PROGRAM 10 15 APPLICATIONS OF THE AEGL VALUES 12 16 17 2. DERIVATION OF AEGL VALUES 15 18 2.1 DEFINITIONS OF AEGL-1, AEGL-2 AND AEGL-3 15 19 PREFACE 15 20 2.2 EMPIRICAL TOXICOLOGIC ENDPOINTS, AND METHODS FOR 21 DETERMINING EXPOSURE CONCENTRATIONS USED TO DERIVE 22 AEGL-1,2, AND 3 LEVELS 17 23 2.2.1 SELECTION OF THE HIGHEST EXPOSURE LEVEL WHERE THE 24 EFFECTS USED TO DEFINE AN AEGL LEVEL WERE NOT 25 OBSERVED 17 26 2.2.2 SELECTION OF HEALTH EFFECTS ENDPOINTS FOR AEGL-1, 27 AEGL-2, AND AEGL-3 20 28 2.2.2.1 AEGL-1 Endpomts 21 29 2.2.2.1.1 No Value Established - AEGL-1 Exceeds AEGL-2 . 22 30 2.2.2.1.2 No Value Established - Insufficient Data 22 31 2.2.2.1.3 Highest Experimental Exposure Without an AEGL-1 32 Effect 22 33 2.2.2.1.4 Effect Level for a Response 22 34 2.2.2.2 AEGL-2 Endpoints 23 35 2.2.2.2.1 Highest Experimental Exposure Without an AEGL-2 36 Effect 23 37 2.2.2.2.2 Effect Level for a Toxic Response Which was Not 38 Incapacitating or Not Irreversible 23 39 2 2.2.2.3 A Fraction of the AEGL-3 Level 24 40 2.2.2.3 AEGL-3 Endpomts 24 41 2.2.2.3.1 Highest Exposure Level Which Does Not Cause Lethality 42 - Experimentally Observed Threshold (AEGL-3 NOEL) 43 24 ------- 1 2.2.2.3.2 Highest Exposure Level Which Does Not Cause Lethality 2 - Estimated Lethality Threshold -1/3 of the LC50 24 3 2.2.2.3.3 Highest Exposure Level Which Does Not Cause Lethality 4 - Benchmark Exposure Calculation of the 5 % and 1% 5 Response 25 6 2.2.2.3.4 Effect Level for a Response 25 7 2.3 GUIDELINES/CRITERIA FOR THE SEARCH STRATEGY, EVALUATION, 8 SELECTION AND DOCUMENTATION OF KEY DATA AND SUPPORTING 9 DATA USED FOR THE DERIVATION OF AEGL VALUES 27 10 2.3.1 Search Strategy 27 11 2.3.2 Evaluation, Selection and Documentation of Key and Supporting Data ..31 12 Elements for the Evaluation of Key and Supporting Data and Studies .. 36 13 2.3.3 Elements for Discussion on Data Adequacy and Research Needs 38 14 2.4 DOSIMETRY CORRECTIONS FROM ANIMAL TO HUMAN EXPOSURES .. 39 15 2.4.1 Discussion of Potential Dosimetry Correction Methodologies for Gases 16 39 17 2.4.1.1 The Respiratory System as a Target Organ 39 18 2.4.1.2 Systemic Toxicity 41 19 2.4.2 Current Approach of the NAC/AEGL Committee to Dosimetry Corrections 20 42 21 2.5 GUIDELINES/CRITERIA FOR SELECTION OF UNCERTAINTY FACTORS TO 22 ADDRESS THE VARIABILITY BETWEEN ANIMALS AND HUMANS AND 23 WITHIN THE HUMAN POPULATION 44 24 2.5.1 Introduction 44 25 2.5.2 Background 45 26 2.5.3 Considerations and Approaches to the Selection of Uncertainty Factors for 27 Developing AEGLs 48 28 2.5.3.1 Interspecies Uncertainty Factors - Use in the Development of 29 AEGL Values - Discussion 50 30 2.5.3.2 Interspecies Uncertainty Factors - NAC/AEGL Committee 31 Guidelines 51 32 2.5.3.2.1 Most Appropriate Species Used 52 33 2.5.3.2.2 Most Sensitive Species Not Used 52 34 2.5.3.2.3 Mechanism of Action is Unlikely to Differ Among 35 Species 52 36 2.5.3.2.4 Mechanism of Action is Unknown 53 37 2.5.3.2.5 Variability in Response Between Species 53 38 2.5.3.2.6 Humans More Sensitive than Animals 53 39 2.5.3.2.7 Use of an Uncertainty Factor of 10 54 40 2.5.3.2.8 A Selected Uncertainty Factor Applied to Animal Data 41 Would Drive the AEGL-2 or -3 Level to a Value Which 42 Humans can Tolerate without Lethal or Senous Adverse 43 Effects 54 ------- 1 2.5.3.2.9 A multiple exposure study was used to set the level... 54 2 2.5.3.3 Intraspecies Uncertainty Factors - Use m the Development of 3 AEGL Values - Discussion 56 4 2.5.3.3.1 Range of Susceptibility 61 5 2.5.3.3.2 Selection of Intraspecies Uncertainty Factors 64 6 2.5.3.3.3 Distinguishing Susceptible and Hypersusceptible 7 Individuals 64 8 2.5.3.3.4 Estimating the Range of Variability in a Human 9 Population 65 10 2.5.3.4 Intraspecies Uncertainty Factors - NAC/AEGL Guidelines 68 11 2.5.3.4.1 Toxic Effect is Less Severe than Defined for the AEGL 12 Tier 68 13 2.5.3.4.2 Sensitive/Naive Individual Used 68 14 2.5.3.4.3 Age/Life Stage/Condition Differences 69 15 2.5.3.4.4 Response by Normal and Sensitive Individuals to 16 Chemical Exposure is Unlikely to Differ for Mechanistic 17 Reasons 69 18 2.5.3.4.5 Mode or Mechanism of Action is Unknown 69 19 2.5.3.4.6 Uncertainty Factors Which Result in AEGL Values That 20 Conflict with Actual Human Exposure Data 70 21 2.6 GUIDELINES/CRITERIA FOR SELECTION OF MODIFYING FACTORS .... 71 22 2.6.1 Definition 71 23 2.6.2 Use of Modifying Factors to Date in the Preparation of AEGL Values .. 71 24 2.7 GUIDELINES/CRITERIA FOR TIME SCALING 72 25 2.7.1 Overview 72 26 2.7.2 Summary of Key Publications on Time Scaling 74 27 2.7.3 Summary of the Approaches that may be Taken for Time Scaling 75 28 2.7.4 Use of Empirical Data that is Available for AEGL-Specified Exposure 29 Durations 76 30 2.7.5 Derivation of Values of n When Adequate Empirical Data are Available for 31 Other than the AEGL-Specified Exposure Durations 76 32 2.7.5.1 Selection of Appropriate Health Effect End Point for Deriving a 33 Value for n 76 34 2.7.5.2 Criteria for Adequate Empirical Data for Deriving Values of n 35 77 36 2.7.5.3 Curve Fitting and Statistical Testing of the Generated Curve . 77 37 2.75.4 Examples of NAC/AEGL Committee Derivations of Values of n 38 from Empirical Data 80 39 2.7.6 Selection of Values of n When Adequate Empirical Data are Not Available 40 to Derive Values for n 80 41 2.7.6.1 Selection of Values of n When Extrapolating from Shorter to 42 Longer Exposure Periods 82 43 2.7.6.2 Selection of Values of n When Extrapolating from Longer to 111 ------- 1 Shorter Exposure Periods 82 2 2.7.7 Special Considerations in the Time Scaling of AEGL-1 and AEGL-2 3 Values 85 4 2.7.8 Time Scaling - Guidelines for NAC/AEGL Committee Approach 86 5 2.7.8.1 Use of Empirical Data to Determine the Exposure Concentration- 6 Exposure Duration Relationship 86 7 2.7.8.2 Estimating the Concentration-Exposure Relationship using a 8 Surrogate Chemical 86 9 2.7.8.3 Estimating the Concentration-Exposure Duration Relationship 10 when Data are not Available to Derive a Value for n and 11 Supporting Data are Available 87 12 2.7.8.4 Determining Concentration-Exposure Relationships when Data 13 are not Available to Derive a Value for n and no Supporting Data 14 are Available 88 15 2.7.8.5 AEGL Exposure Values are Constant Across Time 88 16 2.8 GUIDELINES/CRITERIA FOR ADDRESSING SHORT TERM EXPOSURE 17 KNOWN AND SUSPECT CARCINOGENS 89 18 2.8.1 NRC/NAS Guidance 89 19 2.8.2 Precedents for Developing Short-Term Exposure Limits Based on 20 Carcinogenicity 90 21 2.8.3 Scientific Basis for Credible Theoretical Excess Carcinogenic Risk 22 Assessments for Single Exposures of 8 Hours or Less 91 23 2.8.4 Practical Issues of Using Quantitative, Carcinogenic Risk Assessments for 24 Developing AEGLs 93 25 2.8.5 Current Approach of the NAC/AEGL Committee to Assessing Potential 26 Single Exposure Carcinogenic Risks 94 27 2.8.5.1 Evaluation of Carcinogenicity Data 94 28 2.8.5.2 Methodology Used for Assessing the Carcinogenic Risk of a 29 Single Exposure 95 30 2.8.5.2.1 The Determination of an Adjustment Factor Dealing with 31 the Dose-Dependent Stage of Carcinogenesis 95 32 2.8.5.3 Summary of Cancer Assessment Methodology used by the 33 NAC/AEGL Committee 97 34 2.9 GUIDELINES/CRITERIA FOR MISCELLANEOUS PROCEDURES AND 35 METHODS 99 36 2.9.1 Mathematical Rounding of AEGL Values 99 37 2.9.2 Multiplication of Uncertainty Factors 99 38 39 3. FORMAT AND CONTENT OF TECHNICAL SUPPORT DOCUMENTS 100 40 3.1 FORMAT AND CONTENT OF THE TECHNICAL SUPPORT DOCUMENT 41 (TSD) 100 42 PREFACE 100 43 TABLE OF CONTENTS 100 IV ------- 1 EXECUTIVE SUMMARY 101 2 OUTLINE OF THE MAIN BODY OF THE TECHNICAL SUPPORT 3 DOCUMENT 103 4 3.2 Potential Inclusion of Graphic Descriptions of Data 107 5 6 4. CURRENT ADMINISTRATIVE PROCESSES AND PROCEDURES FOR THE 7 DEVELOPMENT OF AEGL VALUES Ill 8 4 1 COMMITTEE MEMBERSHIP AND ORGANIZATIONAL STRUCTURE 112 9 4.2 THE AEGL DEVELOPMENT AND PEER REVIEW PROCESS 113 10 4.3 OPERATION OF THE COMMITTEE 116 11 44 ROLE OF THE DIRECTOR OF THE AEGL PROGRAM 117 12 4.5 ROLE OF THE DESIGNATED FEDERAL OFFICER 118 13 4.6 ROLE OF THE NAC/AEGL COMMITTEE CHAIR 118 14 4.7 CLASSIFICATION OF THE STATUS OF AEGL VALUES 119 15 48 ROLE OF AEGL DEVELOPMENT TEAMS 120 16 4 8.1 Role of a Chemical Manager 120 17 4.8.2 Role of a Chemical Reviewer 121 18 4.8.3 Role of an Staff Scientist at the Organization which Drafts Technical 19 Support Documents 122 20 4.9 ROLE OF NAC/AEGL COMMITTEE MEMBERS 122 21 4.10 ROLE OF THE ORGANIZATION THAT DRAFTS TECHNICAL SUPPORT 22 DOCUMENTS 123 23 24 5 REFERENCES 125 25 26 APPENDIX A. NAC/AEGL PROGRAM PERSONNEL A 1 27 28 APPEND1XB. PRIORITY LISTS OF CHEMICALS B 1 29 30 APPENDIX C. DIAGRAM OF THE AEGL DEVELOPMENT PROCESS C 1 31 32 APPENDIX D. GLOSSARY - ACRONYMS, ABBREVIATIONS, AND SYMBOLS D 2 33 34 APPENDIX E. EXAMPLE OF A TABLE OF CONTENTS IN A TECHNICAL SUPPORT 35 DOCUMENT El 36 37 APPENDIX F. EXAMPLE OF AN EXECUTIVE SUMMARY IN A TECHNICAL SUPPORT 38 DOCUMENT F 1 39 40 APPENDIX G EXAMPLE OF THE DERIVATION OF AEGL VALUES APPENDIX IN A 41 TECHNICAL SUPPORT DOCUMENT G 1 42 43 APPENDIX H. EXAMPLE OF A TIME SCALING CALCULATIONS APPENDIX IN A ------- 1 TECHNICAL SUPPORT DOCUMENT HI 2 3 APPENDIX I. EXAMPLE OF A CARdNOGENICITY ASSESSMENT APPENDIX IN A 4 TECHNICAL SUPPORT DOCUMENT II 5 6 APPENDIX J. EXAMPLE OF THE DERIVATION SUMMARY APPENDK IN A 7 TECHNICAL SUPPORT DOCUMENT J 1 8 9 APPENDIX K. LIST OF EXTANT STANDARDS AND GUIDELINES IN A TECHNICAL 10 SUPPORT DOCUMENT K 1 11 12 LIST OF TABLES 13 14 TABLE 2.7-1. VALUES OF n FROM TEN BERGE ET AL. (1986) 81 15 TABLE 3.2-1 GROUPING DATA INTO CATEGORIES FOR PLOTTING 109 16 TABLE B-l. PRIORITY LIST OF CHEMICALS B 4 17 18 19 20 LIST OF FIGURES 21 22 FIGURE 1-1 HAZARD ASSESSMENT 14 23 FIGURE 2.3-1 ALLOCATION OF STUDY REPORTS DECISION TREE 35 24 FIGURE 2.7-1 EFFECTS OF VARYING n IN THE EQUATION Cn x t = k 84 25 FIGURE 3.2-1 PLOT OF CATEGORIES OF DATA 110 26 FIGURE 4.2-1 THE AEGL DEVELOPMENT PROCESS 115 27 FIGURE C-l THE AEGL DEVELOPMENT PROCESS C 1 28 29 30 31 VI ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30. 2000 1 2 1. OVERVIEW OF CURRENT AEGL PROGRAM AND 3 NAC/AEGL COMMITTEE 4 5 6 HISTORY 7 8 The concerns of EPA, other U.S. federal agencies, state and local agencies, private 9 industry and other organizations in the private sector regarding short-term exposures due to 10 chemical accidents became sharply focused following the accidental release of methyl isocyanate 11 in Bhopal, India in December of 1984. In November 1985, as part of EPA's National Strategy 12 for Toxic Air Pollutants, the agency developed the Chemical Emergency Preparedness Program. 13 This voluntary program identified a list of over 400 acutely toxic chemicals and provided this 14 information, together with interim technical guidance, for the development of emergency 15 response plans at the local community level. At that time the agency adopted the NIOSH 16 Immediately Dangerous to Life and Health (IDLH) exposure values, or an approximation of these 17 values in instances where IDLH values were not published, to serve as the initial airborne 18 concentrations of concern for each chemical. 19 20 During this same period, the U.S. Chemical Manufacturers Association (CMA) 21 developed and implemented the Community Awareness and Emergency Response (CAER) 22 program. This program encouraged chemical plant managers to assist community leaders in 23 preparing for potential accidental releases of acutely toxic chemicals. The program was intended 24 to provide local communities with information on existing chemicals and chemical processes, 25 technical expertise to assist in emergency planning, notification and response, as well as the 26 training of response personnel. 27 28 In October, 1986 as part of the reauthonzation of Superfund, Congress wrote into law an 29 emergency planning program under the Superfund Amendments and Reauthonzation Act (SARA 30 Title HI). Under this act, states were required to have emergency response plans for chemical 31 accidents developed at the local community level. The EPA subsequently adjusted the level of 32 concern values to one-tenth of the IDLH value or its equivalent as an approach to improving the 33 safety of the levels used for the general public. Since that time, the agency and other 34 organizations, including private industry, have been interested in adopting more rigorous 35 methodologies for determining values that would be deemed safe for the general public. During 36 this period, the American Industrial Hygiene Association (AIHA) established a committee, the 37 Emergency Response and Planning Guidelines (ERPG) Committee to develop ERPGs and 38 pioneered the concept of developing three different airborne concentrations for each chemical 39 that would reflect the thresholds for important health effect endpoints. The Committee was later 40 renamed the Emergency Response Planning (ERP) Committee. Although constrained by limited 41 resources, the ERP Committee has managed to develop one-hour exposure limits for more than Sop08-02 wpd Printed July 6, 2000 1 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 70 chemicals during the past 10 years. 2 3 At a workshop hosted by EPA in 1987, it was proposed by EPA that the ERP Committee 4 and scientists from federal and state agencies, as well as scientists and clinicians from academia 5 and public interest groups pool their technical and financial resources and form a single 6 committee comprised of scientists from both the public sector and the private sector to develop 7 Acute Exposure Guideline Levels (AEGL values). EPA conceived the idea to formulate general 8 guidance for developing short-term exposure limits and together with ATSDR subsequently 9 funded a subcommittee of the Committee of Toxicology of the National Research Council, U. S. 10 National Academy of Sciences (NRC/NAS) to develop guidance on the use of procedures and 11 methodologies to establish emergency exposure guideline levels for the general public. 12 13 Since the 1940's, the NRC/NAS Committee on Toxicology has developed emergency 14 exposure guidelines for 41 chemicals of concern to the U. S. Department of Defense (DOD). 15 These values are referred to as "Emergency Exposure Guideline Levels" (EEGLs). Although the 16 EEGLs were developed for use with military personnel, the NAS also developed special 17 exposure guidelines for the general public, termed "Short-term Public Exposure Guidance 18 Levels" (SPEGLs). Based on this extensive experience and the high level scientific and technical 19 expertise continually available to the NAS, this organization was considered the most qualified 20 entity to develop guidance on the methodologies and procedures used in the establishment of 21 short-term exposure limits for acutely toxic chemicals. 22 23 The NAS guidance document, entitled Guidelines for Developing Community Emergency 24 Exposure Levels for Hazardous Substances, was published in 1993. The Community Emergency 25 Exposure Levels (CEELs) and the Acute Exposure Guideline Levels (AEGLs) represent the 26 identical short-term emergency exposure levels. The NAS name (CEELs) has been replaced by a 27 new name (AEGLs) only to convey the broad applications of these values for planning, 28 response, and prevention in the community, the workplace, transportation, the military, and the 29 remediation of superfund sites. A discussion of how AEGLs might be used for emergency 30 planning, response, and prevention appears later in this chapter. 31 32 The efforts to mobilize the federal and state agencies and individuals and organizations in 33 the private sector to form the committee began shortly thereafter. In October, 1995 the 34 committee was formally chartered and the charter filed with the U.S. Congress under the Federal 35 Advisory Committee Act (FACA) with approval by the Office of Management and Budgets 36 (OMB) and concurrence by the General Services Administration (GSA). Due to EPA budgetary 37 constraints, the first meeting of the NAC/AEGL Committee was not held until June, 1996. This 38 meeting represented the culmination of the efforts to solicit stakeholders, identify committee 39 members, form the committee, obtain the technical support of the Oak Ridge National 40 Laboratories (ORNL), and begin the development of the AEGL values. 41 42 SopOB-02 wpd Printed July 6, 2000 2 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 PURPOSE AND OBJECTIVES OF THE AEGL PROGRAM AND THE 2 NAC/AEGL COMMITTEE 3 4 The primary purpose of the AEGL Program and the NAC/AEGL Committee is to develop 5 guideline levels for once in a lifetime short-term exposures to airborne concentrations of acutely 6 toxic, high priority chemicals. These Acute Exposure Guideline Levels (AEGLs) are needed for 7 a wide range of planning, response, and prevention applications. These applications may include 8 the EPA's SARA Title ffl Section 302-304 emergency planning program, the U. S. Clean Air Act 9 Amendments (CAAA) Section 112(r) accident prevention program, and the remediation of 10 Superfund sites program; the DOE environmental restoration, waste management, waste 11 transport, and fixed facility programs; the DOT emergency waste response program; the DOD 12 environmental restoration, waste management, and fixed facility programs; ATSDR health 13 consultation and risk assessment programs; NIOSH/OSHA regulations and guidelines for 14 workplace exposure; State CAA Section 112(b) programs and other state programs; and private 15 sector programs such as the AIHA-ERPG and the CMA Chemtrec programs. 16 17 A principal objective of the NAC/AEGL Committee is to develop scientifically credible, 18 acute (short-term) once in a lifetime exposure guideline levels within the constraints of data 19 availability, resources and time. This includes highly effective and efficient efforts in data 20 gathering, data evaluation and data summarization, fostering the participation of a large cross- 21 section of the relevant scientific community, and the adoption of procedures and methods that 22 facilitate consensus-building for AEGL values within the Committee. *>3 ^4 Another principal objective of the committee is to develop these AEGL values for 25 approximately 400 to 500 acutely hazardous substances within the next ten (10) years. 26 Therefore, the near-term objective is to increase the level of production of AEGL development to 27 approximately forty (40) to fifty (50) chemicals per year without exceeding budgetary limitations 28 or compromising the scientific credibility of the values developed. 29 30 Further, in addition to determining AEGL values for three different health effect end- 31 points, it is intended to derive exposure values for the general public that are applicable to 32 emergency (accidental) once m a lifetime exposure periods ranging from 10 minutes to 8 hours 33 duration. Therefore, exposure limits will be developed for a minimum of 5 exposure periods (10 34 minutes, 30 minutes, 1 hour, 4 hours, 8 hours). Each AEGL tier is distinguished by varying 35 degrees of severity of toxic effects, as initially conceived by the AIHA ERP Committee and 36 further defined in the NAS' National Research Council report, Guidelines for Developing 37 Community Emergency Exposure Levels for Hazardous Substances, published by the National 38 Academy of Sciences in 1993 (NAS Guidance), and further defined by the NAC/AEGL 39 Committee. These AEGL-1, AEGL-2, and AEGL-3 definitions are presented elsewhere in this 40 SOP manual. 41 42 As stated in the NAS guidelines and described in the AEGL definitions, these exposure Sop08-02 wpd Printed July 6, 2000 3 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 limits are intended to protect most individuals in the general population, including those that may 2 be particularly sensitive or susceptible to the deleterious effects of the chemicals. However, as 3 stated in the guidelines and the definitions, it is recognized that certain individuals, subject to 4 unique and idiosyncratic responses, could experience effects at concentrations below the 5 corresponding AEGL. 6 7 An important objective of the NAC/AEGL Committee is the establishment and 8 maintenance of a comprehensive "Standing Operating Procedures" manual (SOP Manual) that 9 adheres to the 1993a NRC/NAS guidelines and supplements, clarifies, interprets or defines these 10 guidelines with regard to the specific use of certain procedures and methods such as the selection 11 of NOAELs, LOELs, etc., use of uncertainty factors, modifying factors, interspecies/intraspecies 12 extrapolation methodologies, time scaling, carcinogenic risk assessment, and other methods and 13 procedures relevant to the development of AEGL values. 14 15 16 COMMITTEE MEMBERSHIP AND ORGANIZATIONAL STRUCTURE 17 ] 8 The NAC/AEGL Committee is comprised of representatives of federal, state and local 19 agencies, and organizations in the private sector that derive programmatic or operational benefits 20 from the AEGL values. This includes federal representatives from the U.S. Environmental 21 Protection Agency (EPA), the Department of Energy (DOE), the Agency for Toxic Substances 22 and Disease Registry (ATSDR), the National Institute for Occupational Safety and Health 23 (NIOSH), Occupational Safety and Health Administration (OSHA), the Department of 24 Transportation (DOT), the Department of Defense (DOD), the Center for Disease Control 25 (CDC), the Food and Drug Administration (FDA), and the Federal Emergency Management 26 Agency (FEMA). States providing committee representatives include New York, New Jersey, 27 Texas, California, Minnesota, Dlinois, Connecticut, and Vermont. Private companies with 28 representatives include Allied Signal Corporation, Exxon Corporation, and Olin Chemical 29 Company. Other organizations with representatives include the American Industrial Hygiene 30 Association (AM A), American College of Occupational and Environmental Medicine 31 (ACOEM), American Association of Poison Control Centers (AAPCC), and the AFL-CIO. In 32 addition, the committee membership includes individuals from academia, a representative of 33 environmental justice, and other organizations in the private sector. A current list of the 34 NAC/AEGL Committee members and their affiliations is shown in Appendix A of this SOP 35 manual. At present, the Committee is comprised of 32 members. 36 37 Recently, the Organization of Economic and Cooperation Development (OECD) and 38 various OECD member countries have expressed an interest in the AEGL Program. Several 39 OECD member countries such as Germany and the Netherlands have been participating m the 40 Committee's activities and actively pursuing formal membership on the NAC/AEGL Committee 41 It is envisioned that the Committee and the AEGL Program in general will progressively expand 42 its scope and participation to include the international community. Sop08-02 wpd Printed July 6, 2000 4 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 The Director of the AEGL Program has the overall responsibility for the entire AEGL 2 Program and the NAC/AEGL Committee and its activities. A Designated Federal Officer (DFO) 3 is responsible for all administrative matters related to the Committee to insure that it functions 4 properly and efficiently. These individuals are not voting members of the Committee. The 5 NAC/AEGL Committee Chair is appointed by EPA and is selected from among the committee 6 members. In concert with the Program Director and the DFO, the Chair coordinates the activities 7 of the Committee and also directs all formal meetings of the Committee. From time to time, the 8 members of the Committee serve as Chemical Managers and Chemical Reviewers in a 9 collaborative effort with assigned scientist-authors (non-Committee members) to develop AEGLs 10 for a specific chemical. These groups of individuals are referred to as the AEGL Development 11 Teams and their function is discussed in Section 4.8 of this manual.. 12 13 SELECTION OF CHEMICALS FOR AEGL DEVELOPMENT 14 15 A master list of approximately 1,000 acutely toxic chemicals was initially compiled 16 through the integration of individual priority lists of chemicals submitted by each U. S. federal 17 agency placing a representative on the Committee. The master list was subsequently reviewed by 18 individuals from certain state agencies and representatives from organizations in the private 19 sector and modified as a result of comments and suggestions received. The various priority 20 chemical lists were compiled separately by each federal agency based on their individual 21 assessments of the hazards, potential exposure, risk, and relevance of a chemical to their 22 programmatic needs. A list of approximately 400 chemicals representing the higher pnonty 23 chemicals was tentatively identified from the original master list. It was acknowledged that this 24 list was subject to change based on the changing needs of the stakeholders. 25 26 On May 21, 1997, a list of 85 chemicals was published in the Federal Register. This list 27 identified those chemicals from the list of approximately 400 chemicals considered to be of 28 highest priority across all U. S. federal agencies and represented the selection of chemicals for 29 AEGL development by the NAC/AEGL Committee for the first two to three years of the 30 program. The Committee has now addressed these chemicals and they are presently in the Draft, 31 Proposed, Interim, or Final stages of development. Certain chemicals did not contain an 32 adequate database for AEGL development and, consequently, are on hold pending decisions 33 regarding further testing This initial "highest" priority list of 85 chemicals is shown in 34 Appendix B. 35 36 A second "working list" of approximately 100 priority chemicals is being selected from 37 (1) the original master list, (2) the intermediate list of approximately 400 chemicals (which is a 38 subset of the master list) and (3) from new, high priority candidate chemicals submitted by U. S. 39 Agencies and organizations and OECD member countries that are planning to participate in the 40 AEGL Program. Although "working lists" will be published in the U. S. Federal Register and 41 elsewhere from time-to-time to indicate the NAC/AEGL Committee's agenda, the priority of 42 chemicals addressed, and , hence, the "working list" is subject to modification if priorities of the Sop08-02 wpd Printed July 6. 2000 5 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30,2000 1 NAC/AEGL Committee or individual stakeholder organizations, including international 2 members, change during that period. 3 4 5 SCIENTIFIC CREDIBILITY OF AEGLS 6 7 The scientific credibility of the AEGL values is based on adherence to the National 8 Academy of Sciences 1993a guidelines for developing short-term exposure limits, the 9 comprehensive nature of data collection and evaluation, the consistency of the methods and 10 procedures used to develop the values, the potential of acute toxicity testing in cases of 11 inadequate data, and the adoption of the most comprehensive peer review process ever used to 12 establish short-term exposure limits for acutely toxic chemicals. 13 14 The comprehensive data gathering process involves literature searches for all relevant 15 published data and the mobilization of all relevant unpublished data. Data and information from 16 unpublished sources is obtained through individual companies in the private sector and the 17 cooperation of trade associations. The completeness of the data searches is enhanced through the 18 oversight and supplemental searches conducted by individual Committee members and interested 19 parties during the peer review process. 20 21 Data evaluation and selection is performed by scientists with expertise in toxicology and 22 related disciplines from staff at the organization which drafts Technical Support Documents and 23 the assigned members of the NAC/AEGL Committee. Additionally, input on data evaluation and 24 selection is provided by interested parties who participate in the open meetings of the Committee 25 or who formally comment on the Federal Register notices of Proposed AEGL values 26 27 The work of the NAC/AEGL Committee adheres to the 1993a NRC/NAS publication 28 Guidelines for Developing Community Emergency Exposure Levels for Hazardous Substances. 29 Since this guidance document represents a more general guidance for methods and procedures, 30 the NAC/AEGL Committee interprets and develops greater detail related to the methodologies 31 and procedures that it follows. These Standing Operating Procedures (SOPs) are documented by 32 the SOP Workgroup and represent a consensus or two-thirds majority vote of the NAC/AEGL 33 Committee. SOPs also represent concurrence of the National Academy of Sciences' 34 Subcommittee on Acute Exposure Guideline Levels (NAS/AEGL Subcommittee). Therefore, 35 each step of the AEGL development process follows specific methodologies, criteria or other 36 guidelines to insure consistent, scientifically sound values. 37 38 In instances where AEGL values cannot be developed because of poor data or no data, the 39 chemical may be subjected to appropriate acute toxicity testing. The AEGL program is 40 committed to insuring that AEGL values are derived from adequate data and information based 41 on a consensus or two-thirds majority vote of the NAC/AEGL Committee and concurrence of the 42 NAS/AEGL Subcommittee. Sop08-02 wpd Printed July 6. 2000 g ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 To further assure the scientific credibility of the AEGL values and their supporting 2 rationale, the most comprehensive peer review process ever employed in the development of 3 short-term exposure limits has been established (see next section). This review process has been 4 designed to effectively, yet efficiently, encourage and enable the participation of the scientific 5 community and other interested parties from the public and private sectors in the development of 6 the AEGLs. Further, the review process utilizes an expert committee of the National Academy 7 of Sciences, the NAS/AEGL Subcommittee as the final scientific review . Hence, the final 8 judgement of scientifically credible values rests with the United State's ultimate scientific body, 9 the NAS. A detailed summary of the AEGL development process is presented in the next 10 section. 11 12 13 THE AEGL DEVELOPMENT AND PEER REVIEW PROCESS 14 15 The process that has been established for the development of the AEGL values is the 16 most comprehensive ever employed for the determination of short-term exposure limits for 17 acutely toxic chemicals. A summary of the overall process is presented in diagram form in 18 Appendix C. The process consists of four basic stages in the development and status of the 19 AEGLs and they are identified according to the review level and concurrent status of the AEGL 20 values. They include (1) "Draft" AEGLs, (2) "Proposed" AEGLs, (3) "Interim" AEGLs and (4) 21 "Final" AEGLs. The entire development process can be described by individually describing the 22 four basic stages in the development of AEGL values. 23 24 25 Stage 1: "Draft" AEGLs 26 27 This first stage begins with a comprehensive search of the published scientific literature. 28 Attempts are made to mobilize all relevant, non-published data through industry trade 29 associations and from individual companies in the private sector. A more detailed description of 30 the published and unpublished sources of data and information utilized is provided in Section 2.3 31 of this document which addresses search strategies. The data are evaluated following the 32 guidelines published in the NRC/NAS guidance document and this SOP manual and selected 33 data are used as the basis for the derivation of the AEGL values and the supporting scientific 34 rationale. Data evaluation, data selection, and the development of a technical support document 35 are all performed as a collaborative effort among the Staff Scientist at the organization which 36 drafts Technical Support Documents, the Chemical Manager, and two Chemical Reviewers. 37 This group is referred to as an "AEGL Development Team". NAC/AEGL Committee members 38 are specifically assigned this responsibility for each chemical under review. Hence, a separate 39 team comprised of different Committee members is formed for each chemical under review The 40 product of this effort is a technical support document (TSD) that contains "Draft" AEGLs. The 41 Draft TSD is subsequently circulated to all other NAC/AEGL Committee members for review 42 and comment prior to a formal meeting of the Committee. Revisions to the initial TSD and the Sop08-02 wpd Printed July 6, 2000 7 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30,2000 1 "Draft" AEGLs are made up to the time of the NAC/AEGL Committee meeting scheduled for 2 formal presentation and discussion of the AEGL values and the documents. Following 3 deliberations during the committee meeting, an attempt is made to reach consensus, or the 4 minimum of a two-thirds majority of a quorum present, to elevate the AEGLs to "Proposed" 5 status. If agreement cannot be reached, the Committee conveys its issues and concerns to the 6 AEGL Development Team and further work is conducted by this group. After completion of 7 additional work, the chemical is resubmitted for consideration at a future meeting. If a consensus 8 or two-thirds majority vote of the Committee cannot be achieved because of inadequate data 9 unrelated to the completeness of the data search, the chemical becomes a candidate for 10 appropriate toxicity studies. 11 12 13 Stage 2: "Proposed" AEGLs 14 15 Once the NAC/AEGL Committee has reached a consensus, or the minimum two-thirds 16 majority vote, on the AEGL values and supporting rationale, they are referred to as "Proposed" 17 AEGLs and are published in the Federal Register for a thirty (30) day review and comment 18 period. Following publication of the "Proposed" AEGLs in the Federal Register, the Committee 19 reviews the public comments, addresses and resolves relevant issues and seeks a consensus or 20 minimum two-thirds majority of those present on the Committee on the original or modified 21 AEGL values and the accompanying scientific rationale. 22 23 24 Stage 3: "Interim" AEGLs 25 26 Following resolution of relevant issues raised through public review and comment and 27 subsequent approval of the Committee, the AEGL values are classified as "Interim". The 28 "Interim" AEGL status represents the best efforts of the NAC/AEGL Committee to establish 29 exposure limits and the values are available for use as deemed appropriate on an interim basis by 30 federal and state regulatory agencies and the private sector. The "Interim" AEGLs, the supporting 31 scientific rationale, and the TSD, are subsequently presented to the National Academy of 32 Sciences (NAS/AEGL Subcommittee) for its review and concurrence. If concurrence cannot be 33 achieved, the NAS/AEGL Subcommittee will submit its issues and concerns to the NAC/AEGL 34 Committee for further work and resolution. 35 36 37 Stage 4: "Final" AEGLs 38 39 When concurrence by the NAS/AEGL Subcommittee is achieved, the AEGL values are 40 considered "Final" and published by the U. S. NAS. Final AEGLs may be used on a permanent 41 basis by all federal, state and local agencies and private sector organizations. It is possible that 42 from time to time new data will become available that challenges the scientific credibility of Sop08-02 wpd Printed July 6,2000 8 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 "Final" AEGLs. If this occurs, the chemical will be resubmitted to the NAC/AEGL Committee 2 and recycled through the review process. 3 4 5 OPERATION OF THE COMMITTEE 6 7 The NAC/AEGL Committee meets formally four (4) times each year for two and one-half 8 (2-1/2) days. The meetings are scheduled for each quarter of the calendar year and are generally 9 held m the months of March, June, September, and December. Based on overall cost 10 considerations, the meetings are generally held in Washington, D.C. However, from time to 11 time, committee meetings may be held at other locations for justifiable reasons. 12 13 At least 15 days prior to the committee meetings, a notice of the meeting is published in 14 the Federal Register together with a list of chemicals and other matters to be addressed by the 15 Committee and provides dates, times and location of the meetings. The agenda is finalized and 16 distributed to committee members approximately one week prior to the meeting. The agenda 17 also is available to other interested parties at that time, upon request, through the Designated 18 Federal Officer (DFO). 19 20 All NAC/AEGL Committee meetings are open to the public and interested parties may 21 schedule individual presentations of relevant data and information by contacting the DFO to 22 establish a date and time. Relevant data and information from interested parties also may be 23 provided to the Committee through the DFO during the period of development of the Draft 24 AEGLs so that it can be considered during the early stage of development. Data and information 25 also may be submitted during the Proposed and Interim stages of AEGL development as well. 26 27 The NAC/AEGL Committee meetings are conducted by the Chair who is appointed by 28 the U.S. Environmental Protection Agency in accordance with the Federal Advisory Committee 29 Act (FACA). At the time of the meeting, both the Chair and all other committee members will 30 have received the initial draft and one or more revisions of the Technical Support Document 31 (TSD) and "Draft", "Proposed", or "Interim" AEGL values for each chemical on the agenda. 32 Reviews, comments, and revisions are continuous up to the time of the meeting and committee 33 members are expected to be familiar with the "Draft", "Proposed", or "Interim" AEGLs, 34 supporting rationale, and other data and information in each TSD and to participate in the 35 resolution of residual issues at the meeting. Procedures for the AEGL Development Teams and 36 the other Committee members regarding work on AEGLs in the Proposed or Interim status are 37 similar to those for Draft AEGLs. 38 39 All decisions of the NAC/AEGL Committee related to the development of Draft, 40 Proposed, Interim, and Final AEGLs and their supporting rationale are made by consensus or a 41 minimum of two-thirds (2/3) majority of a quorum of committee members. A quorum of the 42 NAC/AEGL Committee is defined as fifty-one percent (51 %) or more of the total NAC/AEGL Sop08-02 wpd Printed July 6.2000 9 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Committee membership in attendance. 2 3 The highlights of each meeting are recorded by the scientists who draft the Technical 4 Support Documents and written minutes are prepared, ratified and maintained in the 5 Committee's permanent records. Deliberations of each meeting also are tape-recorded when 6 possible and stored in the Committee's permanent records by the Designated Federal Officer 7 (DFO) for future reference as necessary. 8 9 All Proposed AEGL values and supporting scientific rationale are published in the 10 Federal Register. Review and comment by interested parties and the general public are requested 11 and encouraged. The Committee's response to official comments on Federal Register notices on 12 Proposed AEGL values consists of an evaluation of the comments received, discussions and 13 deliberations that take place at Committee meetings regarding the considerations of elevation of 14 AEGLs from "Proposed" to "Interim" status, and changes to the Technical Support Documents 15 as deemed appropriate by the NAC/AEGL Committee. This information is reflected on the tapes 16 and in the minutes of the meetings and will be maintained for future reference. 17 18 As previously mentioned a "Standing Operating Procedures" Workgroup (SOP 19 Workgroup) was established in March, 1997 to document, summarize, and evaluate the various 20 procedures, methodologies, and guidelines employed by the Committee in the gathering and 21 evaluation of scientific data and information and the development of the AEGL values. The SOP 22 Workgroup performs a critical function by continually providing the Committee with detailed 23 information on the Committee's interpretation of the NAS guidelines and the approaches the 24 Committee has taken in the denvation of each AEGL value for each chemical addressed. This 25 documentation enables the Committee to continually assess the basis for its decision-making, 26 insure consistency with the NAS guidelines, and maintain the scientific credibility of the AEGL 27 values and accompanying scientific rationale. This ongoing effort is continuously documented 28 and is identified as the "SOP Manual". 29 30 31 VALUE OF A COLLABORATIVE EFFORT IN THE AEGL PROGRAM 32 33 The value of a collaborative effort in the AEGL Program is related primarily to the 34 pooling of substantial resources of the various stakeholders and the direct or indirect involvement 35 of a significant portion of the relevant scientific community from both the public and private 36 sectors. These factors, in turn, promote greater productivity, efficiency and cost effectiveness of 37 such an effort and greatly enhance the scientific credibility of the Acute Exposure Guideline 38 Levels (AEGLs) that are developed by the Committee. 39 40 The formation of the Federal Advisory Committee for Acute Exposure Guideline Levels 41 for Hazardous Substances (NAC/AEGL Committee) with approximately 30 to 35 members has 42 provided an important forum for scientists, clinicians, and others to develop the AEGLs and Sop08-02 wpd Printed July 6, 2000 10 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 related scientific issues. The composition of the Committee represents a balanced cross-section 2 of relevant scientific disciplines and a balance of U. S. federal and state agencies, academia, the 3 medical community, private industry, public interest groups, and other organizations in the 4 private sector. This mutual participation of stakeholders, including the regulators and the 5 regulated community, in the development of the AEGLs promotes the acceptance of the AEGLs 6 by all parties involved. Additionally, the diverse composition of the committee represents the 7 nucleus of a broad network of scientists, clinicians, and other technical personnel that fosters 8 information and data exchange and the resolution of relevant scientific and technical issues well 9 beyond the committee membership. This network also facilitates the identification of national 10 and international experts with particular expertise that may provide important data, information 11 or insight on a specific chemical or scientific issue. 12 13 The collaborative effort also results in greater scientific credibility of the exposure values 14 developed. The pooling of resources enables a very comprehensive gathering and evaluation 15 effort of both published and unpublished data and information. Collaboration provides a broad 16 base of relevant scientific knowledge and expertise that is highly focused on the chemicals and 17 issues addressed by the Committee. This approach provides sufficient scientific and technical 18 resources for the SOP Workgroup to document and evaluate procedures and methodologies that 19 instill rigor and consistency into the process and the resultant AEGL values. The documentation 20 of these procedures and methodologies are contained in this Standing Operating Procedures 21 Manual (SOP Manual). Finally, the collaborative effort has enabled the establishment of the most 22 comprehensive peer review process ever implemented for the development of short-term 23 exposure limits. 24 25 Recently the AEGL Program has extended invitations to all OECD member countries to 26 participate on the NAC/AEGL Committee and the program activities in general. It is believed 27 that expanding the scope of the AEGL Program to include the international community will be of 28 great benefit. Their participation will provide even greater resources, further broaden the base of 29 scientific and technical expertise, provide new toxicological data and insights, and foster the 30 harmonization of emergency exposure limits at the international level. 31 32 In summary, the establishment of a collaborative effort, with its pooling of resources, 33 represents the most productive, efficient, and cost-effective approach to the development of 34 exposure guideline levels. Further, the effort results in the development of uniform values for a 35 wide range of applications. This eliminates inconsistencies and confusion among individuals and 36 organizations involved in emergency planning, response and prevention of chemical accidents. 37 In global terms, the NAC/AEGL Committee represents an approach to unifying the international 38 community in the development and use of chemical emergency exposure limits. In the interest of 39 multinational companies seeking uniform operating parameters and the mandates placed on 40 federal agencies to achieve international harmonization of standards and guidelines, the 41 participation of the international community in the AEGL Program represents an important goal 42 of the AEGL program. Sop08-02 wpd Printed July 6, 2000 11 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2 APPLICATIONS OF THE AEGL VALUES 3 4 As previously stated, it is anticipated that the AEGL values will be used for both 5 regulatory and non-regulatory purposes by federal and state agencies in conjunction with 6 chemical emergency prevention, preparedness, and/or response programs. This includes the 7 implementation of these chemical emergency activities at the local community level. 8 9 More specifically, the AEGL values will be used for conducting various risk assessments 10 to aid in the development of emergency preparedness and prevention plans, as well as real-time 11 emergency response actions, for accidental chemical releases at fixed facilities and from transport 12 carriers. The AEGL values, which represent defined toxic endpoints, are used in conjunction 13 with various chemical release and dispersion models to determine geographical areas, or 14 "vulnerable zones", associated with accidental or terrorist releases of chemical substances. By 15 determining these geographical areas, and the presence of human populations and facilities 16 within these zones, the potential risks associated with accidental chemical releases can be 17 estimated. For example, the release and dispersion models, which take into account the quantity 18 and rate of release of the chemical, the volatility of the substance, the wind speed and wind 19 stability at the time of the release, and a consideration of the topographical characteristics in the 20 area of the release, will define the geographical areas exposed, and quantitatively, the airborne 21 concentration of the "plume" or the chemical cloud as it is dispersed. By comparing the 22 projected airborne concentrations of the chemical substance in question with the exposed 23 populations, human health risks associated with a chemical release can be estimated. Using these 24 nsk estimates, emergency response personnel can make effective risk management and risk 25 communication decisions to minimize the adverse impact of the release on human health. Figure 26 1-1 is a summary diagram that indicates the overall effects that are expected to occur above each 27 of the three AEGL threshold tiers, as well as sensory and non-sensory or asymptomatic effects 28 below the AEGL-1 threshold level. Figure 1-1 also indicates the expected increase in occurrence 29 and severity of the various adverse health effects as the airborne concentration increases beyond 30 each of the three AEGLs. 31 32 Because of the complex nature of chemical accidents, the populations at risk, the variable 33 capabilities among emergency response units, and many other considerations related to a specific 34 event, it is beyond the scope of this document to discuss or speculate on specific actions that 35 should or could be taken at any point in tune 01 at a given level of exposure to a specific 36 chemical. However, it is known by emergency responders and planners that vanous options are 37 available, depending upon the circumstances, for reducing or even preventing the adverse 38 impacts of chemical releases. In general they include public notification and instruction, 39 sheltering-in-place, selective of major evacuation procedures, procedures to enable or facilitate 40 medical attention or some combination of these approaches. These are important decisions best 41 left to local emergency planners and responders to be addressed on a case-by-case basis. Further, 42 information regarding the applications of short-term exposure limits such as AEGLs may be Sop08-02 wpd Printed July 6,2000 12 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 obtained m Technical Guidance for Hazards Analysis (U.S. EPA, 1987). 2 3 4 Sop08-02 wpd Printed July 6. 2000 13 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30,2000 1 FIGURE 1-1 HAZARD ASSESSMENT 2 Threshold Levels HAZARD ASSESSMENT Effects DEATH AEGL-3 Increasing likelihood of death DISABLING -Impairment of ability to escape -Increasing severity of irreversible or ofr.er long-lasting effects AEGL-2 DISCOMFORT AEGL-1 -Increase in notable discomfort -increasing severity of reversible effects (with or without signs/symptoms) DETECTABILITY Increasing complaints of objectionable odor, taste, sensory irritation or other mild, non-sensory or asymptomatic effects SopOB-02 wpd Printed July 6,2000 14 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 i 2. DERIVATION OF AEGL VALUES 2 3 2.1 DEFINITIONS OF AEGL-1, AEGL-2 AND AEGL-3 4 5 AEGL seventy levels represent short-term exposure values which are a threshold for 6 specific biological effects for the general public and are applicable to specified exposure 7 durations. The values for these specified durations are "... ceiling exposure values for the public 8 (i.e., a ceiling is a concentration of a substance that should never be exceeded)..." (NRC 1993a, 9 p2). Three AEGLs are developed for each of five exposure durations (10 and 30 minutes, 1 hour, 10 4 hours, and 8 hours) and are distinguished by varying degrees of severity of toxic effects. 11 AEGLs for 10 minute durations will be developed for the chemicals included in the first 12 publication of AEGLs by the National Academy of Sciences at a future date. 13 14 PREFACE 15 16 Under the authority of the Federal Advisory Committee Act (FACA) P. L. 92-463 of 17 1972, the National Advisory Committee for Acute Exposure Guideline Levels for Hazardous 18 Substances (NAC/AEGL Committee) has been established to identify, review and interpret 19 relevant toxicologic and other scientific data and develop AEGLs for high priority, acutely toxic 20 chemicals. 21 22 AEGLs represent threshold exposure limits for the general public and are applicable to 23 emergency exposure periods ranging from 10 minutes to 8 hours. AEGL-2 and AEGL-3 levels, 24 and AEGL-1 levels as appropriate, will be developed for each of five exposure periods (10 and 25 30 minutes, 1 hour, 4 hours, and 8 hours) and will be distinguished by varying degrees of seventy 26 of toxic effects. It is believed that the recommended exposure levels are applicable to the general 27 population including infants and children, and other individuals who may be sensitive or 28 susceptible. The three AEGLs have been defined as follows: 29 30 AEGL-1 is the airborne concentration (expressed as ppm or mg/m3) of a substance above 31 which it is predicted that the general population, including susceptible individuals, could 32 experience notable discomfort, irritation, or certain asymptomatic, non-sensory effects. 33 However, the effects are not disabling and are transient and reversible upon cessation of 34 exposure. 35 36 AEGL-2 is the airborne concentration (expressed as ppm or mg/m3) of a substance above 37 which it is predicted that the general population, including susceptible individuals, could 38 experience irreversible or other senous, long-lasting adverse health effects, or an impaired ability 39 to escape. 40 41 AEGL-3 is the airborne concentration (expressed as ppm or mg/m3) of a substance above Sop08-02 wpd Printed July 6, 2000 15 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30.2000 1 which it is predicted that the general population, including susceptible individuals, could 2 experience life-threatening health effects or death. 3 4 Airborne concentrations below the AEGL-1 represent exposure levels that could produce 5 mild and progressively increasing odor, taste, and sensory irritation, or certain asymptomatic, 6 non-sensory effects. With increasing airborne concentrations above each AEGL level, there is a 7 progressive increase in the likelihood of occurrence and the seventy of effects described for each 8 corresponding AEGL level. Although the AEGL values represent threshold levels for the general 9 public, including sensitive subpopulations, it is recognized that certain individuals, subject to 10 unique or idiosyncratic responses, could experience the effects described at concentrations below 11 the corresponding AEGL level. 12 SopOB-02 wpd Printed July 6, 2000 16 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2.2 EMPIRICAL TOXICOLOGIC ENDPOINTS, AND METHODS FOR 2 DETERMINING EXPOSURE CONCENTRATIONS USED TO DERIVE 3 AEGL-1,2, AND 3 LEVELS 4 5 The selection of the biological endpoints that serve as the thresholds for each of the 6 AEGL severity levels are based on the definitions for the Community Emergency Exposure 7 Levels (CEELs) that were published in the 1993a National Academy of Sciences' (NAS) 8 guidelines for developing short-term exposure limits. The AEGLs address the same defined 9 population as the NAS CEELs. The NAS definitions of the 3 CEEL tiers have been modified 10 slightly by the NAC/AEGL Committee only to improve the clarity of description of the threshold 11 levels. Hence, the defined threshold levels for CEELs and AEGLs are the same. 12 13 The NAS guidelines describe CEELs (or AEGLs) as ceiling exposure values (i.e., a 14 concentration of a substance that should never be exceeded) that are applicable to emergency 15 exposures to hazardous substances for a specified duration (NAS, 1993). The NAS guidance 16 further states that the CEELs (or AEGLs) must be set low enough to protect most of the 17 population that might be exposed, including those with increased susceptibilities such as 18 children, pregnant women, asthmatics and persons with other specific illnesses (NAS, 1993). 19 The NAS definition of CEELs/AEGLs for each of the three different tiers of adverse health 20 effects states that the adverse effects for each CEEL/AEGL tier is not likely to occur below that 21 level for a specified exposure duration, but becomes increasingly likely to occur at concentrations 22 above that level in a general population, including susceptible individuals. For this reason the 23 NAS also refers to the CEELs/AEGLs as threshold levels (NAS, 1993). 24 25 Because the data and methodologies used to derive AEGLs or any other short-term 26 exposure limits are not sufficiently precise to make a distinction between a ceiling value and a 27 threshold value, no distinction has been made with respect to AEGL values. No fine line can be 28 drawn to precisely differentiate between a ceiling level, which represents the highest exposure 29 concentration for which an effect is unlikely to occur, and a threshold level, which represents the 30 lowest exposure concentration for the likelihood of onset of a given set of effects. Hence, 31 AEGLs are not true effect levels. Rather, they are considered threshold levels that represent an 32 estimated point of transition and reflect the best efforts to quantitatively establish a demarcation 33 between one defined set of symptoms or adverse effects and another defined set of symptoms or 34 adverse effects. Therefore, in the development of AEGLs the NAC/AEGL Committee selects the 35 highest exposure level from animal or human data where the effects used to define a given AEGL 36 tier are not observed. 37 38 39 2.2.1 SELECTION OF THE HIGHEST EXPOSURE LEVEL WHERE THE 40 EFFECTS USED TO DEFINE AN AEGL LEVEL WERE NOT OBSERVED 41 Sop08-02 wpd Printed July 6, 2000 17 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Traditionally, when setting acceptable (typically considered "safe") levels of exposure the 2 evaluator will select the highest experimental exposure which does not cause an adverse effect 3 (No Observed Adverse Effect Level - NOAEL) in an experiment which demonstrated a graded 4 exposure response from no effect to adverse effects. In standard nsk assessment practice (NRC, 5 1993a), the exposure level identified as the NOAEL would then be divided by appropriate 6 uncertainty factors and modifying factors to derive an acceptable exposure level for humans 7 However, there are a number of limitations in this methodology. It does not consider the number 8 of animals used in the expenment and the associated statistical uncertainty around the 9 experimental exposure level chosen. It does not consider the slope of the exposure-response 10 relationship and subjects the evaluator to use the possibly arbitrarily selected exposure levels 11 which were chosen in the face of an unknown exposure-response relationship. Under some 12 conditions, especially a small number of animals exposed per exposure, the NOAEL could be a 13 level associated with significant adverse health effects (Leisenring and Ryan 1992). In recent 14 years Crump (1984), Barnes et al. (1995), US EPA (1995a), Faustman et al. (1994), Gaylor et al. 15 (1998), Gaylor et al. (1999), and Fowles et al. (1999) addressed these problems by using the 16 concept of analyzing all of the data to statistically estimate a benchmark concentration (BMC). 17 The BMC is a statistical estimate of an exposure which will cause a specified incidence of a 18 defined adverse health effect. The BMC is commonly defined as the 95% lower confidence limit 19 (LCL) on the exposure causing a specified level of response (typically 1% to 10%). This 20 exposure is intended to replace the NOAEL and is used like the NOAEL when setting acceptable 21 exposure levels. 22 23 The BMC methodology has a number of advantages over the traditional NOAEL 24 approach. The BMC is derived from a statistical analysis of the exposure-response relationship 25 and is not subject to investigator selection of exposure levels. It is a reflection of the exposure 26 response curve. Although the number of animals used in a study will impact the NOAEL and 27 BMC estimates, the BMC, when compared to the maximum likelihood estimate (MLE), will 28 explicitly reflect the variability in the study and the uncertainty around the number of subjects. 29 The greater the variability and uncertainty, the greater the difference between the BMC and the 30 MLE. The BMC calculation allows for the statistical estimation of a BMC in the absence of an 31 empirical NOAEL. 32 33 The data most relevant to the development of AEGL-3 values and most amenable to a 34 benchmark concentration analysis are inhalation LC50 data. Fowles et al. (1999) analyzed 120 35 inhalation animal lethality data sets using the BMC methodology. The analyses provide the basis 36 for the application of the BMC approach used by the NAC/AEGL Committee in the development 37 of AEGL values. Benchmark concentrations (95% LCL) and maximum likelihood estimates 38 were developed for the 1,5, and 10% response levels using log probit and Weibull models. 39 Species tested included rats, mice, guinea pigs, hamsters, rabbits, and dogs. Exposure times 40 ranged from 5 minutes to 8 hours. Each data set consisted of at least 4 data points. The BMC 41 and MLE values were compared with the empirical NOAEL (highest exposure which did not 42 cause death in the experiment) and LOAEL (lowest exposure which killed at least one animal). Sop08-02 wpd Printed July 6. 2000 18 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 The curve generated by the statistical models was subjected to a chi-squared goodness of fit test 2 (P>0.05). For statistical and data presentation reasons, 100 studies were analyzed with the probit 3 analysis and 93 with the Weibull model. Most of the studies reported NOAELs (81/100 which 4 were considered for the probit analysis and 74/93 considered for the Weibull analysis). 5 6 The benchmark concentrations were generally lower than the NOAELs when analyzed 7 with either statistical estimate. The mean NOAEL/BMC ratios for the 1, 5, and 10% response 8 were 1.60,1.16, and 0.99 when using a probit analysis and 3.59, 1.59, and 1.17 when using the 9 Weibull analysis. It is interesting to note that comparable means from a Weibull analysis of 10 developmental toxicity data were considerably greater, the developmental toxicity means of the 11 NOAEL/BMC ratios were 29, 5.9, and 2.9 (Allen et al., 1994). 12 13 The proportion of times that the NOAEL exceeded the BMC for the 1, 5, and 10% 14 response was 89, 65, and 42% for the probit analysis and 95, 80, and 54% for the Weibull 15 analysis. In all cases the LOAEL/BMC ratio exceeded 1 for the probit and Weibull analysis of 16 the 1 and 5% response but not always for the 10% response (99%). For this reason the BMCIO 17 may be too high a response rate to use to predict a NOAEL. In contrast the corresponding 1 and 18 5% response ratios were always greater than 1. 19 20 The ratios of the MLE/BMC were not great, ranging from a mean of 1.39 for a probit 21 analysis of the 10% response to 3.02 for a Weibull analysis of the 10% response. It is important 22 to note that using the probit analysis the LOAEL/MLE ratios were equal to or greater than 1 in 23 99, 94, and 71% of the cases for the 1, 5, and 10% responses. The MLE would probably be 24 protective at the 1% response level but not for the 5 and 10% response levels. Similar numbers 25 of 99, 97, and 76% were observed for the Weibull analysis. 26 27 The BMC approach can provide a more refined assessment of the prediction of the 28 empirical NOAEL. It must be emphasized that even the empirical NOAEL may represent a 29 response level which is not detected. When 5 to 10 animals are used in an experiment a 10 to 30 20% response can be missed (Leisenring and Ryan, 1992) and even a BMC,0 is similar to a 31 LOAEL with dichotomized data (Gaylor, 1996). It is expected that the BMC is less than the 32 empirical LOAEL. In the Fowles et al. (1999) analysis of the data the BMC05 and BMC0, values 33 were always below the empirical LOAEL for the studies analyzed. The probit analysis of the 34 data by Fowles et al. (1999) provided a better fit with the data as measured by the "chi-squared 35 goodness-of-fit test, mean width of confidence intervals, and number of data sets amenable to 36 analysis by the model." 37 38 It is interesting to note that the BMCOJ is very close to the MLEOI in the Fowles et al. 39 (1999) evaluation of inhalation acute toxicity data. Through 1999 the NAC/AEGL Committee 40 has used the MLE0, to estimate the highest exposure at which lethality is not likely to be 41 observed in a typical acute exposure study. Given the analysis by Fowles et al. (1999) and for the 42 above reasons, the NAC/AEGL Committee will generally use the BMC05 (lower 95% confidence Sop08-02 wpd Printed July 6. 2000 19 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30. 2000 1 limit (LCL) of the exposure required to produce a 5% response to exposure to chemicals) in the 2 future for this estimate although, the MLE0, will also be calculated and considered. This 3 incorporates the uncertainties due to the number of animals used in an experiment, the 4 experimental variability observed, utilizes all of the data and the slope of the exposure response 5 curve, and provides for a reasonable estimate of a predicted experimental NOAEL. In all cases 6 the MLE and BMC at specific response levels will be considered when setting AEGL levels. 7 Statistical models in addition to the log-probit will also be considered. Since goodness of fit tests 8 consider an average fit, they may not be valid predictors of the fit in the low exposure region of 9 interest. In this case the output of the different models will be plotted and compared visually 10 with the experimental data in selection of the most appropriate model. 11 12 It should be emphasized that these methodologies will generally be considered for an 13 acute lethal endpoint. Their use to set AEGL-1 and AEGL-2 levels will be considered on a 14 chemical-by-chemical basis. Different endpoints may require the use of different data sets in 15 different or the same species, a different benchmark dose approach, or identification of a 16 different response level. These factors will be considered for specific chemicals and 17 toxicological endpoints. 18 19 The preferred approach will be to use the BMC approach to identify the highest exposure 20 at which the toxicologic effects used to define an AEGL tier were not observed. If the data are 21 insufficient to use that approach then the level will be determined empirically from experimental 22 data. 23 24 2.2.2 SELECTION OF HEALTH EFFECTS ENDPOINTS FOR AEGL-1, 25 AEGL-2, AND AEGL-3 26 27 In addition to the working definitions of the three AEGL tiers, this section includes a 28 summary of the specific biologic endpoints used to establish the AEGL levels for individual 29 chemicals. Also included are general principals for selection of AEGL health effect endpoints 30 that have been derived from the Committee's selections on a chemical-by-chemical basis. Since 31 ideal data sets for certain chemicals are not available, extrapolation methods and the 32 Committee's scientific judgement are often employed to establish threshold values. In the 33 absence of adequate data, no AEGL value is established. The basis for this decision is the failure 34 to achieve a minimum two-thirds majority of a quorum of the Committee that is in favor of 35 establishing a value, or a formal decision by two-thirds of the Committee not to establish a value. 36 37 38 Under ideal circumstances the specific health effects would be identified that determine 39 each of the AEGL levels. A search of the published literature would be performed for data on 40 the chemical, and AEGL levels would be generated from that data. However, data relating 41 exposure and effect do not always follow an ideal paradigm and may lead to apparent 42 mconsistences in the use of endpoints to set AEGL levels. The general principles laid down in Sop08-02 wpd Printed July 6. 2000 20 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 the NRC (1993a) guidance for evaluating data and selecting appropriate health effects, combined 2 with professional judgement, are used to establish AEGL levels. From the evaluations of the first 3 5 chemicals in this publication, and experience with data sets on chemicals currently under 4 review, the following refinements to the NAS guidelines have been adopted by the NAC/AEGL 5 Committee to set AEGL levels. Following the guidelines are elements of the rationale to capture 6 in the Technical Support Document. 7 8 For the reasons discussed in the introduction to this section, the NAC/AEGL Committee 9 generally selects the highest experimental concentration that does not elicit the symptoms or 10 effects defined by the AEGL tier in question. This concentration represents the starting point for 11 AEGL development. In instances where appropriate data are available, the BMC methodology 12 may be considered and used to select the AEGL endpoints. 13 14 15 2.2.2.1 AEGL-1 Endpoints 16 17 The NRC 1993a guidelines discuss the definition of the AEGL-1 endpoint on pages 10, 18 12, and 21. Above the AEGL-1 level, discomfort becomes increasingly likely. Below the 19 AEGL-1 level (detectability) "Exposure insufficient to cause discomfort or adverse health effects 20 might be perceived nevertheless by means of smell, taste, or sensations (mild sensory irritation) 21 that are not uncomfortable. The awareness of exposure might lead to anxiety and complaints and 22 constitutes what is termed here detectability." (NRC, 1993a, p21). 23 24 Thus at concentrations below the AEGL-1 level there may be specific effects such as the 25 perception of a disagreeable odor, taste, or other sensations (mild sensory irritation). In some 26 people that could result in mild lacrimation or coughing. Since there is a continuum in which it 27 is difficult to judge the appearance of "discomfort" in animal studies and human experiences, the 28 NAC/AEGL Committee has used its best judgement on a case by case basis to arrive at 29 appropriate and reasonable AEGL-1 values. 30 31 One additional factor to consider is that the three tiers of AEGL values "...provide much 32 more information than a single value because the series indicates the slope of the dose-response 33 curve" (NRC, 1993a). If an accident occurs and people smell or otherwise "detect" a chemical, 34 the extent of the concentration range between the AEGL-1 and AEGL-2 levels provides 35 information and insight into the estimated margin of safety between a level of detection or mild 36 sensory irritation (AEGL-1) and a level that may impair escape or lead to a serious long-term or 37 irreversible health effect (AEGL-2). In cases where the biological criteria for the AEGL-1 value 38 would be close to, or exceed the AEGL-2 value, the conclusion is reached that it is "Not 39 Recommended" (NR) to develop AEGL-1 values. In these cases, "detectability" by itself would 40 indicate that a serious situation exists. In instances where the AEGL-1 level approaches or 41 exceeds the AEGL-2 level, it might erroneously be believed that people experiencing mild 42 irritation are not at risk when in fact they have been exposed to extremely hazardous or possibly Sop08-02 wpd Printed July 6. 2000 21 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 lethal concentrations. 2 3 Since a comparison of the AEGL-1 and AEGL-2 levels indicates the slope of the dose- 4 response curve which may be of value in emergency response, planning, or prevention, the 5 NAC/AEGL Committee also attempts to establish AEGL-1 endpomts for adverse effects that are 6 asymptomatic or non-sensory. Examples of such effects include significant (measurable) levels 7 of methemoglobin, elevated blood enzyme levels, or other biological markers related to exposure 8 to a specific chemical. By establishing an AEGL-1 value in these instances, important 9 information on the toxicological behavior of a specific chemical is available to emergency 10 responders and planners. 11 12 The following criteria have been used by the NAC/AEGL Committee to select 13 endpoints for use in setting the AEGL-1 values. 14 15 2.2.2.1.1 No Value Established - AEGL-1 Exceeds AEGL-2 16 17 1. What aspects of the chemical toxicity profile make it inadvisable to generate an 18 AEGL-1 value. 19 20 For example, the AEGL-1 value was not established because levels which are 21 "detectable" are close to, or exceed, an AEGL-2 level. These materials have poor warning 22 properties. 23 24 2.2.2.1.2 No Value Established - Insufficient Data 25 26 Insufficient data were available. 27 28 2.2.2.1.3 Highest Experimental Exposure Without an AEGL-1 Effect 29 30 1. State the species, effect, and concentration and exposure time to cause the effect. 31 2. Describe the toxicologic endpoint of concern. 32 33 The highest experimental exposure levels which did not cause sensory irritation, 34 pulmonary function, and narcosis in humans have been used to set AEGL-1 levels. 35 36 2.2.2.1.4 Effect Level for a Response 37 38 1. State the species, effect, and concentration and exposure time to cause the effect. 39 2. Describe the toxicologic endpoint of concern 40 41 For example, levels for odor detection in humans, mild sensory irritation, asymptomatic Sop08-02 wpd Printed July 6. 2000 22 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 or non-sensory effects such as methemoglobin formation (22%), and pulmonary function 2 (transient changes in clinically insignificant pulmonary functions of a sensitive individual) have 3 been used as AEGL-1 endpoints. 4 5 2.2.2.2 AEGL-2 Endpoints 6 7 NRC (1993a) discussed the AEGL-2 definition on pages 10,12, and 21. The AEGL-2 8 exposure level is the threshold between reversible effects which cause discomfort, and serious or 9 irreversible health effects or effects which impair escape. Above the AEGL-2 level there is an 10 increasing likelihood people may become disabled or are increasingly likely to experience serious 11 or irreversible health effects. "The term disability is used here to indicate the situation where 12 persons will require assistance or where the effects of exposure will be more severe or prolonged 13 without assistance." (NRC, 1993a, p21). In developing AEGL-2 levels the NAC/AEGL 14 Committee has defined a NOEL for serious or irreversible effects or effects which impair escape. 15 It must be emphasized that reversible clinical toxiciry may be observed below the AEGL-2 level. 16 If minor reversible effects are seen at one level of exposure and disabling effects at a higher 17 exposure, the former is used to set the AEGL-2 level. If the exposure associated with disabling 18 effects cannot be determined from experimental data, then the highest level causing reversible 19 effects/discomfort may be used to set the AEGL-2 level. 20 21 The following criteria have been used by the NAC/AEGL Committee to date to select 22 endpoints for use in setting the AEGL-2 values. 23 24 2.2.2.2.1 Highest Experimental Exposure Without an AEGL-2 Effect 25 26 1. State the species, effect, and concentration and exposure time to cause the effect. 27 2. Describe the toxicologic endpoint of concern. 28 29 The highest experimental exposure levels which did not cause decreased hematocrit, 30 kidney pathology, behavioral changes or lethality (effects observed at higher exposures were 31 above the definition for AEGL-2) have been used as the basis for determining AEGL-2 levels. 32 33 2.2.2.2.2 Effect Level for a Toxic Response Which was Not Incapacitating or 34 Not Irreversible 35 36 1. State the species, effect, and concentration and exposure time to cause the effect. 37 2. Describe the toxicologic endpoint of concern. 38 39 40 41 **- a-rwiawijiiyv ui^« LisyvJtsl/llsgll' 1*I1U£JWU1L Ul VU111/&111. For example, strong irritation, dyspnea, pulmonary function, provocation of asthma episodes, pathology (respiratory tract, mild narcosis, methemoglobin formation (41%) have been used to set AEGL-2 levels. Sop08-02 wpd Printed July 6, 2000 23 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2.2.2.2.3 A Fraction of the AEGL-3 Level 2 3 1. State the rationale for using a fraction of the AEGL-3. 4 2. State why the specific fraction chosen is scientifically justified. 5 6 In the absence of specific data used to determine an AEGL-2 value, 1/3 of the AEGL-3 7 value has been used to establish the AEGL-2 level. This approach can only be used if the data 8 indicate a steep exposure-response relationship from serious to no-effects. 9 10 2.2.2.3 AEGL-3 Endpoints 11 12 NRC, (1993a) discussed the AEGL-3 definition on pages 10, 12, and 21. The AEGL-3 13 tier is the threshold exposure level between serious long lasting or irreversible effects or effects 14 which impair escape and death or life-threatening effects. Above the AEGL-3 there is an 15 increasing likelihood of death or life threatening effects occurring. In determining AEGL-3 16 levels, the NAC/AEGL Committee defined the highest exposure which does not cause death or 17 life threatening effects. It must be emphasized that severe toxicity will be observed at the AEGL- 18 3 level. In cases where data to determine the highest exposure level which does not cause life- 19 threatening effects are not available, levels which cause severe toxicity without producing death 20 have been used. 21 22 The following criteria have been used by the NAC/AEGL Committee to date to 23 select endpoints for use in setting the AEGL-3 values. 24 25 2.2.2.3.1 Highest Exposure Level Which Does Not Cause Lethality - 26 Experimentally Observed Threshold (AEGL-3 NOEL) 27 28 1. State the species, effect, and concentration and exposure time to cause the effect. 29 2. Describe the toxicologic endpoint of concern. 30 31 Where experimental lethality data have been insufficient to statistically determine a 32 benchmark concentration, the highest experimental exposure which did not cause lethality in an 33 experiment in which death was observed was used to set the AEGL-3 level. 34 35 2.2.2.3.2 Highest Exposure Level Which Does Not Cause Lethality - Estimated 36 Lethality Threshold - 1/3 of the LC50 37 38 1. State the species, effect, and concentration and exposure time to cause the effect. 39 2. Describe the toxicologic endpoint of concern. 40 3. If an exposure which does not produce death is estimated by dividing an LC50 value by 41 3 (or some other divisor), give the slope of the exposure response curve or enough Sop08-02 wpd Printed July 6, 2000 24 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 data points to support the division by 3 (or some other divisor). 2 3 Where experimental lethality data have been insufficient to statistically determine an LC0, 4 value, but an LC50 value was determined, and all exposure levels caused lethality, a fraction of 5 the LC50 value was used to estimate the threshold for lethality. In all cases the exposure response 6 curve was steep and the LCSO value was divided by three. The Fowles et al. (1999) analysis of 7 inhalation toxicity experiments revealed that for many chemicals, the ratio between the LCSO and 8 the experimentally observed non-lethal level was on average a factor of approximately 2, the 90th 9 percentile was 2.9, and the 95th percentile was 3.5. There was a range of ratios from 1.1 to 6.5. 10 11 2.2.2.3.3 Highest Exposure Level Which Does Not Cause Lethality - 12 Benchmark Exposure Calculation of the 5 % and 1% Response 13 14 1. State the species, effect, and concentration and exposure time to cause the effect. 15 2. Descnbe the toxicologic endpoint of concern. 16 3. State the statistical methodology used to derive a BMC05 and the MLE01. 17 18 Where sufficient information was available, the preferred method through 1999 was a 19 probit analysis (Finney, 1971) to determine the LC01. Actual calculations were performed using 20 the Number Cruncher Statistical System - Version 5.5. This is a probit analysis of the response - 21 log exposure curve. The Maximum Likelihood Estimate (MLE) was used for the LC0, value. 22 The method of Litchfield and Wilcoxon (1948) has also been used. 23 24 In the future both the BMC05 and MLE01 for lethality will be determined, presented and 25 discussed. Results from the above models will be compared with the log probit U. S. EPA 26 (2000) Benchmark Dose Software (http://www.epa.gov/ncea/bmds.htm). In all cases the MLE 27 and BMC at specific response levels will be considered. Other statistical models such as the 28 Weibull may also be considered. Since goodness of fit tests consider an average fit, they may not 29 be valid predictors of the fit in the low exposure region of interest. In this case the output of the 30 different models will be plotted and compared visually with the experimental data to determine 31 the most appropriate model. The methodology which results in values consistent with the 32 experimental data and the shape of the exposure-response curve will be selected for AEGL 33 derivations. 34 35 Because of uncertainties that may be associated with extrapolations beyond the 36 experimental data range, the estimated values are compared with the empmcal data. Estimated 37 data which conflicts with the empirical data will generally not be used. 38 39 2.2.2.3.4 Effect Level for a Response 40 41 1. State the species, effect, and concentration and exposure time to cause the effect. 42 2. Describe the toxicologic endpoint of concern. Sop08-02 wpd Printed July 6, 2000 25 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Where the data were insufficient to estimate the highest exposure which does not cause 2 lethality, exposures which caused severe intoxication in the absence of lethality were used in the 3 selection of exposure levels to set AEGL-3 values. The endpoints of concern included decreased 4 hematocrit, methemoglobin formation (70-80%), cardiac pathology, and severe respiratory 5 pathology. 6 Sop08-02 wpd Printed July 6. 2000 26 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2.3 GUIDELINES/CRITERIA FOR THE SEARCH STRATEGY, 2 EVALUATION, SELECTION AND DOCUMENTATION OF KEY DATA 3 AND SUPPORTING DATA USED FOR THE DERIVATION OF AEGL 4 VALUES 5 6 2.3.1 Search Strategy 7 8 The literature search strategy focuses on three general sources of information: (1) 9 electronic databases, primarily peer-reviewed journals and government databases, (2) published 10 books and documents from the public and pnvate sectors of the U. S. and foreign countries, 11 including references on toxicology, regulatory initiatives, and general chemical information; (3) 12 data from private industry on other pnvate sector organizations. The search strategy also 13 includes the use of search terms to enhance the relevance of the electronic databases identified 14 and retrieved. 15 16 17 (1) ELECTRONIC DATABASE COVERAGE 18 19 The following databases are searched: 20 21 TOXLINE database (1981 - Current) from U. S. National Library Medicine's TOXNET: 22 TOXLINE covers the toxicological effects of chemicals, drugs and physical agents on 23 living systems. Among the areas covered are adverse drug reactions, carcmogenesis, 24 mutagenesis, developmental and reproductive toxicology, environmental pollution and food 25 contamination. 26 27 TOXLINE65 database (1965-1980) 28 Subject coverage is identical to TOXLINE, for tune periods that precede that of 29 TOXLINE. 30 31 HAZARDOUS SUBSTANCES DATA BANK (HSDB) (Current) from TOXNET: 32 HSDB is a comprehensive factual and numeric chemical profile. Each chemical profile is 33 peer reviewed for completeness and accuracy to reflect what is known the about the chemical. 34 35 PUBLIC MEDLINE (PUBMED): 36 PUBMED includes MEDLINE and PREMEDLINE. MEDLINE, the U. S. National 37 Library of Medicine's (NLM) premier bibliographic database covers medicine, nursing, dentistry, 38 veterinary medicine, health care systems, and the preclimcal sciences The above-mentioned 39 TOXLINE searches include MEDLINE citations. PREMEDLINE, also produced by NLM, 40 provides citation and abstract information before full records are added to MEDLINE. For a 41 short period of time, this information is only available in PUBMED. SopOS-02 wpd Printed July 6. 2000 27 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 REGISTRY OF TOXIC EFFECTS OF CHEMICAL SUBSTANCES (RTECS). 2 RTECS, compiled by NIOSH (U. S. National Institute of Safety and Health), is a 3 comprehensive database of basic toxicity information and toxic-effects data on more than 4 100,000 chemicals. 5 6 U. S. NATIONAL TECHNICAL INFORMATION SERVICE (NTIS) 7 The NTIS database provides access to the results of US government-sponsored research, 8 development and engineering, plus analyses prepared by federal agencies, their contractors, or 9 grantees. It is a means through which unclassified, publicly available, unlimited distribution 10 reports are made available from such U. S. agencies as NASA, DDC, DOE, HUD, DOT and 11 some 600 other agencies. In addition, some state and local government agencies contribute their 12 reports to the database. NTIS also provides access to the results of government-sponsored 13 research and development from other countries. 14 15 U.S. INTEGRATED RISK INFORMATION SYSTEM (IRIS) 16 Data from US EPA in support of human health risk assessment, focusing on hazard 17 identification and dose-response assessment for specific chemicals. 18 19 U.S. FEDERAL RESEARCH EN PROGRESS (FEDRIP) 20 FEDRIP provides access to information about ongoing U.S. government funded research 21 projects in the fields of physical sciences, engineenng, and life sciences. 22 23 U. S. DEFENSE TECHNICAL INFORMATION CENTER (DTIC) 24 The central U. S. Department of Defense facility for access to scientific and technical 25 information. The DTIC database includes technical reports, independent research and 26 development summaries, technology transfer information, and research and development 27 descriptive summaries. The scope of the DTIC collection includes areas normally associated 28 with Defense research such as military sciences, aeronautics, missile technology, and nuclear 29 science. The collection also includes information on biology, chemistry, environmental sciences, 30 and engineenng. 31 32 U S. ORNL IN-HOUSE DATABASES 33 34 CHEMICAL UNIT RECORD ESTIMATES (CURE) 35 The CURE database contains selected information from the U.S. Environmental 36 Protection Agency Office of Health and Environmental Assessment documents 37 and Carcinogen Risk Assessment Verification Effort (CRAVE) and Reference 38 Dose (RfD) work groups. Although the groups are not currently active, this 39 database is a valuable compilation of historic information. 40 41 TOXICOLOGY AND RISK ANALYSIS (TARA) DOCUMENT LIST 42 This database lists all types of documents written by TARA staff over the past Sop08-02 wpd Printed July 5, 2000 28 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 fifteen years. These range from Toxicity Summaries to journal articles. This list 2 provides good references for chemicals which overlap the AEGL priority list 3 4 (2) PUBLISHED BOOKS AND DOCUMENTS FROM THE PUBLIC AND PRIVATE 5 SECTORS 6 7 GENERAL REFERENCES FOR TOXICOLOGY AND CHEMICAL INFORMATION 8 9 U. S. ATSDR (Agency for Toxic Substances and Disease Registry) Toxicological 10 Profiles. 11 Chemfmder, Chemical Searching and Information Integration by CambridgeSoft 12 Corporation 13 Current Contents, Life Sciences edition 14 HEAST (Health Effects Assessment Summary Tables) 15 Kirk-Othmer Encyclopedia of Chemical Technology 16 IARC (International Agency for Research on Cancer) Monographs on the Evaluation of 17 the Carcinogenic Risk of Chemicals to Humans 18 Low-dose Extrapolation of Cancer Risks, S. Olin, et al. (editors) 19 Merck Index 20 U. S. NTP (National Toxicology Program) Div. of Toxicology Research and Testing, 21 published reports. 22 Patty's Industrial Hygiene and Toxicology 23 Respiratory System, Monographs on the Pathology of Laboratory Animals, T.C. Jones, et 24 al. (editors) 25 Synthetic Organic Chemicals, U.S. International Trade Commission 26 Toxicology of the Nasal Passages, C.S. Barrow (editor) 27 U.S. Air Force Installation Restoration Program Toxicology Guide 28 29 GENERAL REFERENCES FOR REGULATORY INFORMATION AND STANDARDS 30 31 U. S. AIHA (American Industrial Hygiene Association) Emergency Response Planning 32 Guidelines 33 (ERPGs) and Workplace Exposure Level Guides (WEELs) 34 U. S. ACGIH (American Conference of Government and Industrial Hygienists) Threshold 35 Limit 36 Values for Chemical Substances and Physical Agents and Biological Exposure Indices 3 7 ACGIH Documentation of Threshold Limit Values 38 U. S. NAAQS National Ambient Air Quality Standards 39 U.S. NIOSH Documentation of IDLH's 40 U. S NIOSH (National Institute for Occupational Safety and Health) Pocket Guide to 41 Chemical Hazards 42 U. S. NIOSH Recommendations for Occupational Safety and Health, Compendium of Sop08-02 wpd Printed July 6, 2000 29 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Policy Documents and Statements 2 U. S. OSHA (Occupational Safety and Health Administration) Limits for Air 3 Contaminants 4 U. S. SMACS Spacecraft Maximum Allowable Concentrations for Selected Airborne 5 Contaminants, Committee on Toxicology, Commission on Life Sciences, and 6 National Research Council, sponsored by NAS 7 U. S. EPA Health Effects Documents 8 9 (3) UNPUBLISHED DATA FROM PRIVATE INDUSTRY AND OTHER PRIVATE 10 SECTOR ORGANIZATIONS OF ALL NATIONS 11 12 Reports and data not published in peer reviewed scientific journals that is relevant to the 13 development of AEGLs. Most often this represents acute toxicity data from controlled inhalation 14 exposure studies available from private industry or other organizations in the private sector of all 15 nations that may or may not be published in a peer reviewed journal at some later date. 16 17 SEARCH TERMS 18 19 The U. S. Chemical Abstract Services (CAS) Registry number of the chemical is used as 20 the first choice. Chemical nomenclature or common chemical names and synonyms are used if 21 the CAS Registry number is unknown. 22 23 The CAS Registry number alone is used as the first step. If there are approximately 300 24 citations, then all are retrieved for review. If less than approximately 300 references are found, 25 conduct searches using chemical nomenclature and common chemical name(s) in addition to the 26 CAS number. Searches by chemical name(s) also should be made if limited data of high quality 27 are found, irrespective of the number of citations found. 28 29 If more than 300 citations are found using any form of chemical identification, the 30 references may be enriched in relevance and quality by adding any number of the following 31 characterizations of the desired data to the search strategy: 32 33 short-term 34 threshold limit 35 permissible exposure 36 acute 37 ocular terms 38 inhalation terms 39 dermal terms 40 41 If the number or quality of single exposure toxicity studies found is not deemed to be 42 adequate, multiple exposure studies may be considered but might not achieve a consensus of the Sop08-02 wpd Printed July B. 2000 30 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 NAC/AEGL Committee. If a consensus or 2/3 majority of the Committee cannot agree on the 2 adequacy of the data, the chemical may be placed in a cue for future acute toxicity testing. 3 4 2.3.2 Evaluation, Selection and Documentation of Key and Supporting Data 5 6 As a detailed interpretation and supplementation of the U. S. NAS (NAS, 1993a) 7 guidelines, U. S. EPA's National Advisory Committee on Acute Exposure Guideline Levels 8 (NAC/AEGL Committee) has developed guidelines for evaluating the quality of studies to be 9 used in the calculation of proposed AEGL values. The proposed evaluation and documentation 10 procedure created by the AEGL Committee is intended to provide technical support document 11 (TSD) writers, reviewers, committee members, interested parties and the public with a clear and 12 consistent list of elements that must be considered in their evaluations. The proposed evaluation 13 and documentation system will add technical validity and administrative credibility to the process 14 by providing a transparent, logical and consistent method for selecting key studies used to 15 calculate an AEGL value. Additionally, the system will allow linkage to uncertainty factors and 16 modifying factors in a consistent and logical manner. The process is designed to allow 17 maximum flexibility in professional judgment while promoting scientific uniformity and 18 consistency and providing a sound administrative foundation from which Committee members 19 can function. The NAC/AEGL Committee has the concurrence of the U. S. National Academy 20 of Sciences (NAS) on these guidelines, as well as all other guidelines published in this manual. 21 22 Many toxicology studies used in the development of an AEGL were not designed to meet 23 current regulatory guidelines and are not necessarily consistent in protocol or scientific 24 methodology. As a result, these valuable investigations cannot be judged solely on the basis of 25 currently accepted experimental design criteria for such studies. Current U. S. EPA and OECD 26 guidelines are used as the basis for future studies conducted on behalf of the NAC/AEGL 27 Committee, but lack of consistency of older studies requires evaluation and qualification of each 28 data set for scientific validity within the context of AEGL documentation. A study can be 29 valuable in the derivation of AEGL values without conforming completely to a standard of 30 detailed methodology, data analysis and results reporting. The aim of the subject procedure is to 31 provide specific criteria in the selection and use of specific data sets for development of 32 defensible values, yet retain the ability to use logical scientific thinking and competent 33 professional judgment in the data selection process. If a study or some portion of a study 1) uses 34 scientifically valid methods, 2) contains adequate and reliable data and 3) presents defensible 35 conclusions relevant to the AEGL process, it may be included in the technical support document 36 and used to support the AEGLs. 37 38 It is important to emphasize that only toxicity data obtained directly from a primary 39 reference source is used as the basis for "key" toxicity studies from which the AEGL values are 40 derived. Additionally, all supporting data and information important to the derivation of an 41 AEGL value is obtained solely from the primary references. This includes data used to provide a 42 "weight-of-evidence" rationale in support of the AEGL value derived. Secondary references may Sop08-02 wpd Printed July 6, 2000 31 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 be used to provide data and information on commercial uses, production volumes, chemical and 2 physical properties and other non-toxicological or epidemiological information on a chemical. 3 Secondary references also may be used to present background information on the toxicity or 4 toxicological characteristics of a chemical and any other information not important or directly 5 relevant to the actual derivation of, or the supporting rationale for, the AEGL values. Finally, 6 data and information from secondary references should not be included in data summary tables 7 presented in the Technical Support Documents. 8 9 The evaluation guidelines are more credible if they are drawn from a widely accepted 10 prescription for study protocol design. The list of guidelines for AEGL study evaluation should 11 be based upon the scientific methodologies, but not be so restrictive that it precludes competent 12 professional judgment. Current Good Laboratory Practice (GLP) guidelines provide a basis for 13 selection of a robust list of study elements that, in concert with the professional experience and 14 judgment of the AEGL Development Team and NAC/AEGL Committee members in general, are 15 used to qualify the data which support the AEGLs. Consequently the NAC/AEGL Committee 16 has used NRC (1993a), the OECD's Guidelines for the Testing of Chemicals, and U. S. EPA 17 (Health Effects Test Guidelines) as a basis for selection. 18 19 The NAS (1993a) guidance provides only limited guidance on the use of toxicological 20 data from routes of exposure other than inhalation. The guidance states that the bioavai lability 21 and differences in the pharmacokinetics from the different exposure routes of the chemical in 22 question must be considered. Because of these complex biological phenomena and the paucity of 23 data to enable credible evaluation and consideration, the NAC/AEGL Committee to date has 24 selected and used only inhalation toxicity data to derive AEGL values. Further, toxicity data 25 from alternate routes of exposure will not be included in discussions in the Technical Support 26 Documents unless it is considered important for the support of relevant pharmacokinetics or 27 metabolism data or mechanisms and observed effects of toxicity. In the absence of inhalation 28 data to derive an AEGL value, the NAC/AEGL Committee may use toxicity data from other 29 exposure routes if there are adequate data to perform scientifically credible route-to-route 30 extrapolations. In the absence of acceptable data, the Committee will refer the chemical for 31 toxicity testing. 32 33 Each key and supporting study is evaluated using all listed Elements for Evaluation as 34 guidance. A "Key Study" is defined as the human and/or animal study from which a 35 toxicological value is obtained for use in AEGL calculations. "Supporting Studies" are the 36 human and/or animal studies which are used to support the toxicological findings and values 37 obtained from the Key Study and their use is consistent with the "weight-of-evidence" approach 38 to scientific credibility. While all Elements for Evaluation listed below are considered when 39 evaluating a study, only Elements for Evaluation from key and supporting studies which are 40 relevant to the derivation of the AEGL values will be discussed in the TSD as they impact the 41 derivation. In evaluating a study, a variety of measurement endpoints are preferred. However, a 42 study measuring, for example, only one endpoint may be selected for development of an AEGL if Sop08-02 wpd Printed July 6. 2000 32 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 other studies have shown that other known inhalation toxicology endpoints are less sensitive, 2 provided the data are considered to be reliable. The list of Elements for Evaluation also is used 3 for initial review of all studies evaluated for possible inclusion in the TSD in instances where 4 they are germane to the selection of studies. 5 6 The NAC/AEGL Committee is dependent upon existing human studies published in the 7 literature for data on humans Many of these studies do not necessarily follow current guidelines 8 on ethical standards which require that effective, documented, informed consent from 9 participating humans subjects be required. Further, recent studies which followed such 10 guidelines may not include that fact in the publication. Although human data may be important 11 in deriving AEGL values that protect the general public, utmost care must be exercised to insure 12 first of all that such data have been developed in accordance with ethical standards. No data on 13 humans known to be obtained through force, coercion, misrepresentation, or any other such 14 means will be used in the development of AEGLs. The NAC/AEGL Committee will use its best 15 judgement to determine whether the human studies were ethically conducted and that the human 16 subjects were likely to have provided their informed consent. Additionally, human data from 17 epidemiological studies and chemical accidents may be used. However, in all instances 18 described here, only human data, documents and records will be used from sources that are 19 publicly available or if the information is recorded by the investigator in such a manner that 20 subjects cannot be identified directly or indirectly. These restrictions on the use of human data 21 are consistent with the Common Rule as published in the Code of Federal Regulations (40 CFR 22 Part 26 [The Common Rule], 2000). 23 24 In addition to the discussion of the Elements for Evaluation in the individual studies 25 section of the Technical Support Document (TSD), a section entitled "Data Adequacy and 26 Research Needs" is included in the text of the TSD. A summary of the data adequacy discussion 27 is also included in the Derivation Summary Tables in the appendix of the TSD and in the 28 Executive Summary of the TSD. The text of the TSD relates the studies used to derive, or 29 support the derivation of, the AEGL values to the discussion of the adequacy of the available 30 data. Brief summaries of this discussion are included in the Executive Summary and Derivation 31 Summary Tables. The data adequacy section also presents and integrates the weight-of-evidence 32 by considering all information as a whole for each AEGL developed. In addition to considering 33 the Elements for Evaluation as relevant in the discussion, a number of other factors must be 34 considered. These include repeatability of experiments between laboratories, consistency of data 35 between experiments and laboratories, types and number of species tested, variability of results 36 between species, and comparison of AEGL values with the valid human and animal data. Every 37 data set is a unique, chemical-specific source of information which reflects the investigations 38 conducted on the chemical and the properties of the chemical. This section reflects a "best 39 professional judgement" approach in the evaluation of the data adequacy and future research 40 needs. 41 42 A diagram of the decision process for the selection of key studies and supporting studies Sop08-02 wpd Printed July 6, 2000 33 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 is shown on the following page. A summary of the elements or criteria used to select key studies 2 and supporting studies, and to evaluate their adequacy in deriving AEGL values follows. 3 Sop08-02 wpd Printed July 6. 2000 34 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 FIGURE 2.3-1 ALLOCATION OF STUDY REPORTS DECISION TREE ALLOCATION OF STUDY REPORTS DECISION TREE Non-published, Non-peer Reviewed Industry Data Published Literature Search Other Data / Information Sources Special Toxicity Studies \t V Identification and Selection of Relevant Data / Information Evaluate Data and Information Retain Incorporation in Whole or in Part into TSD? Yes _L Consider for AEGL Derivation? Reject based on major _ deficiencies in specific selection criteria No Does not represent the best key or supporting studies ~ based on adherence to specific selection criteria No Good supporting data _ 'but does not represent the best key studies Not Included in TSD Background Information Supporting Study Yes Selected as Key Study Derivation of one or more Acute Exposure Guideline Levels 2 3 4 Sop08-02 wpd Printed July 6. 2000 35 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Elements for the Evaluation of Key and Supporting Data and Studies 2 3 1. Only toxicity data and information obtained directly from a primary reference source may be 4 used as the basis for "key" lexicological studies. All other studies important to the 5 derivation of an AJEGL value, or that serve as a "weight-of-evidence" rationale are 6 obtained from a primary source. 7 8 2. Secondary references may be used for non-toxicological data such as physical/chemical 9 properties, production locations, quantities and background information on the toxicity of 10 a chemical, provided the information is not directly used in the derivation of the AEGL 11 values. 12 13 3. Only human data from studies that meet the ethical standards discussed in the Evaluation, 14 Selection and Documentation of Key and Supporting Data section of this SOP Manual 15 will be used in the derivation of AEGL values. 16 17 4. Route of exposure. The inhalation route is preferred. Where the endpomt of concern is 18 systemic intoxication and the first pass effect is not significant, oral exposure may be 19 considered. In the absence of scientifically sound data with high confidence in a valid 20 route-to-route extrapolation, routes of exposure other than inhalation will not be used for 21 AEGL derivation. 22 23 5. Scientifically credible exposure concentration and exposure duration are provided. 24 25 6. Analytical procedures used to determine chamber concentration for inhalation exposure in 26 controlled studies and detailed, scientifically credible methods, procedures, and data used 27 to measure chemical concentration in epidemiological or anecdotal cases (accidental 28 chemical releases). For oral exposure, dose may be determined from the amount of test 29 chemical placed into the subject. 30 31 7. Number of subjects. The number is not rigid; e.g., a general rule uses 5-10 rodents/sex/group 32 as a valid measure, but as few as 2-3 primates or dogs/sex/group may be used. The 33 acceptable number of subjects per group is influenced by the relationship between the 34 within group variability and the degree of change that is considered to be detrimental. 35 Smaller numbers per group may be acceptable by increasing the number of treatment 36 groups. 37 38 8. Species studied. Humans are most relevant. Rats, mice, rabbits, guinea pigs, ferrets, dogs or 39 monkeys are acceptable. Other species require evaluation on a case-by-case basis. It is 40 important to use a species for which there are historical control data and relevance to 41 humans. 42 Sop08-02 wpd Printed July 6, 2000 36 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 9. Presence of a concurrent control group composed of the same species as that in the treatment 2 groups. The control subjects should be housed and cared for in the same manner as 3 exposed animals. 4 5 10. Concentration/dose selection that establishes a clear dose-response relationship. 6 7 11. Observation period. The period is variable based on the time of onset of the toxic effect. If 8 it is rapid (minutes to 2-3 hours) and associated with quick recovery, an observation 9 penod of 3-4 days may be sufficient. For effects that are slow in onset (2-3 days) and 10 delayed in time, a minimum observation period of 14 days is recommended. 11 12 12. Signs and symptoms of intoxication noted during and after exposure and reported separately 13 by sex and concentration or dose. 14 15 13. For animal studies, body weights should recorded throughout the study. 16 17 14. For repeated concentration/dose studies, establishment of the highest estimated or 18 experimental (empirical) level of no effect for the specific AEGL endpoint of concern. 19 20 15. Toxicity data from routes of exposure other than inhalation generally will not be used as key 21 or supporting data. Data from alternate routes are considered in the absence of inhalation 22 data if sufficient data are available to perform a credible route-to-route extrapolation. 23 24 16. Number of concentrations or doses used. 25 26 17. If a NOEL is selected or derived as the endpoint for an AEGL seventy level of concern, 27 identifying both the highest dose at which the effect is not seen, and the lowest dose at 28 which it is seen, for each AEGL severity level strengthens the confidence in the study. 29 30 18. Record of time of death if applicable. 31 32 19. For animal studies, necropsy conducted with at least gross effects noted. 33 34 20. As available, data (e.g. histopathological changes, clinical chemistry and hematology) may 35 reduce uncertainty. 36 37 21 Recovery group included in the study and data generated are sufficient to determine the 38 degree of reversibility. 39 40 22. Statistical treatment of data generated from study. 41 42 23. An evaluation of all relevant data should be performed and summarized in the Technical Sop08-02 wpd Printed July 6. 2000 3 7 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Support Document in order to present an integrated "weight-of-evidence" picture for all 2 information considered as a whole. 3 4 2.3.3 Elements for Discussion on Data Adequacy and Research Needs 5 6 The adequacy of the key and supporting data selected for AEGL derivation should be 7 discussed in Section 8.3 of the TSD (Data Adequacy and Research Needs) Because of the 8 different toxic endpoints used for the three AEGL tiers and the use of different data and/or 9 studies for each tier, the data adequacy should be addressed separately for AEGL-1, -2, and -3 10 In addition to any discussion regarding the elements for evaluating key and supporting studies 11 listed in this section of the TSD, the discussion should consider in general terms: (1) repeatability 12 of experiments between laboratories, (2) consistency of data between experiments and 13 laboratories, (3) types and number of species tested, and, (4) comparisons of the AEGLs with 14 valid human and animal data. 15 16 A summary of the discussion in the TSD section "Data Adequacy and Research Needs" 17 also should be included in the Executive Summary and the Derivation Summary Tables. The 18 summary statements also should address the adequacy of the data by AEGL tier. 19 20 Sop08-02 wpd Printed July 6, 2000 3 8 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2.4 DOSIMETRY CORRECTIONS FROM ANIMAL TO HUMAN 2 EXPOSURES 3 4 When extrapolating from observed responses in animals to predicted human responses, 5 the relationship between nominal exposure concentration and delivered dose to the target tissue 6 is often an issue of concern. For inhaled toxicants the target tissue is either some component of 7 the respiratory system and/or other tissue or organ. A number of methods have been proposed to 8 adjust for differences in the dose to target tissue in the respiratory system (U.S. EPA, 1994b) and 9 those located systemically (U. S. EPA, 1994b; NRC, 1993a). The concern has been the lack of 10 validated methodologies that would provide scientifically sound values for gases, vapors and 11 aerosols. This is particularly true where the methodology may predict levels for humans that may 12 not be sufficiently protective. Both methodologies referenced above have not been validated for 13 gases with experimental data, especially in the higher dose ranges required to produce toxicity 14 with acute exposures. Another possible dosimetry correction, using the inhaled dose against the 15 body weight raised to the 3/4 power has support based upon an analysis of chronic toxicity 16 studies (U.S. EPA, 1992). However, this adjustment may not be relevant for acute lethality 17 studies (Wolff and Rhomberg, 1998). Therefore, no dosimetry adjustments have been made to 18 date by the NAC/AEGL Committee for attaining human-equivalent doses in the development of 19 AEGLs for gases, vapors and aerosols. 20 21 If AEGL values are developed for particulates, the methodology developed by the U. S. 22 Environmental Protection Agency, and validated with experimental data on particulate matter, 23 will be reviewed and applied on the basis of the individual material (U. S. EPA, 1994b). Where 24 specific data and validated models are available for chemicals inhaled as gases, a dosimetry 25 correction will be made by the NAC/AEGL Committee. 26 27 28 2.4.1 Discussion of Potential Dosimetry Correction Methodologies for Gases 29 30 2.4.1.1 The Respiratory System as a Target Organ 31 32 The RfC (Reference Concentration) methodology for chronic exposure to gases was 33 proposed by U.S. EPA (1994b) as an approach to the dosimetry correction for effects on the 34 respiratory system. This method has not been used by the NAC/AEGL Committee for the 35 following reasons: (1) The RfC dosimetry corrections from animal to man are based upon 36 theoretical constructs which have not been confirmed and validated with experimental data; (2) 37 Some of the RfC assumptions are questionable and can have a significant impact upon the 38 calculated dosimetry correction between animal and human. Below is a discussion of two key 39 examples and their impact upon the dosimetry adjustment. The assumptions are the requirement 40 of uniform deposition in compartments and equivalent percent of deposition in animals and 41 humans 42 Sop08-02 wpd Printed July 6, 2000 3 9 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 For Category 1 gases (highly water soluble and/or rapidly irreversibly reactive) the RfC 2 methodology assumes that for each respiratory compartment (extrathoracic, tracheobroncial, and 3 pulmonary), the deposition of chemical is equivalent throughout the compartment. This fails to 4 take into account major differences in anatomical structure and deposition (dose) as the gas, 5 vapor or aerosol progresses from proximal to distal regions within any one compartment. The 6 dosimetric adjustment from rodent to man for the extrathoracic region predicts a 5-fold higher 7 delivered dose to humans compared to rodents at equivalent exposures. However, a number of 8 investigators have shown that treating the entire extrathoracic region as a single homogeneous 9 compartment is incorrect. The use of sophisticated computational fluid dynamics computer 10 modeling, correlated with analysis of patterns of lesions induced by chemical exposure, 11 demonstrate that the degree of deposition of chemicals vanes greatly in different extrathoracic 12 regions in rats (Kimbell et al., 1993; Kimbell et al., 1997a; Kimbell et al., 1997b) and the 13 monkey (Kepler et al., 1998). Specific areas such as the olfactory epithelium will receive 14 different regional doses in the rat and humans because of differences in surface area, susceptible 15 location, and degree of ventilation (Frederick, et al., 1998). A recent estimate of a dosimetnc 16 adjustment for vinyl acetate toxicity to the olfactory epithelium was performed using multiple 17 compartments and a physiologically based pharmacokinetic model (PBPK). Bogdanffy et al. 18 (1999) predicted that a time adjusted exposure of 8.7 ppm in the rat would result in the same 19 damage in a human exposed to 10 ppm. In this case the application of the RfC methodology 20 overestimates the risk to humans. 21 22 In the RfC methodology the proportion deposited in each region for Category 1 gases is 23 assumed to be the same in animals and humans. Where the deposition is less than 100% this 24 assumption is incorrect when one considers a rodent breathing at 100 times a minute vs 15 25 breaths a minute for a human. The residence time for the chemical in a rodent lung is 26 approximately 0.6 seconds while it is approximately 4 seconds in a human or about 6 times as 27 long. All things being equal, the longer residence time in the human respiratory system will 28 mean that the human extracts a greater percent of inspired chemical per breath than a rodent. 29 Another factor to consider is that at high exposure levels, a steady state can be rapidly achieved 30 in which relatively little chemical is deposited in each breath so that the concentration becomes 31 the determining factor. 32 33 Of concern is the fact that when dosimetry adjustments are made between rodents and 34 humans for toxicity to the pulmonary region, the delivered dose to the human is predicted to be 35 about 3-times less than the mouse for an equivalent nominal exposure concentration. Using this 36 methodology in the absence of supporting empirical data could seriously underestimate human 37 sensitivity. For example, at lethal concentrations fluorine toxicity is due to pulmonary 38 intoxication in all species tested (Keplinger and Suissa, 1968) Further, the empirically derived 39 LC50 values for the mouse, rat, rabbit, and guinea pig are essentially identical. However, the 40 minute volume to surface area ratio for the pulmonary region of the guinea pig closely resembles 41 the human. If the RfC dosimetry procedure were correct, the LC50 for the guinea pig should be 2- 42 3 times higher than that observed for the rat and mouse, yet the empirical data were essentially Sop08-02 wpd Printed July 6, 2000 40 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 identical for all three species. Using the RfC methodology to extrapolate a dosimetric correction 2 to humans in this case would seriously underestimate the nsk by a factor of 3 from the mouse 3 data. This problem is compounded by the fact that the RfC methodology calls for the use of a 4 lower interspecies uncertainty factor when the dosimetry correction is used. 5 6 2.4.1.2 Systemic Toxicity 7 8 Most systemic toxicants would fall under the definition of a Category 2 gas in the EPA 9 methodology (U.S. EPA, 1994b). Category 2 gases are moderately water soluble and 10 intermediate in their reactivity such that they would be distributed throughout the respiratory tract 11 and absorbed readily into the blood stream. In the case of Category 2 gases, the RfC dosimetry 12 procedure predicts that the human receives a dose ranging from 6,000 to 50,000 times higher 13 than a rodent (depending upon the species) for an equivalent exposure. These numbers do not 14 appear to be biologically reasonable or scientifically credible. Because of the potential errors, the 15 methodology for category 2 gases has not been used. When a corrected methodology is 16 published it will be evaluated for use by the NAC/AEGL Committee. 17 18 For systemic toxicants, the NRC (1993a), proposed that dosimetry correction be 19 conducted by adjusting for minute volume to body weight ratios. It is assumed for this 20 calculation that 100 percent of the chemical, or that equal percentages of the chemical, are 21 absorbed. Given that assumption, the correction is a reasonable approach and may be valid for 22 low concentrations of chemicals. Most animal to human extrapolation is done using mouse or rat 23 data. Using certain typical minute volume and body weight parameters, it is possible to calculate 24 an adjustment factor or multiplier in order to derive an equivalent dose in a human from animal 25 data. The multiplier is approximately 6 for the mouse and 3.5 for the rat. Thus, if the exposure 26 of interest in mice or rats is 100 ppm, then an equivalent internal dose in humans would be 27 predicted to be induced by exposure to 600 ppm and 350 ppm from these two species 28 respectively. Therefore, in order to induce an acutely toxic systemic effect in humans, people 29 would have to be exposed to a concentration 6 times greater and 3.5 times greater than the 30 nominal exposure required to induce the effect in mice or rats respectively. 31 32 If, on the other hand, less than 100 percent of the inspired chemical is absorbed with each 33 breath, the human and animal would absorb a different fraction of the chemical in each minute 34 (see discussion above). As the percent absorbed approaches 0 the multiplier would approach 1. 35 In the example above the multiplier for human dosimetry correction would go from 6 to 1 in the 36 case of mice and 3.5 to 1 in the case of rats as the percent absorbed approaches 0. 37 38 AEGL-2 and AEGL-3 levels represent relatively high exposure concentrations where 39 absorption may not be complete. If the minute volume to body weight correction for dosimetry 40 which assumes 100 percent absorption were used in these cases, the estimated human exposure 41 equivalent to the rodent would be too high, leading to an underestimate of the toxicity and the 42 derivation of AEGL values that are not protective to the human population Sop08-02 wpd Printed July 6. 2000 41 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Another approach to dosimetry correction might be that used by the U.S. Environmental 2 Protection Agency when extrapolating from animal cancer bioassays to theoretical excess human 3 cancer risk levels for lifetime exposures (U.S. EPA, 1992) The cross species scaling factor used 4 is based upon an equivalence of mg/kg^/day. There is reasonable scientific support for utilizing 5 this approach based upon an analysis of a number of multiple exposure studies across a number 6 of animal species (U.S. EPA, 1992) One might assume that the total amount of chemical 7 inhaled is equivalent to the dose (NRC, 1993a) and adjust that across species using the 8 equivalence of mg/kg^/day. However, Vocci and Farber (1988) point out the power law of 9 (body weight)3'4 holds for the ventilation rate such that on a weight to weight basis, the rat 10 receives about 4 times the delivered dose of a human for the same exposure concentration. 11 When this adjustment for breathing rate is combined with the adjustment for toxicity (U.S. EPA, 12 1992), the two cancel each other out and one is left with the conclusion that equivalent exposure 13 concentrations result in equivalent outcomes in animals and humans. 14 15 The situation is further complicated by an analysis of oral acute toxicity experiments by 16 Rhomberg and Wolff (1998) using pair-wise comparisons of LD50 values for different species for 17 a large number of chemicals on the RTECS database. Their findings contrast with the U.S. EPA 18 (1992) findings, which largely evaluated multiple exposure studies, in that the best 19 correspondence of toxicity across species for LD50 values was found when doses were expressed 20 as mg/kg. This finding might argue for the NRC (1993a) recommendation to scale doses across 21 species based upon minute volume to body weight ratios. However, this conclusion would be 22 based upon an evaluation of oral toxicity studies, most of which were probably by gavage. Bolus 23 doses result in a high peak body dose, in contrast to the inhalation of a chemical over a number 24 of hours with a more constant body burden over time. The question then becomes, does 25 inhalation exposure on the order of hours mimic the toxic response seen with multiple exposures 26 (U.S. EPA, 1992) or the acute bolus doses used in the Rhomberg and Wolff (1998) analysis? If 27 the former situation prevails then the rationale by Vocci and Farber would argue for no dosimetry 28 corrections being made. On the other hand, the latter case would argue for the use of the NRC 29 (1993a) methodology. 30 31 2.4.2 Current Approach of the NAC/AEGL Committee to Dosimetry 32 Corrections 33 34 Given the large amount of uncertainty surrounding this issue, and the fact that the use of 35 no dosimetry corrections for gasses across species would be the most conservative approach, the 36 NAC/AEGL Committee has chosen not to use dosimetry corrections across species. However, as 37 the science surrounding this issue progresses the NAC/AEGL Committee will continue to re- 38 evaluate it's conclusions. If data are available, on a chemical-by-chemical basis, which would 39 scientifically support dosimetry corrections for gases in the development of AEGL values, they 40 will be used to do so. 41 42 As AEGL values are developed for particulates, the methodology developed by the U. S Sop08-02 wpd Printed July 6, 2000 42 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Environmental Protection Agency, and validated with experimental data on particulate matter, 2 will be reviewed and applied on the basis of the individual material (U. S. EPA, 1994b). 3 Sop08-02 wpd Printed July 6. 2000 43 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2.5 GUIDELINES/CRITERIA FOR SELECTION OF UNCERTAINTY 2 FACTORS TO ADDRESS THE VARIABILITY BETWEEN ANIMALS 3 AND HUMANS AND WITHIN THE HUMAN POPULATION 4 5 2.5.1 Introduction 6 7 The variation in the toxicological response of organisms to chemical exposures is well 8 known. This variability can be expressed across species and among individuals within the same 9 species. Lack of knowledge about the range of variability introduces uncertainties into any 10 estimate of AEGL values based upon biological data. To account for known and unknown 11 variability in response, the value derived from experimental data is adjusted by a value that 12 reflects the degree of uncertainty. This value is referred to here and by most agencies and 13 organizations as the uncertainty factor (UF). If an extrapolation is being made from animal data 14 to humans the total UF is a composite of an mterspecies UF to account for possible differences 15 between animal and human response to the chemical, and an mtraspecies UF to account for 16 differences in response to the chemical within the human population. The mtraspecies UF is 17 needed to account for possible variabilities in response by "... those at either extreme of age, 18 those with poor nutritional status, those with preexisting diseases, such as certain heart diseases, 19 that are fairly widespread in the general population, those with enhanced hereditary 20 susceptibility, or those who are overexposed because of unusual physical exertion." (NRC 1993a, 21 p88). 22 23 Inter- and intraspecies uncertainty factors have been used in the development of "safe" or 24 threshold exposure levels for chronic, non-cancer toxicity by health organizations throughout the 25 world. Examples include the acceptable daily intake (ADI) (Lu, 1988; Truhaut, 1991; Lu and 26 Sielken, 1991), the tolerable daily intake (TDI) or tolerable concentration (TC) (Meek et al., 27 1994; IPCS, 1994), the minimal risk level (MRL) (Pohl and Abdin, 1995), the reference dose 28 (R.fD) (Barnes and Dourson, 1988; Dourson, 1994), and the reference concentration (RfC) (U. S. 29 EPA, 1994b; Jarabeck, 1994). The importance of using distribution based analyses to assess the 30 degree of variability and uncertainty in risk assessments has been emphasized in recent trends in 31 risk analysis. This will enable risk managers to make more informed decisions and better inform 32 the public about possible risks and the distribution of those risks among the population (Hattis 33 and Anderson, 1999). These techniques can be used to assess variability from differences in 34 individual exposure and susceptibility for specific risk assessments in order to reduce the 35 uncertainty in estimating the real variability which exists in a population (Hattis and Burmaster, 36 1994; Hattis and Barlow, 1996). 37 38 The use of uncertainty factors in the development of AEGL values is designed to protect 39 the general public, including sensitive subpopulations, from short-term exposures to acutely toxic 40 chemicals. However, it is recognized that certain individuals, subject to unique or idiosyncratic 41 responses, could expenence adverse effects at concentrations below the corresponding AEGL 42 level. "In the case of CEEL-2 (AEGL-2), uncertainty factors must be balanced against the Sop08-02 wpd Printed July 6, 2000 44 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Barnes and Dourson (1988) described the U.S. EPA's approach and rationale to assessing 2 non-carcinogenic health risks from chronic chemical exposure. The U. S. EPA approach follows 3 the general format as set forth by the National Research Council (NRC, 1983). The conceptual 4 difference between "safety" and "uncertainty" is discussed within the context of the terms safety 5 factor (SF) versus uncertainty factor (UF) and acceptable daily intake (ADI) versus reference 6 dose (RfD). The authors state that "safety factor" suggests the notion of absolute safety and that 7 the ADI is generally and erroneously interpreted as a strict demarcation between what is 8 "acceptable" and what is "safe" in terms of chronic exposure. In reality, the ADI represents an 9 estimate of a level where the probability of adverse effects is low but a level where the complete 10 absence of all risk to all people cannot be assured. Consequently, the RfD and UF terminology 11 was developed and adopted by the U. S. EPA. The U. S. EPA considers the RfD to be an 12 estimate (with uncertainty spanning perhaps an order of magnitude of a daily exposure to a 13 human population, including sensitive subpopulations), that is likely to be without an appreciable 14 nsk of deleterious effects during a lifetime. 15 16 Dourson, et al. (1992) conducted an analysis of chronic and subchronic toxicity data on 17 69 pesticides obtained from EPA's Integrated Risk Information System (IRIS) to determine the 18 potential impact of missing studies on the quality of the RfD values derived. Certain of these 19 data proved useful in determining interspecies variations in toxic response to long term oral 20 ingestion of a wide range of pesticides. The authors' analyses of 1- to 2- year studies indicated 21 that the probability of the rat NOAEL for each of 67 pesticides exceeding the dog NOAEL by 22 greater than 3.16-fold was 28 percent and the probability of the rat NOAEL exceeding the dog 23 NOAEL by greater than 10-fold was 10 percent. Also, the probability of the dog NOAEL in the 24 same studies exceeding the rat NOAEL by greater than 3.16-fold was 19 percent and the 25 probability of the dog NOAEL exceeding the rat NOAEL by greater than 10-fold was 4 percent. 26 These data support the value of using uncertainty factors (UFs) derived from data in developing 27 RfDs and suggests that UFs between species may be significantly less than 10-fold for a wide 28 range of structurally diverse chemicals. 29 30 Renwick (1993) considered the expression of toxicity to be the combined result of 31 toxicokinetics (all processes contributing to the concentration and duration of exposure of the 32 active chemical toxicant at the target tissue) and toxicodynamics (mode or mechanism of action 33 of the active toxicant at the target tissue site). Therefore, he reasoned that since both 34 toxicokinetics and toxicodynamics contribute quantitatively to the uncertainty factor, it is 35 necessary to subdivide each of the 10-fold UFs (inter- and intraspecies) into these two 36 components to effectively accommodate differences in contributions made by toxicokmetic and 37 toxicodynamic factors. Hence, for any chemical, appropriate data may be used to derive a 38 specific data-derived factor for that component. The overall inter-and intraspecies UFs would 39 subsequently be determined as the product of the known data-derived factor or factors and the 40 "default" values for the remaining unknown factors. The author evaluated published data for 41 parameters that measure interspecies differences in plasma kinetics (physiological changes, 42 differences in rates of absorption, biotransformation, and elimination) in laboratory animals and Sop08-02 wpd Printed July 6. 2000 47 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 little or no evaluation of scientific data to support or reject the use of this value. Today there is 2 greater knowledge and insight and defined methods to evaluate sensitivity or variability in 3 responses and selecting or deriving more scientifically credible UFs. 4 5 Dourson and Stara (1983) introduced the concept that empirical data are available to 6 support the use of UFs for both inter- and intraspecies adjustment. This was followed by the 7 publication of an analysis of the chronic and subchronic toxicity data obtained from U. S. EPA's 8 Integrated Risk Information System (IRIS). Certain of these data proved useful in determining 9 the extent of interspecies variations in toxic response to long-term oral ingestion of a wide range 10 of pesticides. More recently the concept of data-derived UFs has been introduced (Renwick, 11 1993; Dourson et al., 1996). Finally, the concept of dividing, evaluating, and quantifying 12 separately the toxicokinetic and toxicodynamic factors from each of the inter- and intraspecies 13 UFs has been proposed (Renwick, 1993). 14 15 One important consideration in the selection or derivation and use of UFs for the 16 development of AEGLs is the nature of the toxicant and the exposure period. Much of the data, 17 information, and emphasis to date on non-carcinogenic and non-mutagenic substances has 18 addressed chronic effects from long-term or life-time exposures. Certain of the reports 19 discussing the toxicokinetic and toxicodynamic factors as related to variability of response have 20 drawn on carcinogenic or mutagenic mechanisms as a basis for scientific support. By contrast, 21 the AEGL values address relatively high concentration, short-term exposures to threshold effects 22 of acutely toxic chemicals. In attempting to draw on the scientific foundations upon which UFs 23 are being selected for use in developing chronic guideline levels such as RfDs and RfCs, it is 24 important to maintain an awareness of certain potential differences when considering acute 25 guideline levels such as AEGLs. Responses to chronic exposures may be greater between 26 species or between individuals as compared to responses to acute exposures. For example, the 27 impact of individual differences in absorption, excretion, metabolism, rate of repair or 28 accumulation of unrepaired damage may be magnified through exposure to lower concentrations 29 over extended time periods. The higher concentrations associated with acute exposure may tend 30 to overwhelm existing defense mechanisms, possibly ameliorating certain differences in response 31 among species and among individuals within the same species. The higher concentrations 32 associated with single exposures, together with the short-term nature of the exposure period, may 33 nullify existing differences in absorption, metabolism, and excretion of a substance, as well as 34 differences in repair mechanism rates, and other factors. Hence, acute exposure to acutely toxic 35 substances in some instances may reduce the variability in response between species and among 36 individuals of the same species depending upon the mode of action of the chemical. 37 Additionally, the fact that AEGLs are based on, and intended for, inhalation exposure adds one 38 more important dimension to the complexity of differences between individuals and species. 39 40 Based on the considerations presented above, the acceptance and use of default UFs based 41 upon chronic exposure data should be carried out only after careful evaluation of chemical 42 specific data for single exposures. However, the concepts, ideas, and approaches to developing Sop08-02 wpd Printed July 6. 2000 49 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 UFs that have emanated from the chronic exposure studies of the past 10 years is of substantial 2 value in the development of AEGLs and will be employed as appropriate m the selection or 3 derivation of UFs used in the AEGL program. 4 5 2.5.3.1 Interspecies Uncertainty Factors - Use in the Development of AEGL 6 Values - Discussion 7 8 Where data are insufficient to determine the relative sensitivity of animals to man, an 9 uncertainty factor of 10 has been used by U. S. EPA, U. S. ATSDR, Health Canada, International 10 Program on Chemical Safety (IPCS), and Rijksinstituut voor Volksgesondheid en Milieu 11 (RTVM) when developing the equivalent of reference doses for chronic exposure to chemicals 12 (Dourson et al, 1996). When extrapolations are made from animals to humans based upon mg/kg 13 of body weight the factor of 10-fold is usually adequate to account for differences in response. 14 Dourson and Stara (1983) found that a factor of 10 accounted for many of the animal to human 15 differences observed when the dose was adjusted for differences between human and animal 16 body weights and body surface areas. 17 18 Brown and Fabro (1983) compared the lowest effective dose to cause teratogemcity in 19 animals (mouse, rat, rabbit, cat, monkey) and humans for 8 chemicals (methyl mercury, diethyl 20 stilbesterol, methotrexate, aminoptenn, PCBs, thalidomide, phenytoin, alcohol). The LOAEL 21 ratios ranged from 1.8 to 50 with a geometric mean of 7. Humans were generally more sensitive 22 on an administered oral dose/body weight basis but by less than an order of magnitude. This 23 analysis is complicated by the fact that the criteria and confidence in determining the lowest 24 effective dose are not discussed, and the 8 chemicals may represent potent developmental 25 toxicants in humans since their effect in humans represented the basis for their selection. The 26 potency estimates in humans may represent only the sensitive part of the distribution of human 27 response to exposure. The animal response dose may be closer to the mean response level, and 28 therefore presents a higher LOAEL for the species. However, the retrospective nature allows the 29 choice of the most sensitive animal species. In most instances the animal database is incomplete. 30 Thus, this analysis may represent the spectrum of results in which humans are more sensitive 31 than animals to developmental toxicants. 32 33 Renwick 1993 subdivided the inter- and intraspecies UFs into two components to address 34 toxicokinetics and toxicodynamics separately. Although the supporting data for this concept is 35 from chronic animal feeding studies and in vitro cell cultures, the concept of considering the 36 kinetics and dynamics separately across species has relevance to UFs for AEGLs. Renwick 37 proposed specific quantitative values of 4-fold and 2.5-fold for the kinetics and dynamics 38 components, respectively. Although this approach has merit, the NAC/AEGL Committee does 39 not make such a precise quantitative differentiation. To date the NAC/AEGL Committee uses 40 only general information on the kinetic and dynamic components of toxicity to adjust the 41 interspecies uncertainty factor from 10 to 3 or 1. This approach is also consistent with the 42 recommendation by Dourson et al. (1996) to use data-denved uncertainty factors when SopO&-02 wpd Printed July 6, 2000 50 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 appropriate data are available. This approach is in keeping with the U. S. EPA's general 2 approach in the development of RfDs. For example in the case of Aroclor 1016 the default 3 tnterspecies UF of 10 was reduced to 3 because of the similarity with which monkeys and 4 humans respond to, and metabolize PCBs (toxicokinetics) and the physiologic similarity 5 (toxicokinetics) between the two species (U. S. EPA, 1996b). 6 7 Comparisons of the current approach to determine UFs for AEGLs with other short-term 8 exposure limits has not altered the current thinking of the NAC/AEGL Committee. In the 9 development of Emergency Exposure Guideline Levels (EEGLs) by the National Research 10 Council (NRC, 1986) a factor of 10-fold was used for mterspecies extrapolation. However, no 11 EEGLs have been developed in the last 15 years so it is not known if different uncertainty factors 12 might be used in light of the more recent concepts and data on interspecies differences. 13 14 The NAS Guidelines for Developing Spacecraft Maximum Allowable Concentrations for 15 Space Station Contaminants (SMACs) states that uncertainty factors between 1- and 10-fold are 16 used for each source of uncertainty (NRC, 1992a). The sources include intraspecies (human) 17 response variabilities, interspecies variabilities, the extrapolation of a LOAEL to a NOAEL, and 18 the extrapolation from an inadequate or incomplete data base. For 1 hour SMACs the NAS 19 employed an overall (combined intra- and interspecies) UF of 10-fold when only animal data 20 were available or when the route of human exposure differed from the study. However, the 21 population for which SMACs is intended does not include infants, children, the elderly, or the 22 infirm and is, therefore, a more homogeneous and healthier subpopulation. 23 24 The National Research Council (NRC, 1993a) recommended the use of an interspecies 25 uncertainty factor (UF) within the range of 1- to 10-fold to account for differences between 26 animals and humans. The guidance suggests that the UF should be based on the quality of the 27 data available. In this regard, the NAC/AEGL Committee evaluates data on a chemical-by- 28 chemical basis, considers the weight of evidence, and uses scientific judgement in the selection 29 of interspecies UFs. As data become available, the NAC/AEGL Committee will use data-derived 30 interspecies uncertainty factors. 31 32 Information bearing on the toxicokinetics and toxicodynamics of the chemical under 33 consideration, as well as structurally related analogues and/or chemicals which act by a similar 34 mechanism of action, will be used to derive an appropriate interspecies factor which may range 35 from 10 to 3 or 1. In the absence of information on a subject, or analogous, chemical to set data- 36 derived uncertainty factors, the use of a default uncertainty factor of 10 is considered to be 37 protective in most cases. As always, all information on the chemical, its mechanism of action, 38 structurally related chemical analogs, and informed professional judgement will be used when 39 determining appropriate uncertainty factors and evaluating the resultant AEGL values. 40 41 42 2.5.3.2 Interspecies Uncertainty Factors - NAC/AEGL Committee Guidelines Sop08-02 wpd Printed July 6, 2000 51 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2 General guidelines followed by the NAC/AEGL Committee to select UFs are presented 3 below. In each section there is a list of questions which should be addressed to support the 4 rationale for the choice of the uncertainty factor used. The guidelines are organized into 5 categories for convenience. However, more than one guideline may be applied to the selection of 6 any one uncertainty factor. 7 8 2.5.3.2.1 Most Appropriate Species Used 9 10 In cases where there is little interspecies variability (e.g., within a factor of 3), and/or the 11 most sensitive species is selected, and/or a species closely related to humans was selected, the 12 interspecies uncertainty factor is typically reduced from 10 to 3. It should be noted that m these 13 cases the mechanism of action can be identified and there is evidence that it is not expected to 14 vary significantly between species. 15 16 17 THE RATIONALE FOR THE SELECTION OF AN UF SHOULD INCLUDE: 18 1. The species tested. 19 2. The toxicologic endpoint used for the AEGL derivation. 20 3. The qualitative and quantitative range of response of the species tested. 21 4. Discussion of why the species/study chosen was the most appropriate. 22 5. Discussion of the variability among studies with the same species or among strains. 23 24 2.5.3.2.2 Most Sensitive Species Not Used 25 26 In instances where the most sensitive species is not used, an uncertainty factor of 10 is 27 generally used. 28 29 30 THE RATIONALE FOR THE SELECTION OF AN UF SHOULD INCLUDE: 31 1. The species tested. 32 2. The toxicologic endpoint used for the AEGL derivation. 33 3. The qualitative and quantitative range of responses of the species tested. 34 4. Discussion of why the most sensitive species was not used, and/or why the less 35 sensitive species was selected. 36 37 2.5.3.2.3 Mechanism of Action is Unlikely to Differ Among Species 38 39 If evidence is available indicating the mechanism of action, such as direct acting irritation 40 or alkylation is not expected to differ significantly between species an interspecies uncertainty 41 factor of 3 is generally used. 42 Sop08-02 wpd Printed July 6. 2000 52 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30. 2000 1 THE RATIONALE FOR THE SELECTION OF AN UF SHOULD INCLUDE: 2 1. A description of the mechanism of action. 3 2. A discussion of why the mechanism of action is unlikely/likely to differ? Is 4 bioavailability/metabolism/detoxification/elimination likely to be an issue? 5 6 7 2.5.3.2.4 Mechanism of Action is Unknown 8 9 In cases where the mechanism of action is unknown, or insufficient data between species 10 are available, or there are likely to be substantial (but inadequately quantified) differences in 11 metabolic and physiological response between species, an interspecies uncertainty factor of 10 is 12 applied. 13 14 THE RATIONALE FOR THE SELECTION OF AN UF SHOULD INCLUDE: 15 1. Description of the toxicological effects observed. 16 2. Description of the range of uncertainty in toxicologic response and how that relates to 17 this assessment. 18 3. Discussion of what is known/unknown about the mechanism of action. 19 4. Discussion of the extent of data available among species. 20 21 2.5.3.2.5 Variability in Response Between Species 22 23 When there is a wide degree of variability between species, or strains or experiments 24 which cannot be adequately explained, an interspecies uncertainty factor of 10 is applied. 25 26 THE RATIONALE FOR THE SELECTION OF AN UF SHOULD INCLUDE: 27 1. Description of the response. 28 2. Discussion of the differences or similarities in pharmacokinetic parameters 29 (absorption/metabolism/detoxification/elimination) among species. 30 3. Discussion of the range of dose-dependent response(s) of the species tested and the 31 qualitative and quantitative aspects of the data. 32 33 2.5.3.2.6 Humans More Sensitive than Animals 34 35 Where published data show humans are more sensitive than animals, an interspecies 36 uncertainty factor of 10 is used unless published results demonstrate otherwise. 37 38 THE RATIONALE FOR THE SELECTION OF AN UF SHOULD INCLUDE: 39 1. Description of the toxicologic endpomts for which humans and animals show 40 differential sensitivity. 41 2. Discussion of the factors where humans are thought to be more/less sensitive than 42 animals. Sop08-02 wpd Printed Juty 6, 2000 53 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 3. State which species were tested. 2 4. Discussion of the range of response of the species tested. This discussion should 3 address qualitative and quantitative aspects of the data. 4 5. Discussion of why humans are more susceptible than test animals. 5 6 2.5.3.2.7 Use of an Uncertainty Factor of 10 7 8 The uncertainty factor for interspecies response adjustment is 10 when there is 9 insufficient information about the chemical or its mechanism of action to justify a lower UF, or if 10 data are available suggesting a high degree of variability between species 11 12 THE RATIONALE FOR THE SELECTION OF AN UF SHOULD INCLUDE: 13 1. Discussion of why an uncertainty factor of 10 is chosen. For example, the analysis 14 may depend upon data collected in only one species, high variability of response, 15 uncertainties in exposure measurement, etc. This statement could point to data 16 gaps which could be filled if the need exists. 17 18 19 2.5.3.2.8 A Selected Uncertainty Factor Applied to Animal Data Would Drive 20 the AEGL-2 or -3 Level to a Value Which Humans can Tolerate without 21 Lethal or Serious Adverse Effects 22 23 Where the application of an interspecies uncertainty factor of 10 reduces the AEGL-3 24 level, the threshold for lethality, or the AEGL-2 level, the threshold for irreversible or disabling 25 effects, to an exposure concentration which humans are known to tolerate without adverse effect, 26 the interspecies uncertainty factor is reduced to 3 or 1. 27 28 THE RATIONALE FOR THE SELECTION OF AN UF SHOULD INCLUDE- 29 1. Citations and explanations of the human data and how it relates to the AEGL value 30 denved with an UF selected on the basis of the existing guidelines. 31 32 2.5.3.2.9 A multiple exposure study was used to set the level. 33 34 In cases where a single exposure AEGL value is denved from a multiple exposure study 35 because the acute data set for a single exposure is lacking, the multiple exposure data are 36 considered an inherently conservative estimate because a biological organism is expected to have 37 greater tolerance to a single exposure as compared to multiple exposures to the same chemical. 38 If the adverse effect identified in the multiple exposure study is cumulative for the AEGL level of 39 concern, the interspecies uncertainty factor used to adjust the multiple exposure animal data 40 might be reduced to 1 or 3. Careful judgement should be used when making this assessment. If 41 a chemical is cleared very rapidly, or there is evidence that the concentration causing the effect 42 does not vary with duration or number of exposures, then the animal may be able to sustain Sop08-02 wpd Printed July 6. 2000 54 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 repeated insult at a level close to a single acutely toxic exposure. Thus in these instances the 2 reduction of the uncertainty factor based on multiple exposures versus a single exposure would 3 not be justified. 4 5 THE RATIONALE FOR THE SELECTION OF AN UF SHOULD INCLUDE: 6 1. A description of the study. 7 2. Discussion of the known or suspected clearance rate and other toxicokinetic properties 8 of the chemical. For example, does the concentration causing the effect vary significantly with 9 time or number of exposures? Sop08-02.wpd Printed July 6. 2000 55 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 l 2.5.3.3 Intraspecies Uncertainty Factors - Use in the Development of AEGL 2 Values - Discussion 3 4 Intraspecies uncertainty factors (UFs) are used to address the variability in biological 5 response that exists within a human population exposed to a toxic agent. Their use represents an 6 important step in the AEGL development methodology and is designed to account for the 7 differences which can occur within the general population. 8 9 The National Research Council, National Academy of Sciences' guidelines for developing 10 emergency exposure limits state that the exposure limits are "designed to protect almost all 11 people in the general population..." (NRC, 1993a). The NRC guidelines state that, although the 12 levels "...are designed to protect 'sensitive' individuals, some hyper-susceptible individuals might 13 not be protected...". This distinction is based on the premise that emergency exposure limits 14 must be set low enough to protect the general population but must also be set at levels that 15 minimize the risks associated with inappropriate or unwarranted response to chemical 16 emergencies as a result of rare or exceptional circumstances. Consequently, the AEGL values 17 may not be expected to necessarily protect certain individuals with unique or idiosyncratic 18 susceptibilities. This consideration is clearly communicated in the NAC/AEGL Committee's 19 definition of the AEGLs. 20 21 When data are insufficient to determine the relative sensitivity of individuals in a human 22 population exposed to a specific chemical, a default uncertainty factor of 10 has been used by U. 23 S. EPA, U. S. ATSDR, Health Canada, IPCS, and RIVM when developing the equivalent of 24 reference doses for chronic exposure to chemicals (Dourson et al, 1996). This value of 10 is 25 generally applied to the NOAEL (the highest observed or calculated dose which did not cause an 26 adverse effect in an experiment). A number of studies have tried to address the issue of the 27 reasonableness or validity of this factor. Under ideal circumstances an analysis would provide 28 information on the ratios of the experimentally observed NOAELs for different human groups 29 within a population for a wide range of defined exposures to chemicals. Groups would be 30 identified based upon biochemical or physiological differences which might cause members of 31 the group to respond to chemical exposure in a fundamentally different manner - either 32 quantitatively or qualitatively. Sample sizes would be large and include a wide variety of genetic 33 backgrounds. Such examples would include differences among newborns, infants, children, 34 adults, the elderly, the infirm, and those compromised by illness, including asthmatics. The 35 NOAELs also would represent a distinct relationship between dose level and response. These 36 data would encompass all variables due to the toxicokinetics and toxicodynarrucs factors. Such 37 data are not available, even in carefully controlled, double blind clinical trials for new therapeutic 38 drugs. However, surrogates have been developed which provide information on the 39 reasonableness of the choice of the intraspecies uncertainty factor of 10 or less. This approach is 40 referred to as the use of data-derived uncertainty factors. 41 42 Dourson and Stara (1983) analyzed the slopes of 490 adult p.o. rat LD50 studies reported Sop08-02 vypd Printed July 6,2000 56 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30.2000 1 by Weil (1972). They calculated the intraspecies adjustment factor required to reduce the dose 3 2 standard deviations below the median LDSO response using a probit, log-dose analysis. This gives 3 a z value of 0.4987 from the mean or a calculated response of 1.3/1000 (Spiegel, 1996). This 4 was used to predict the response of a sensitive subgroup in the population. An adjustment factor 5 of 10 was adequate to reduce the response from a dose killing 50% of the animal population to a 6 level which would kill only the most sensitive members of the inbred rat population in 92% of 7 the chemicals studied. These data support the contention that a 10-fold uncertainty factor is 8 adequate in many instances to account for intraspecies differences in response to acute exposures. 9 However, in some instances this UF may not protect the more sensitive members of the 10 population. The extrapolation reported here represents a measure of 3 standard deviations from 11 the median response data points. Statistically, an extrapolation of 3 standard deviations from the 12 mean includes more than 99 percent of the population in question, or approximately 999 13 individuals out of a population of one thousand. The extrapolation of three standard deviations 14 as performed by Dourson and Stara (1983) includes a similar proportion of the population in 15 question, 998.7 out of 1000. It is interesting to note that the Fowles et al. (1999) analyses of 16 inhalation toxicity experiments revealed that for many chemicals, the ratio between the LCJO and 17 the experimentally observed non-lethal level was on average a factor of approximately 2, the 90* 18 percentile was 2.9, and the 95th percentile was 3.5. There was a range of ratios from 1.1 to 6.5. 19 Therefore, the use of an UF of 3-fold with a NOAEL for lethality can achieve the same reduction 20 in acute lethality as that reported by Dourson and Stara (1983). The 490 LD50 studies with rats 21 were undoubtedly based on a wide range of chemical substances exhibiting many different 22 toxicological mechanisms. Hence, the variability due to chemical-specific properties was 23 included in this evaluation and was accounted for by an adjustment factor of 10-fold in 92 24 percent of the chemicals tested. This type of statistical analysis makes the untested hypothesis 25 that the slope of the dose response was the same in the experimental dose range and at the 26 untested tails of the experiment. It also reflects the response in a homogeneous (inbred) adult 27 animal population and does not measure the difference in values between potentially sensitive 28 subgroups such as adult vs newborn. 29 30 A number of authors have presented data and analyzed adulf.newborn LD50 ratios to 31 assess the differential sensitivity of young and adult animals. Done (1964 as cited in NRC, 32 1993b) compiled LDSO ratios between immature and mature animals. He found that for 34 of 58 33 chemicals the immature animals were more sensitive that adults, and for 24 of 58 chemicals the 34 adults were more sensitive than the immature animals (NRC, 1993b). A similar compilation of 35 newbom/neonate and adult LD50 ratios for rat and mouse was done by Goldenthal (1977) on data 36 submitted to FDA in drug applications. This included a broad range of chemicals such as 37 analgesics, bronchodilators, CNS depressants and stimulants, anti-depressants, tranquilizers, etc. 38 NRC (1993b) analyzed these data and found that about 225 of the compounds were more toxic to 39 neonates and 45were more toxic to adults. Almost all of the age related differences from the 40 Done (1964 as cited in NRC 1993b) and Goldenthal (1977) data collections were within a factor 41 of 10 of each other and most of the ratios were within a factor of 3 (NRC, 1993b). Sheenan and 42 Gaylor (1990) analyzed adult:newbom LDSO ratios for 238 chemicals. The median ratio of the Sop08-02 wpd Printed July 6. 2000 5 7 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30,2000 1 LDJO values between age groups was 2.6. Approximately 86% of the ratios were less than 10 2 indicating that this factor is adequate to account for differences in response to chemical exposure 3 between adult and young in most cases but may be insufficient for 14% of the cases. In these 4 studies the comparison was made from the median response. 5 6 Another indirect approach to quantify biological uncertainty is to measure the observed 7 variability in human populations. Calabrese (1985) examined a number of parameters related to 8 toxicokinetics (metabolism, binding of chemicals to protein and DN A, and activity levels of. 9 enzymes). In studies which included between 10 and 349 subjects he concluded that generally 10 75-95% of the population fell within a range of 10-fold. However, his conclusion was based on 11 the supposition that the 10-fold factor was to account for the total range of human variability as 12 opposed to the range from an experimental NOEL to the most sensitive person. In a similar 13 study, Hattis et al. (1987) evaluated toxicokinetic parameters in 101 data sets (5 or more healthy 14 adults) on 49 chemicals (primarily drugs). They found that 96% of the variation was within a 15 factor of 10. However, this analysis also measured the total range of human variability. These 16 analyses measured the range of responses for toxicokinetic parameters and give some sense of 17 the variability in an adult population only and not in a potentially sensitive subpopulation. They 18 do not measure how far the tail for response goes beyond the lowest dose/activity in the 19 population measured, nor the response of different populations. Another consideration is the fact 20 that these data represent measures of toxicokinetic variables which may not directly reflect the 21 threshold of toxicologic response to chemical exposure. 22 23 Ideally, one would like to be able to compare NOAEL levels observed in an experiment 24 to the tail of the NOAEL distribution in order to assess the actual frequency of response in the 25 total human population when the intraspecies uncertainty factor of 10 is applied and obtain a 26 measure of the sensitive person. Determining the experimental NOAEL is fraught with problems 27 of sample size and dose selection. The response of the sensitive population at a dose 10 fold 28 lower than the experimental NOAEL will never be known. Hattis et al. (1999) performed 29 statistical modeling analyses designed to determine the efficacy of applying the intraspecies 30 uncertainty factor of 10 to a NOAEL. They statistically analyzed clinical studies on humans 31 which measured parameters related to toxicokinetics and toxicodynamics. The studies had at 32 least 5 subjects each and included approximately 2700 data points for the toxicokinetic 33 endpoints. They demonstrated that the population distribution of the data were lognormal in the 34 data region and assumed that they were lognormally distributed out to the extreme tails. From 35 the data, and assuming a lognormal distribution, they calculated the dose required to produce an 36 incidence in 5% of the population. This is essentially an experimental NOAEL which is divided 37 by the intraspecies uncertainty factor when a risk assessment is performed. The dose at the 5% 38 incidence level was divided by 10 and the response at that dose calculated, assuming a lognormal 39 distribution of data, to the extreme tails. This approach was used to assess the response rate 40 when a 10 fold uncertainty factor is applied to a NOAEL. They found that "...acting by itself, a 41 10-fold reduction in dose from a 5% effect level could be associated with effect incidences 42 ranging from slightly less than one in ten thousand for a median chemical/response to a few per Sop08-02 wpd Printed July 6, 2000 58 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version OB-02 - June 30, 2000 1 thousand for chemicals and responses that have more human interindividual variability than 19 2 out of 20 typical chemicals/responses." The analysis did not include sensitive subpopulations so 3 the variability seen could be greater. This type of analysis assumes a lognormal distribution of 4 the data to the extreme tails. It does not allow for a threshold which is generally assumed to be 5 true for non-cancer effects. Thus, the calculated response at doses 10-fold less than the 5% 6 response level may be overly conservative. There are no data, human or animal, that far out in 7 the tail of the distribution curve. The analysis by Hattis et al. (1999) indicates that a human 8 intraspecies UF of 10 would be protective of sensitive individuals and may be overly 9 conservative in many instances. 10 11 Another approach to measuring variability between different groups of a human 12 population is to compare maximum tolerated doses (MTDs) OT effect levels between groups. 13 Reports comparing the MTDs of chemotherapeutic agents in child and adult cancer patients 14 indicate that most of the substances studied were tolerated as well, and, in many instances, 15 tolerated better by children than by adults when the dose was expressed as mg/kg body weight or 16 mg/m2 (Glaubiger et. al., 1982; Marsom, et. al., 1985). In those instances where children 17 demonstrate a greater response at equivalent dose to these substances, the differences were less 18 than a factor of two-fold. Although MTDs are not entirely a precise measure of a toxicological 19 threshold, they represent a credible parameter by which relative toxicities between groups can be 20 measured in humans. It is important to acknowledge that although the substances studied 21 represent a diverse group of chemical classes, these substances exhibit similar mechanisms of 22 cytotoxicity. Therefore, the results observed cannot be applied to a large number of other 23 chemicals with different mechanisms of action. In addition, only MTDs were reported, not the 24 variability within each group in response to the drugs. Thus, this type of study gives a measure 25 of response between groups within a population but not the variability within each group. 26 27 Other studies regarding differences in sensitivities between specific groups in humans to 28 various anaesthetic gases have been reported. These studies indicate children, particularly 29 infants, are more resistant than adults to the effects of various volatile anesthetics (Gregory, et. 30 al, 1969; Katoh and Dceda, 1992; Lerman et. al., 1983; Matthew, et al., 1996; Stevens, et ah, 31 1975; LeDez and Lerman, 1987). The susceptibility of individuals of different ages has been 32 extensively studied in the anesthesia literature where the concentrations of various anesthetic 33 gases in the lung which produce "anesthesia" (ie lack of movement) have been measured. The 34 results are usually reported as the Mean Alveolar Concentration (MAC) which produces lack of 35 movement in 50% of persons exposed to that concentration. Occasionally the ED9S - the alveolar 36 concentration which prevents movement in 95% of those exposed is also reported. MACs for 37 several anesthetic gases have been measured as a function of age. The results consistently show 38 a pattern with maximal sensitivity (lowest MAC values) in newborns, particularly prematures, 39 pregnant women, and the elderly. The least sensitive (highest MAC values) occur in older 40 infants, toddlers and children as compared to adults. The total range of sensitivity was 2-3 fold. 41 Many organic solvents for which AEGLs are developed can also produce anaesthesia in humans 42 at high doses. As previously stated, this type of study gives a measure of response between Sop08-02 wpd Printed July 6,2000 59 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 groups within a population but not the variability within each group. 2 3 Intraspecies uncertainty factors (UFs) are used to address the variability in biological 4 response that exists within a human population exposed to a toxic agent. Their use is designed to 5 account for the range of response to exposure by individuals within the general population. As 6 the studies above demonstrate, an uncertainty factor of 10 is adequate to account for variability m 7 the majority of cases and a factor of 2-3 is often adequate. 8 9 It has been proposed that data on the differences in kinetics and dynamics be used to 10 modify the uncertainty factors from defaults of 10 (Renwick, 1993; Dourson et ah, 1996). 11 Renwick (1993) proposed dividing inter- and intraspecies uncertainty factors into two 12 components. Toxicity is considered to be the combined function of toxicokinetics (all processes 13 contributing to the concentration and duration of exposure of the active chemical toxicant at the 14 target tissue) and toxicodynamics (mode or mechanism of action of the active toxicant at the 15 target tissue site). If data are available on the differences between or within species on one or 16 both of these two processes, then it should be possible to reduce the total uncertainty factor by 17 developing a data derived uncertainty factor. This approach has in fact been taken by the U.S. 18 Environmental Protection Agency in the examples below. 19 20 The U. S. EPA (I996b) reduced the default intraspecies uncertainty factor of 10 to 3 for 21 Aroclor 1016 because data from animal and human studies indicate that infants who were 22 exposed transplacentally represent a sensitive subpopulation and this information (toxicity in 23 monkeys) was used to derive the RfD value (toxicodynamics). 24 25 In the case of methyl mercury toxicodynanucs data were used to reduce the intraspecies 26 UF to 3 (U. S. EPA, 1995b). The RfD was based upon a benchmark dose computed lower 95% 27 confidence limit on the 10% increase over the background for human childhood neurological 28 abnormalities (this level has been used to represent the NOAEL) in the sensitive subpopulation 29 (the developing fetus). Therefore, the default intraspecies uncertainty factor of 10 was reduced to 30 3. Since the sensitive subpopulation had been identified, the toxicodynamic part of the 31 uncertainty factor had been addressed. However, variability due to toxicokinetics was 32 main tamed with the use of the 3 fold uncertainty factor. 33 34 For styrene the default intraspecies UF of 10 was reduced to 3 in the calculation of the 35 RfC value because the lower 95% limit of the exposure extrapolation for a NOAEL in a human 36 cross-sectional study was used and the biological exposure index had been shown to account for 37 variation in pharmacokinetic and physiological measures such as the alveolar ventilation rate (U. 38 S.EPA, 1993). 39 40 In the absence of information to set data derived uncertainty factors, an uncertainty factor 41 of 10 is considered to account for intraspecies variability in most cases. When information is 42 available about the response of a sensitive population, mechanism of action in different species SopOS-02 wpd Pnnled July 6. 2000 60 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 and/or subgroups within an exposed population, toxicokinetics or toxicodynamics, it will be 2 factored into the development of a data derived uncertainty factor which may vary between 10, 3, 3 or 1. All information on the chemical, its mechanism of action, structurally related chemical 4 analogs, a discussion of the weight of evidence and informed professional judgement are used 5 when determining appropriate uncertainty factors. 6 7 8 2.5.3.3.1 Range of Susceptibility 9 10 The National Research Council, National Academy of Sciences' guidelines for developing 11 emergency exposure limits state that the exposure limits are "designed to protect almost all 12 people in the general population..." (NRC, 1993a). The NRC guideline levels "...are designed to 13 protect 'sensitive1 individuals, some hyper-susceptible individuals might not be protected...". 14 This distinction is based on the premise that emergency exposure limits must be set low enough 15 to protect most of the general population but must also be set at levels that minimize 16 over-response to chemical emergencies as a result of rare or exceptional circumstances. 17 Consequently, the AEGL values may not necessarily protect certain individuals with unique or 18 idiosyncratic susceptibilities. This consideration is clearly communicated in the definitions of 19 the three AEGL tiers. 20 21 The definition, and intended application of AEGL values make distinctions between 22 susceptible and "hypersusceptible" individuals. It is important to characterize these two terms 23 and the potential subpopulations they may represent for purposes of uncertainty factor selection . 24 It is also important to distinguish between these two populations for purposes of risk 25 communication to emergency planners, emergency responders, and to the public. 26 27 Individual susceptibility within a population will vary according to both individual 28 determinants and the specific properties of a given chemical. The origins of susceptibility are 29 multifactorial and distributed across populations. According to the U. S. 30 Presidential/Congressional Commission of Risk Assessment and Risk Management, "Genetic, 31 nutritional, metabolic, and other differences make some segments of a population more 32 susceptible than others...susceptibility is influenced by many factors" (P/CC, 1997). The factors 33 are based on intrinsic and/or acquired differences among individuals and may include age, 34 gender, genetic factors, ethnicity and race, quality of life and life-style considerations. The latter 35 considerations may be further classified as preexisting illnesses, prior exposure(s), nutritional 36 status, personal behavior (e.g. occupation, smoking, alcohol, obesity, etc.), and socio-economic 37 factors. The NRC also characterizes such determinants: "[S]ome of the individual determinants 38 of susceptibility are distributed bimodally...other determinants seem to be distributed more or 39 less continuously and unimodally" (NRC, 1994). 40 41 Hypersusceptibility describes extreme examples of responses. Hypersusceptibility may 42 represent biological reactions that are unique, idiosyncratic and/or stem from determinants that Sop08-02 wpd Printed July 6. 2000 61 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30,2000 1 are generally discontinuous with, and lay outside of, the range of distributions expected for the 2 general population. 3 4 The determination of susceptibility entails the presence of observable changes in 5 biochemical or physiological processes reflecting dose-response relationships unique to a 6 chemical (e.g., sulfur dioxide) or class of chemicals (e.g., acid aerosols). Susceptibility and 7 hypersusceptibility are not meaningful concepts outside of the context of specific exposures; 8 "Dose-response relationships are chemical-specific and depend on modes of action; people are 9 not hyper-susceptible to all kinds of exposures" (P/CC, 1997). 10 11 Susceptibility and hypersusceptibility may reflect transient, rather than permanent states. 12 For example, infants are susceptible to some chemicals (e.g. ingested nitrates and nitrites as a 13 result of their relatively high gastric pH), but lose that susceptibility as they mature. Susceptible 14 populations may also experience transient penods of hypersusceptibility. For example, 15 asthmatics represent 5 to 10 percent of the general population and can be more susceptible than 16 non-asthmatics to challenge by respiratory irritants. Moreover, at any given time some 17 asthmatics may be suffering acute asthmatic attacks, which might lead to a hypersusceptible 18 condition, just prior to an irritant exposure. Based on the transient condition, these individuals 19 might not be accounted for in the published AEGL values. Similarly, otherwise normal 20 individuals may suffer transient periods of hypersusceptibility during penods of illness. For 21 example, following very severe, acute respiratory infections, many non-asthmatic individuals will 22 experience several weeks or more of bronchiolar hyper-reactivity and bronchospasm following 23 non-specific exposure to respiratory irritants. This condition can be considered an example of 24 transient hypersusceptibility. In general, since there is little or no information regarding the 25 responses of transiently hypersusceptible individuals to chemical exposures, the corresponding 26 AEGL values might not be protective for this group. 27 28 During the past 15 years, a wide range of symptoms and complaints in patients thought to 29 be related to extreme sensitivity to low-levels of diverse and often non-quantifiable chemical 30 exposures have been reported by clinicians and researchers. This syndrome has been referred to 31 as "Multiple Chemical Sensitivity" or MCS (Cullen, 1987). MCS has been characterized as the 32 heightened, extraordinary, or unusual response of individuals to known or unknown exposures 33 whose symptoms do not completely resolve post exposure and/or whose sensitivities seem to 34 spread to other chemicals (Ashford, 1999). The syndrome is thought by Ashford to be a 2-step 35 process with an initial acute exposure to high concentrations of a substance and the subsequent 36 triggering of symptoms at extraordinarily low-levels of exposure to the same substance or 37 different substances. He believes that repeated or continuous lower level exposures may also 38 lead to the same type of sensitization. Ashford and Miller (1998) also postulate that this 39 sensitivity may be the consequence of a variety of disease processes resulting from "toxicant- 40 induced loss of tolerance" (TILT) - described as "a new theory of disease providing a 41 phenomenological description of those disease processes". 42 Sop08-02 wpd Printed July 6. 2000 62 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 In response to the increasing public demand for government attention to a problem 2 frequently identified as MCS, the Environmental Health Policy Committee (EHPC) of the U.S. 3 Public Health Service (USPHS), formed the Interagency Workgroup on MCS in 1995 to address 4 this issue. The workgroup's charge was to review the scientific literature on MCS, consider the 5 recommendations from various expert panels on MCS, review current and past federal actions, 6 and make recommendations to policy makers and researchers at government agencies concerned 7 with evaluating public health issues that might relate to MCS-like syndromes. The Workgroup 8 comprised scientists from the U. S. federal agencies, including, ATSDR, DOD, DOE, DVA, 9 NCEH (CDC), NEEHS (NIH), and EPA. The original draft report was peer reviewed by 12 10 independent experts in occupational and/or environmental medicine, toxicology, immunology, 11 psychology, psychiatry, and physiology. A Predecisional Draft Report was issued for public 12 comment on August 24,1998 (U. S. PHS, 1998). Although a final report has not yet been issued, 13 the Draft Report concluded that MCS remains a poorly defined problem where the experts 14 disagree on possible causes (e.g., physical or mental) while the sufferers complain of a wide 15 range of symptoms (not associated with any "end-organ" damage) that may result from a 16 disruption of homeostasis by environmental stressors. 17 18 In addition to the EHPC Interagency Workgroup on MCS, the U. S. National Academy of 19 Sciences (NRC, 1992b, 1992c), professional organizations (ACOEM [McLellan et al., 1999]; 20 AAAJ, 1986; AAAA1, 1999;), and others (Kreutzer, et al., 1999; Kipen and Fiedler, 1999; 21 Graveling, et al, 1999) have attempted to address this issue. Despite these attempts, the 22 diagnosis, treatment and etiologic assessment of MCS has remained a troublesome medical and 23 social concern for individuals, physicians, government and organizations (McClellan et al., 24 1999). No consensus has yet been reached for a case definition (U. S DHHS, 1995; ACOEM 25 [McLellan et al., 1999]; Graveling, et al, 1999), diagnostic methods (U. S. DHHS, 1995; 26 AAAAI, 1999; ACOEM [McLellan et al., 1999], or treatment (AAAAI, 1999). Further, despite 27 extensive literature on the existence of MCS, "there is no unequivocal epidemiological evidence; 28 quantitative exposure data are lacking; and qualitative exposure data are patchy" (Graveling et 29 al., 1999). Although most reviewers contend that symptoms characteristic of chemical 30 sensitivities exist, they agree that symptoms may be exaggerated and may be "differentially 31 precipitated by psycho social events or stress, or by different physical or chemical exposures" 32 (Ashford, 1999). All researchers and clinicians familiar with the problem agree more work must 33 be done to understand the unexplained symptoms that are attributed to MCS (Kipen and Fiedler, 34 1999). 35 36 The American College of Occupational and Environmental Medicine (ACOEM), the 37 American Academy of Allergy, Asthmatics, and Immunology (AAAAI) and the International 38 Programme on Chemical Safety (IPCS) have all recommended that the term "idiopathic 39 environmental intolerance" be used to replace the term MCS (McClellan et al., 1999; IPCS, 40 1996; AAAAI, 1999). These authors believe that the term MCS incorrectly implies that the 41 condition affects the immune system and that chemical exposure is its sine qua non (McLellan et 42 al., 1999). No immunological dysfunction has been identified in these patients (Graveling, et al., SopOe-02 wpd Printed July 6. 2000 63 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30. 2000 1 1999; AAAAI, 1999). Further, they concur with other prominent medical organizations in 2 maintaining that evidence does not exist to define MCS as a distinct entity (ACOEM [McLellan 3 etal., 1999]). 4 5 While some clinicians hold that MCS occurs as a result of environmental exposures, 6 mechanism(s) by which this may take place have not been proven scientifically. No single 7 widely accepted test of physiologic function can be shown to correlate with observed symptoms 8 (U. S. PHS, 1998; Brown-DeGagne and McGlone, 1999; AAAAI, 1999; McLellan et al., 1999). 9 Immunologic, allergic, neuropsychological, and traditional psychiatric disorders have all been 10 postulated to cause MCS, but to date, they have not been supported by well designed studies (U. 11 S. PHS, 1998; McLellan et al., 1999; Brown-DeGagne and McGlone, 1999). 12 13 As a result of the considerations presented here, it is not believed that MCS represents a 14 viable scientific basis for developing AEGL values, including further adjustments for sensitive 15 subpopulations, at this time. However, the need for scientific research on this proposed 16 syndrome that may help explain and describe its features, enable scientifically valid approaches 17 to hazard or risk assessment, and define appropnate clinical interventions is recognized. Also, 18 the NAC/AEGL Committee will remain vigilant and will consider any new data or information 19 that is scientifically credible and relevant to the development of AEGL values. 20 21 22 2.5.3.3.2 Selection of Intraspecies Uncertainty Factors 23 24 To meet the AEGL definitions that protect susceptible individuals but not necessarily 25 hypersusceptible individuals, the NAC/AEGL Committee evaluates two separate considerations 26 regarding susceptibility. First, evidence is reviewed to attempt to distinguish "susceptible" from 27 "hypersusceptible" individuals for each chemical of concern. Second, estimation of the range of 28 response variability in the general population that includes susceptible (but not necessarily 29 hypersusceptible) individuals and selection of appropriate intraspecies uncertainty factor(s) for 30 development of the AEGL values(s) is carried out. 31 32 2.5.3.3.3 Distinguishing Susceptible and Hypersusceptible Individuals 33 34 A clear distinction between susceptible and hypersusceptible individuals in all cases for 35 all chemicals is not achievable with the clinical and lexicological information available to date. 36 However, the NAC/AEGL Committee has identified specific categories and populations that may 37 be considered sensitive and part of the general population that the AEGL values are intended to 38 protect. These categories include children and infants, the elderly, asthmatics, pregnant women 39 and the fetus, and individuals with preexisting illnesses, diseases or metabolic disorders who 40 would not ordinarily be considered in a severe or critical medical condition. Examples of 41 sensitive individuals based on preexisting illnesses include those with compromised pulmonary 42 function (typical respiratory infections, smokers, immunologically sensitized due to prior SopOB-02 wpd Printed July 6, 2000 64 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 exposures, etc.), hepatic function (alcoholism, hepatitis, prior chemical exposures, etc.), cardiac 2 ftmction (dysrhythmias), and those with impaired renal function. 3 4 Hypersusceptible individuals are considered as those individuals whose reactions to 5 chemical exposure are unique and idiosyncratic, lie outside of the range of distributions expected 6 for the general public, including sensitive individuals, and constitute a relatively small 7 component of the general public. For example, the AEGLs are intended to be protective of mild 8 to moderate asthmatics, but may not necessarily be protective of severe asthmatics. Additionally, 9 there are some asthmatics who, at any given time, could be coincidentally suffering acute 10 asthmatic episodes at the time of a chemical emergency. Such individuals may be considered 11 transient hypersusceptible individuals and would not necessarily be protected by the published 12 AEGLs. Examples of hypersusceptible individuals might include those with severely debilitating 13 pulmonary, hepatic, or renal disorders or diseases, the elderly with serious debilities of primary 14 physiological systems, and those individuals with unique hypersensitivities to specific chemicals 15 or chemical classes such as the isocyanates. 16 17 Certain otherwise healthy individuals in the general population also may suffer transient 18 periods of hypersusceptibility as a result of highly severe, but reversible, short-term illnesses. 19 For example, during recovery from a severe episode of acute upper respiratory infection, many 20 non-asthmatic individuals will expenence several weeks or more of bronchiolar hyper-reactivity 21 and bronchospasm following non-specific exposure to respiratory irritants. This reversible 22 condition is considered an example of transient hypersusceptibility and it is acknowledged that 23 the AEGL values may not be protective of individuals in such circumstances. 24 25 The nature of the dose-response relationships among sensitive and hypersensitive 26 individuals is highly complex and not well-understood. In almost all instances there is no clear 27 line of demarcation that distinguishes susceptible individuals from hypersusceptible individuals 28 and there is no generic or medical guidance that can be followed for a wide range of chemical 29 exposures. However, since most biological responses are chemical-specific and are dependent 30 on the mode of action of the substance in question, the issue of identifying and protecting groups 31 or populations of sensitive individuals is addressed by the NAC/AEGL Committee on a 32 chemical-by-chemical basis. The Committee uses all available data on the properties of the 33 chemical and their relationship to both normal and compromised biochemical, physiological, and 34 anatomical systems in humans to identify and protect sensitive populations. In the absence of 35 data on the chemical in question, the use of structurally related chemicals and scientific 36 judgement may be employed to select uncertainty factors that provide protection for the public 37 health. 38 39 2.5.3.3.4 Estimating the Range of Variability in a Human Population 40 41 The NAC/AEGL Committee estimates the range in variability of response to specific 42 chemical exposures primarily on the basis of quantitative human data. Acceptable experimental Sop08-02 wpd Printed July 6. 2000 65 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 data are more likely to be available for AEGL-1 and AEGL-2 endpomts than for AEGL-3 2 endpoints. For example, numerous studies have considered induction of bronchospasm after 3 controlled exposures to sulfur dioxide in asthmatics and non-asthmatics. There is marked 4 individual variability in the severity of reaction to inhalation of low concentrations of sulfur 5 dioxide. Asthmatics, individuals with hyper-reactive airways, smokers and those with chronic 6 respiratory or cardiac disease react to relatively lower concentrations (Aleksieva, 1983; Simon, 7 1986). Susceptibility may also be increased in people over 60 years of age, but reports have not 8 been consistent (Rondinelli et al., 1987; Koenig, et al., 1993). By contrast, comparable human 9 data for AEGL-3 tier concentrations are limited to anecdotal case reports. 10 11 For example, during the course of the Committee's deliberations on phosphine AEGL 12 development, the possibility that children are more susceptible to phosphine exposure was 13 suggested by two case reports describing the deaths of children, but not adults, after 14 "comparable" phosphine exposures. As with most case reports, the exposure concentrations were 15 not quantified. However, both the children and the adults in question were present in somewhat 16 restricted environments, suggesting comparable exposure levels. Based on these case reports, the 17 Committee concluded that children may be more sensitive to phosphine exposure and selected 18 uncertainty factors that would provide additional protection for children. 19 20 In cases where quantitative human data are lacking for specific chemicals, but adequate 21 data can be found for structurally or mechanistically similar agents, uncertainty factors may be 22 selected by analogy to structurally similar chemicals and/or mechanism of action. For example, 23 asthmatics are particularly sensitive to sulfur dioxide. Declines of >20% in FEV1 have been 24 documented after inhalation of 0.4-1 ppm for 2-15 minutes. The effects of sulfur dioxide 25 exposure are enhanced in normal and asthmatic individuals by moderate exertion (ventilation 26 >40 1/m with mouth breathing), hyperventilation, and use of oral airways (Horstman, et al., 1988; 27 Frank, 1980; Koenig, et al., 1981; Koenig, et al., 1982; Balmes, et al., 1987; Linn, et al., 1987; 28 Roger, et al., 1985). Duration of bronchospasm is generally limited and these patients may 29 develop tolerance with prolonged or repeated exposure. These studies suggest that 30 mouth-breathing asthmatics exposed to sulfur dioxide develop bronchospasm at levels of 31 approximately 33 percent of comparably exposed non-asthmatics. Accordingly, the Committee 32 generally has used an uncertainty factor of 3 when considering the differences in human 33 susceptibility to most respiratory irritants. However, the NAC/AEGL Committee is aware that 34 the variation in response of asthmatics may differ among respiratory irritants ranging from mild 35 to severe in their effects. The most appropriate uncertainty factor will be considered based upon 36 the degree of severity of the imtant chemical and the biological data available, for known or 37 suspected differences among humans. 38 39 Children and infants are often considered as susceptible populations. There is a general 40 belief that children and infants are more susceptible to the effects of toxic substances than adults. 41 Much of this belief is predicated upon the fact that children, and particularly infants, possess 42 immature or developing biochemical, physiological, and anatomical systems that are not Sop08-02 wpd Printed July 6. 2000 66 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30,2000 I adequate to combat the adverse affects of toxic chemicals. Further, it is believed that in certain 2 instances, the toxic effects of chemicals may permanently damage or alter the growth and 3 function of developing organs and organ systems in the young. The potential for greater 4 sensitivity to chemical substances by children and infants has been reviewed by the National 5 Research Council of the National Academy of Sciences (NRC, 1993b). The report indicates that 6 there are limited data on the relative toxiciry of pesticides and other xenobiotic compounds in 7 immature and mature humans. Consequently, the NRC focused on laboratory animal studies, 8 age-related pharmacokinetic and pharmacodynamic differences, and pharmacological data from 9 controlled clinical investigations with humans. The NRC concluded that the mode of action is 10 generally similar in mammalian species and across age and development stages within species. 11 They also concluded that children may be more sensitive or less sensitive than adults to pesticide 12 toxicity, depending on the chemical, but that the quantitative differences in toxicity between the 13 age groups are usually less than a factor of approximately 10-fold. 14 15 Although many reports have been published on the pharmacokinetic differences of 16 pharmacologic agents and other chemicals in children and adults, the data cannot be translated 17 into meaningful dose-response relationships to make valid quantitative comparisons in the 18 absence of specific biologically relevant endpoints. Bruckner and Weil (1999) summarized the 19 biological factors which may influence the responses of adolescents to chemical exposure. 20 21 Based on the limited data available, the extent to which significant differences in the 22 susceptibility of children/infants and adults exists is largely unknown. However, the difference is 23 generally considered to be within a factor of 10-fold (NRC, 1993b) with most of the differences 24 in susceptibility on the order of 2-3 fold. It is highly probable that any differences are 25 chemical-specific and also related to specific developmental stages of children and infants. 26 Within the context of the AEGL program, this issue is further complicated by the consideration 27 of once-in-a-lifetime inhalation exposures of 1 hour or less to 8 hours. The discussion at the 28 beginning of this section indicates that there is a paucity of data on age related differences and 29 the young can be more or less susceptible than adults to exposure to chemicals, depending upon 30 the chemical or chemical class in question. However, it is believed that uncertainty factors 31 applicable to other sensitive subpopulations are adequate to protect children and infants with 32 decisions based on a weight-of-the-evidence on a chemical-specific basis. It is important that all 33 of the relevant information on the chemical be considered when making judgements about 34 selection of the appropriate uncertainty factors for age differences and all other factors that 35 contribute to differences in susceptibility. 36 37 In summary, the maximum variation in responses of susceptible populations are believed 38 to generally range within 3 to 10-fold of a for healthy individuals. All information on the 39 chemical, including its mechanism of action, the biological responses, and data on structurally 40 related chemical analogs is considered as well as informed professional judgement when 41 determining appropriate uncertainty factors. Information about similarities and differences in 42 toxicokinetics and toxicodynarmcs are used where available to modify as necessary the Sop08-02 wpd Printed July 6, 2000 67 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30.2000 1 qualitatively and quantitatively as compared to non-sensitive individuals. 2 3 2.5.3.4.3 Age/Life Stage/Condition Differences 4 5 When available data indicate certain age groups may be uniquely sensitive as contrasted 6 to the general population, an intraspecies uncertainty factor of 10 is generally used. 7 8 THE RATIONALE FOR THE SELECTION OF THIS UF SHOULD INCLUDE: 9 1. Description of the toxicologic endpoints which differ between humans of different age 10 groups. 11 2. Discussion of the magnitude of this difference. For example, quantitatively, how 12 much does the response differ, or what qualitative information indicates there may 13 be differences among age groups? 14 15 16 2.5.3.4.4 Response by Normal and Sensitive Individuals to Chemical 17 Exposure is Unlikely to Differ for Mechanistic Reasons IS 19 In cases where the mode or mechanism of action is such that the response elicited by 20 exposure to the chemical by different subpopulations is unlikely to differ, an intraspecies 21 uncertainty factor of 3-fold is generally used. Typically this involves a direct acting mechanism 22 of toxicity where metabolism is unlikely to play a major role. A steep dose response curve may 23 also be an indication of little variation within a population, and is factored into the weight-of- 4 evidence considerations for UF determination. 25 26 THE RATIONALE FOR THE SELECTION OF THIS UF SHOULD INCLUDE: 27 1. Description of the mechanism of action. 28 2. Discussion of why the response to chemical exposure is unlikely to differ and whether 29 metabolism/detoxification is likely to be an issue. 30 31 32 2.5.3.4.5 Mode or Mechanism of Action is Unknown 33 34 When the mode or mechanism of toxic action is uncertain, or unknown metabolic factors 35 may play an important role, and/or a broad range of responses to chemical exposure is observed, 36 there is concern that there may be large differences in susceptibility between individuals. In 37 these cases an intraspecies uncertainty factor of 10 may be applied. 38 39 THE RATIONALE FOR THE SELECTION OF THIS UF SHOULD INCLUDE: 40 1. Description of the toxicity reported and the uncertainty associated with the chemical's 41 mechanism of action or other factors. 42 2. Statement as to why the effects seen add uncertainty to the assessment. SopOB-02 wpd Printed July 6. 2000 69 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30,2000 1 2 2.5.3.4.6 Uncertainty Factors Which Result in AEGL Values That Conflict 3 with Actual Human Exposure Data 4 5 When AEGL values are initially proposed, the candidate range of values are compared to 6 the known spectrum of supporting data on the chemical. In a weight-of-the-evidence approach, 7 conflicts between the candidate AEGLs (generally derived from animal data) and the supporting 8 data (either animal data or human data) may lead to the conclusion that the uncertainty factors 9 utilized in the calculations are inappropriate because they conflict with other specific and highly 10 relevant data. In this case, the candidate AEGLs are revised to reflect the supporting data. In 11 other cases where the AEGL may conflict with an existing standard or guideline, the comparative 12 basis of the two values may be evaluated to see if the discrepancy is justified or resolvable. 13 14 THE RATIONALE FOR THE SELECTION OF THIS UF SHOULD INCLUDE: 15 1. A statement on why the use of uncertainty factors initially selected conflict(s) with the 16 published evidence. 17 Sop08-02.wpd Printed July 6, 2000 70 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2.6 GUIDELINES/CRITERIA FOR SELECTION OF MODIFYING 2 FACTORS 3 4 2.6.1 Definition 5 6 In addition to the uncertainty factors discussed above, an additional Modifying Factor 7 may be necessary when an incomplete data base exists. Hence, the modifying factors represent 8 an adjustment for uncertainties in the overall database or for known differences in toxicity among 9 structurally similar chemicals. The modifying factor "... reflects professional judgment of the 10 entire data base available on the specific agent" and is applied on a case by case basis (NRC, 11 1993a, p88). The Modifying Factor may range from Ito 10-fold. The default value is 1-fold. 12 13 2.6.2 Use of Modifying Factors to Date in the Preparation of AEGL Values 14 15 Modifying factors have been used for chemicals currently published "Final" by the U. S. 16 National Academy of Sciences. Modifying factors of 2 or 3 are under consideration for 17 chemicals currently undergoing review to account for (1) a limited data set, (2) instances where 18 the adverse effects used to set the AEGL level are more severe than those described in the AEGL 19 definition, and (3) to account for the differential toxicity of chemical isomers. 20 Sop08-02 wpd Printed July 6, 2000 71 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2.7 GUIDELINES/CRITERIA FOR TIME SCALING 2 3 Acute Exposure Guideline Levels (AEGLs) are derived for 30-minute, 1-hour, 4-hour, 4 and 8-hour exposure durations to meet a wide range of needs for government and private sector 5 organizations. AEGLs for 10 minute exposure durations will be developed for all future 6 chemicals addressed by the NAC/AEGL Committee, and 10 minute AEGLs will be developed 7 for the first six chemicals published by the U. S. NAS in the near future. Experimental animal 8 and controlled human exposure-response data and data from human exposure incidents often 9 involve exposure durations differing from those specified for AEGLs. Therefore, extrapolation 10 from the reported exposure duration and chemical concentration of a toxic endpoint to an 11 equivalent concentration for an AEGL-specified period is usually required. The discussion in 12 this section covers the concept, the published scientific literature, the methodologies used for 13 extrapolation, and examples of the application of these methodologies to specific chemicals for 14 the development of AEGL values. 15 16 2.7.1 Overview 17 18 In accordance with the needs of stakeholders, AEGLs are derived for 30-minute, 1-hour, 19 4-hour, and 8-hour exposure durations. Stakeholders have requested that the NAC/AEGL 20 Committee also develop 10 minute AEGL values. AEGL values for 10 minute durations will be 21 developed for chemicals in future U. S. NAS publications. Experimental exposure-response 22 data from animal studies or human exposure incidents often involve exposure durations differing 23 from AEGL-specified durations or may coincide with only one or two AEGL exposure durations. 24 Therefore, extrapolation from a reported toxic endpoint concentration and exposure duration to 25 an equivalent concentration for an AEGL-specified exposure period is usually required. 26 27 The 1993a NRC/NAS guidelines for developing short-term exposure limits address the 28 extrapolation of the effects of genotoxic carcinogens from long-term to short-term exposures. 29 Only limited NRC/NAS guidance is provided for approaches or methodologies for the 3 0 extrapolation of reported acutely toxic effects to shorter or longer durations of exposure. 31 Therefore, the NAC/AEGL Committee and ORNL have reviewed the scientific literature related 32 to time exposure relationships and current approaches and methodologies used for time-scaling. 33 Documented here are the NAC/AEGL Committee's approaches to making exposure duration 34 adjustments to develop of AEGL values from 10 minutes to 8 hours. This approach also has 35 been reviewed by scientists representing certain OECD-member countries. 36 37 The relationship between dose and time for any given chemical is a function of the 38 physical and chemical properties of the substance and the unique lexicological and 39 pharmacological properties of the individual substance. Historically, the relationship according 40 to Haber (1924), commonly called Haber's Law (NRC, 1993a) or Haber's Rule (i.e., Cxt-k, 41 where C = exposure concentration, t = exposure duration, and k = a constant) has been used to 42 relate exposure concentration and duration to effect (Rinehart and Hatch, 1964). This concept Sop08-02 wpd Printed Jufy 6.2000 11 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 states that exposure concentration and exposure duration may be reciprocally adjusted to 2 maintain a cumulative exposure constant (fc) and that this cumulative exposure constant will 3 always reflect a specific quantitative and qualitative response. This inverse relationship of 4 concentration and time may be valid when the toxic response to a chemical is equally dependent 5 upon the concentration and the exposure duration. However, an assessment by ten Berge et al. 6 (1986) of LCJO data for certain chemicals revealed chemical-specific relationships between 7 exposure concentration and exposure duration that were often exponential. This relationship can 8 be expressed by the equation C" x t = k, where n represents a chemical specific, and even a toxic 9 endpomt specific, exponent. The relationship described by this equation is basically the form of 10 a linear regression analysis of the log-log transformation of a plot of C vs t (see Curve Fitting and 11 Statistical Testing of the Generated Curve below). Ten Berge et al. (1986) examined the airborne 12 concentration (C) and short-term exposure duration (t) relationship relative to death for 13 approximately 20 chemicals and found that the empirically derived value of n ranged from 0.8 to 14 3.5 among this group of chemicals (See Table 2.7-1). Hence, these workers showed that the 15 value of the exponent («) in the equation C1 x t = k quantitatively defines the relationship 16 between exposure concentration and exposure duration for a given chemical and for a specific 17 health effect endpoint. Haber's Rule is the special case where n = 1. As the value of n increases, 18 the plot of concentration vs time yields a progressive decrease in the slope of the curve. 19 20 In cases where adequate data are available, the NAC/AEGLCommittee conducts an 21 analysis of chemical-specific toxicity and exposure data to denve a chemical-specific and health 22 effect-specific exponent (n) for use in extrapolating available exposure data to AEGL-specified 23 exposure durations. If data are not available for empirically deriving the exponent n, the 4 NAC/AEGL Committee identifies the most appropriate value for n by comparing the resultant ^5 AEGL values derived using n=l and n=3. The value of n=l has been used historically by others 26 and results in rapid reductions in concentrations when extrapolations are made to longer 27 exposure periods and rapidly increasing concentrations when extrapolated to shorter exposure 28 periods. Based on the work often Berge et. al. (1986), 1 represents the estimate of the lower 29 boundary of the value of n. The value of n=3, an estimate of the upper boundary of the value of n 30 (ten Berge, 1986), results in less rapid rates of decrease in estimated effect concentrations when 31 extrapolations are made to longer exposure periods and to less rapid rates of increase in 32 estimated effect concentrations when extrapolated to shorter exposure periods. This range of 33 values in n from 1 to 3 encompasses approximately 90 percent of the chemicals examined by ten 34 Berge et al. (1986). In selecting a value for n when the derivation of n is not possible, the 35 NAC/AEGL Committee evaluates the resultant AEGL values determined with either the upper or 36 the lower boundary value of n (1 or 3) within the context of other supporting data to determine 37 the reasonableness of the extrapolated AEGL value. A value of n=l is used when extrapolating 38 from shorter to longer exposure durations and a value of n=3 when extrapolating from longer to 39 shorter durations. The resultant AEGL value is then compared to supporting data to determine 40 the scientific reasonableness of the derived AEGL value. A consensus of the Committee 41 generally favors the use of a value for n that results in an AEGL value that best fits the 42 supporting data. Sop08-02 wpd Printed July 6, 2000 73 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 In summary, analyses of relevant data, together with scientific judgement are used to 2 determine the extent of temporal extrapolation and its validity in AEGL derivations. For 3 example, extrapolation of 10-minute exposure data to a 4 or 8-hour AEGL value requires more 4 supporting data and or/assumptions than the extrapolation of 10-minute exposure data to a 30- 5 minute or 1-hour AEGL. Errors in the estimated exposure concentration-exposure duration 6 relationship (i.e., the value of n) will progressively increase the magnitude of the error of the 7 derived AEGL value as the time from the empirical data point to the extrapolated data point 8 increases. Since toxicity data are often not available for any or all of the AEGL-specified time 9 periods, temporal extrapolation is usually necessary to generate scientifically credible values for 10 the AEGL time points. 11 12 2.7.2 Summary of Key Publications on Time Scaling 13 14 Several investigators have studied the relationship of exposure duration and exposure 15 concentration as related to the toxic response to airborne chemicals (Haber, 1924; Flury, 1921; 16 Rinehart and Hatch, 1964; ten Berge et al., 1986; ECETOC, 1991, and Pieters and Kramer, 17 1994). 18 19 Based on observations and studies with chemical warfare gases such as phosgene, Haber 20 (1924) found that for certain chemicals the product of the exposure duration multiplied by the 21 exposure concentration was constant for a specific response or toxic endpoint (i.e., lethality). In 22 experiments with cats, Haber found that a specific concentration x time product would result in 23 100% lethal response and that as long as this product value was maintained, regardless of the 24 specific exposure concentration or duration, the response was consistent. This linear relationship 25 became known as Haber's Rule; or Cx t = k where C = concentration of the chemical of the 26 chemical in question, / = exposure duration, and k = a cumulative exposure constant. Similarly, 27 Flury (1921) found that inhalation of phosgene exhibited a linear relationship, Cx t = E, where 28 £ represents the onset of pulmonary edema. Obviously, the cumulative exposure constant may 29 relate to any number of responses or toxic endpoints. However, the information reported by 30 Haber is limited to a small number of chemicals or chemical classes and substantial quantitative 31 data derived from controlled studies is lacking. 32 33 Historically Haber's Rule has been used for time concentration extrapolations U. S. EPA 34 (1994b). This relationship assumes that each unit of damage is irreversible, that no repair takes 35 place during the exposure period and, therefore, that each unit of exposure is 100 percent 36 cumulative. However, this is generally not the case for acutely toxic responses to short-term 37 exposures. The relationship between concentration and duration of exposure as related to 38 lethality was examined by ten Berge et al. (1986) for approximately 20 irritant or systemically- 39 acting vapors and gases. The authors subjected the entire individual animal data set to probit 40 analysis with exposure duration and exposure concentration as independent variables. They used 41 the methodology of Finney (1971) to investigate the fit of the data into a probit model on the 42 basis of a maximum likelihood estimate. In re-evaluating the raw data for these chemicals, it was Sop08-02wpdPnnted July 6. 2000 74 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 found that the linear relationship described by Haber's, C x t = k, was not always a valid 2 predictor of lethality. An exponential function (C1 x t = k), where the value of n ranged from 0.8 3 to 3.5 for different chemicals, was a more accurate quantitative descriptor These authors 4 derived empirically-based, chemical-specific regression coefficients for exposure duration and 5 exposure concentration, as well as chemical-specific values for n. The values for n for the 20 6 chemicals studied ranged from 0.8 to 3.5. The analyses indicated that the concentration-duration 7 relationship for lethality was described more accurately by the exponential function (C1 x t = K) 8 and that Haber's Rule was appropriate for only a limited number of the chemicals. Based upon 9 the results of the analyses, ten Berge et al. (1986) concluded that the concentration-time 10 relationship (i.e., value for n) should be determined empirically from acute inhalation exposure 11 toxicity data on a chemical-specific basis. 12 13 2.7.3 Summary of the Approaches that may be Taken for Time Scaling 14 15 A tiered approach to generating toxicity values for time scaling is taken by the 16 NAC/AEGL Committee to derive AEGL values from empirical data. This approach is 17 summarized below. Each of the approaches and the circumstances under which they are, or 18 could be, used are discussed subsequently in this section of the SOP Manual. 19 20 (1) If appropriate lexicological data for the exposure concentration-exposure duration 21 relationship of a specific health effect endpoint are available for the AEGL-specified 22 exposure periods, use the empirical data directly. 23 24 (2) If empirical exposure concentration-exposure duration relationship data are available, 25 albeit they do not coincide with AEGL - specified exposure periods, use the available 26 data to derive values of n and extrapolate the AEGL values using the equation C" x t = k. 27 28 (3) If no empirical exposure concentration-exposure duration relationship data are 29 available to derive a value of n, a value of n=l for extrapolating from shorter to longer 30 exposure durations, and a value of n=3 for extrapolating from longer to shorter exposure 31 durations, should be selected initially. The scientific reasonableness of the selection of 32 the estimated lower and upper boundaries of n (n=l and n=3) is then evaluated by 33 comparing the resultant AEGL values with all other supporting data. If appropriate, the 34 final value(s) of n may be modified to reconcile differences between extrapolated AEGL 35 values and the supporting data. 36 37 (4) If there are no supporting data to evaluate selected values of n, a value of n=l for 38 extrapolating from shorter to longer exposure periods and a value of n=3 for extrapolating 39 from longer to shorter exposure periods should be selected. In the absence of other data, 40 the resultant AEGL values are thought to be protective and scientifically credible. 41 42 The balance of this section of the guidance will provide more detailed information on the Sop08-02 wpd Printed July 6, 2000 75 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 approaches stated above. 2 3 2.7.4 Use of Empirical Data that is Available for AEGL-Specified Exposure 4 Durations 5 6 If toxicity data are available for all four AEGL-specified exposure periods, there is no 7 need to derive values of n and the data for each exposure period can be used directly. However, 8 it is rare that toxicity data are sufficiently comprehensive to encompass all of the AEGL- 9 specified exposure periods from 10 minutes to 8 hours. Further, there are instances where 10 empirical data are not available to estimate n and predict the exposure concentration-exposure 11 duration relationship using Cn x t = k. Therefore, the sequential approaches used by, or available 12 to the NAC/AEGL Committee to establish AEGL values for the specified exposure periods are 13 discussed in the following sections. 14 15 2.7.5 Derivation of Values of n When Adequate Empirical Data are Available 16 for Other than the AEGL-Specified Exposure Durations 17 18 A key element in the procedure of time-scaling is the use of a value or values for n in the 19 equation C" x t = k. If empirical data for exposure durations other than the AEGL-specified 20 exposure periods are available to quantify the exposure concentration - exposure duration 21 relationships for a health effect endpomt, including lethality, the value of n should be derived 22 using the method of calculation described in this section. It is believed empirically derived 23 values of n are scientifically more credible than simply choosing n=l (Haber's Rule) or 24 attempting to select some other value of n. 25 26 2.7.5.1 Selection of Appropriate Health Effect End Point for Deriving a Value 27 for n 28 29 The first step in any time scaling methodology is the selection of the health effect 30 endpoint of concern. Clearly the health effect endpoint selected should be consistent with the 31 definition of the AEGL tier being determined. Further, the endpoint should be unambiguous and 32 consistently observed at all reported exposure durations. For example, death is an unambiguous 33 endpoint and a quantitatively determined index of toxicity, the LC50, is a response rate which can 34 be compared reliably among exposures at different time periods. The use of the LC50 as an index 35 of toxicity is ideal because it is a statistically derived concentration which is not subject to the 36 vagaries of dose selection and exhibits less variability in response than any other experimental 37 endpoint. Death is included in the AEGL-3 definition and is used for estimating the value of n. 38 39 A comparable endpoint for the AEGL-1 and AEGL-2 tiers would be an ED50 (the dose 40 which causes a specific response in 50 percent of the animals) for a precisely defined toxic or 41 health effect endpoint that is consistent with the definition of the AEGL tier in question. The Sop08-02 wpd Printed July 6, 2000 76 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 actual endpoint is often difficult to determine in most experiments because the observed effects 2 are often a continuum from mild to severe and generally not precise enough to determine an ED50 3 value with reliability. Further, incidence data for non-lethal effects is not always reported. For 4 these reasons, the concentration/response relationship and the value of n derived from lethality 5 data have often been applied to both the AEGL-3 and the AEGL-2 exposure period 6 extrapolations. However, in instances where the mechanism of toxicity causing the health effect 7 of concern at the AEGL-2 tier is thought to be different than that which causes lethality, the value 8 of n derived from LCSO data should not be used. Under these circumstances, AEGL-2 values can 9 be developed by selecting the upper and lower boundaries of n (n=3 and n=l) for extrapolation 10 from longer to shorter and shorter to longer exposure periods, respectively. The resultant AEGL- 11 2 values should be evaluated within the context of other supporting data to evaluate the 12 reasonableness of the values of n selected. In the absence of supporting data, the AEGL values 13 determined using n=3 and n= 1 should be utilized. 14 15 Selection of appropriate endpoints for AEGL-1 values per se represents a unique and 16 often difficult task. Based on the NAC/AEGL Committee's experience to date, no rigorous data 17 for any chemical have been available from which values of n could be denved for the AEGL-1 18 type of endpoints. The derivation of AEGL-1 values is discussed later in this section. 19 20 2.7.5.2 Criteria for Adequate Empirical Data for Deriving Values of n 21 22 After determining the health effect endpoint to be used in deriving the value(s) for n, the 23 next step is to evaluate the quality and the quantity of the data to be used in the derivation. 24 Obviously, two data points will define the slope of a curve descnbing the exposure 25 time/exposure concentration relationship. However, the validity and, hence, the value(s) of n 26 will depend on many factors including the scientific soundness of the concentration exposure- 27 duration data, the length of the empirical exposure duration(s) relative to the AEGL-specified 28 exposure periods, and the known or perceived similarities in effects and mechanism of action of 29 the chemical at the reported exposure concentrations and durations. Generally three empirical 30 data points will improve the scientific validity of the slope and the estimated values(s) for n, and 31 the validity is likely to increase with an increase in the number of empirical data points used to 32 derive n, provided that there is a reasonable fit of these data points. 33 34 2.7.5.3 Curve Fitting and Statistical Testing of the Generated Curve 35 36 Once the health effect endpoint and data points describing the concentration-exposure 37 duration relationship have been selected, the values are plotted and fit to a mathematical equation 38 from which the AEGL values are developed. There may be issues regarding the placement of the 39 exponential function in the equation describing the concentration-exposure duration relationship 40 (e.g. C'xt = kvsCxfn = k2vsC*x? = k}). It is clear that the concentration-exposure duration 41 relationship for a given chemical is directly related to its pharmacokmetic and pharmacodynamic 42 properties. Hence, the use and proper placement of an exponent or exponents to quantitatively Sop08-02wpd Printed July 6. 2000 77 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 describe these properties is highly complex and not well understood. 2 3 The quantitative description of actual empirical data of this relationship can be expressed 4 by any of a number of linear regression equations. In the assessment of empirical data reported 5 by ten Berge et al. (1986) these workers quantified the concentration-exposure duration 6 relationship by varying the concentration to the nlh power. Since raising c or t or both to a power 7 can be used to quantitatively define the same relationship or slope of the curve, and to be 8 consistent with data and information presented in the peer reviewed scientific literature, the 9 equation C" x t = k is used for extrapolation. It must be emphasized that the relationship between 10 C and t is an empirical fit of the log transformed data to a line. No conclusions about specific 11 biological mechanisms of action can be drawn from this relationship. 12 13 The preferred method is to use a statistical methodology which utilizes all of the 14 individual animal data and generates a maximum likelihood estimate with 95% confidence 15 limits. Where individual animal data are available, the NAC/AEGL Committee will explore 16 using the methodology of Finney (1971). This methodology has been incorporated into a 17 computer program and provided to the Committee by Dr. ten Berge from the Netherlands. 18 19 Unfortunately, the individual animal data are often not available and only LC50 values are 20 listed. In this case a linear regression analysis of the log-log transformation of the 21 concentration/time data will be performed as described below. 22 23 When time-concentration data are plotted on a log-log plot, they generally fall along a 24 straight line. For that reason a simple linear regression (Alder and Roessler, 1968) is run on the 25 data to generate the mathematical curve. The basic linear regression equation is in the form: 26 27 Y = a + bX 28 29 where Y is the predicted value of the dependent variable, X is the value of the independent 30 variable, a is the Y intercept and b is the slope of the line. 31 32 This is the form of the log-transformation of the nonlinear C" * t = k equation to a linear 33 equation (see below): 34 35 log C = (log k)/n + (-1/n)* log t 36 37 where C is the predicted value of the concentration to cause an effect at exposure duration t. The 38 (logk)/n is the Y intercept of the plot of logC against logT", and -1/n is the slope of the plot of 39 logC against Iog7. 40 41 C"*t = k 42 logfC1 * t) = log k SopOB-02 wpd Printed July 6, 2000 78 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 n*logC + logt = log k 2 n*logC-logk- log t 3 logC= (log k)/n - (log t)/n 4 log C = (log k)/n - (1/n) * log t 5 6 The regression coefficient or slope, b, returns the slope of the linear regression line 7 through data points X and Y. The slope (rate of change along the regression line) is the distance 8 between the Y values of the two points divided by the distance between their respective X values. 9 The regression coefficient is calculated as: 10 11 , _ N^XY - (LX) (LY) where N = the number of observations 19 or 20 NLflog t)(log C) - (Llog t) (Llog Q _1/n = NL(logtf-(Llog(? 21 The above is solved in a spreadsheet for n. 28 29 The validity of the derived value(s) of n is dependent on the degree of correlation among 30 the various concentration/time data points used to construct the curve and the equation. 3 1 Normally a coefficient of determination (r2) is calculated as a measure of how well the generated 32 curve (linear in this case) fits the data points. If r2 = 0 the data do not fit a linear relationship. If 33 r2 = 1 the data exhibit a strong linear relationship. If the number of data points are 3 and the real 34 value of r = 0 "... the chance of obtaining a fairly high correlation coefficient for the sample is 35 greater than the chance of obtaining a small correlation coefficient." (Alder and Roessler 1968, 36 p!91) If the number of data points are 4 "... the chance of obtaining a particular correlation 37 coefficient is equal to that of obtaining any other." (Alder and Roessler 1968, pi 91). Since the 38 number of data points typically available are only in the range of 3 or 4 values, the use of r2 to 39 measure how well the data fit the generated curve is not a meaningful test to perform. Therefore 40 informed professional judgement is exercised by the NAC/AEGL Committee. 41 42 Given the fact that the distribution of r for low numbers of observations (typically 3 or 4 43 data points for time scaling) cannot be fit to a normal curve, meaningful statistical tests of the fit 44 of the regression line (used to derive n) to the data cannot be performed. Even with these 45 shortcomings, a regression analysis of the data as previously described gives the best fit of a line 46 to the data. A visual inspection of the regression line vs the data also will show the 47 reasonableness of the fit and, hence, the reasonableness of the derived value for n. This is 48 generally the best approach empirical data are used to derive n values for developing AEGL Sop08-02 wpd Printed July 6. 2000 79 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30. 2000 1 values for specified exposure durations. As stated earlier, it must be emphasized that: When 2 deriving or selecting a value for n, the NAC/AEGL Committee evaluates the resultant 3 AEGL values within the context of other supporting data to determine the reasonableness 4 of the extrapolated values. This is true even when the value of n is derived from empirical 5 data that describes the concentration-exposure duration relationship. The NAC/AEGL 6 Committee uses a value for n that results in AEGL values that best fit the supporting data. 7 Therefore, there is no substitute for informed professional judgement based on careful review, 8 evaluation and discussion of all available data. 9 10 2.7.5.4 Examples of NAC/AEGL Committee Derivations of Values of n from 11 Empirical Data 12 13 During the course of AEGL development, the NAC/AEGL Committee has used 14 empirically-based derivations of n in the equation C" x t = k for time-scaling to AEGL-specified 15 exposure penods. Guidelines have been developed from this experience and are presented in the 16 final part of this section. 17 18 19 2.7.6 Selection of Values of n When Adequate Empirical Data are Not 20 Available to Derive Values for n 21 22 When adequate data describing concentration-exposure duration period relationships for a 23 specific chemical and toxic endpomt of interest are not available, an alternative approach to 24 quantitatively estimating this relationship must be followed. The approach used by the 25 NAC/AEGL Committee involves the application of the equation Cn x t = k and the selection of a 26 value or values of n that results in AEGL values that best fit the supporting data for the chemical 27 and toxic endpoint in question. It is important to distinguish the difference between the 28 derivation of values of n as described in the preceding section and the selection of values of n as 29 described in this section. 30 31 An evaluation of the analysis of values of n by ten Berge et al. (1986) served as the basis 32 to select the limits used by the NAC/AEGL Committee. 33 34 Table 2.7-1 is a summary of the airborne concentration-exposure duration relationships 35 for 20 chemicals based on their LC50 values. 36 Sop08-02 wpd Printed July 6, 2000 80 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 TABLE 2.7-1. VALUES OF n FROM TEN BERGE ET AL. (1986). 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 §i 33 34 35 The lowest value of n was 0.8 and the highest value of n was 3.5. Approximately 90 36 percent of the values of n range between n=l and n=3. Consequently, these values were selected 37 as the reasonable lower and upper bounds of n. 38 39 In the absence of data to derive a value for n, the NAC/AEGL Committee selects values 40 for n of 1 and 3, depending on an extrapolation from shorter to longer durations or longer to 41 shorter durations. The value of n is then used in the equation Cn x t = k to extrapolate from 42 empirically reported concentration and exposure durations to the AEGL-specified exposure 43 duration(s). The Committee then selects the derived AEGL values in accord with the supporting Sop08-02 wpd Printed July 6. 2000 81 SYSTEMIC CHEMICALS HCN H2S methyl t-butyl ether methylenechlorobromide ethylenedibromide tetrachloroethylene trichloroethylene carbon tetrachloride acrylonitrile Value of n(ave) 2.7 2.2 2 1.6 1.2 2 0.8 2.8 1.1 IRRITANTS ammonia HC1 chlorine pentafluoride nitrogen dioxide chlorine perfluoroisobutylene crotonaldehyde HF ethylene inline bromine dibutylhexamethylenediamine Range of n # Chemicals/range 0.8-1.5 8 1.51-2.0 6 2.01-2.5 2 2.51-3.0 2 3.01-3.5 2 2 1 2 3.5 3.5 1.2 1.2 2 1.1 2.2 1 Cumulative # chemicals 8 14 16 18 90%withn<3 20 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 data. 2 3 4 2.7.6.1 Selection of Values of n When Extrapolating from Shorter to Longer 5 Exposure Periods 6 7 As discussed previously, a value of n=l represents the lower range of the concentration- 8 exposure period relationship. If the exponent n=l is used in the equation C"xt = k, there is a 9 rapid decrease in extrapolated values when extrapolations are made from shorter to longer 10 exposure periods (see Figure 2.7-1). The extrapolated values are lower and, hence, represent a 11 conservative estimate of the AEGL value. A value of n=3 represents a value in the upper range 12 for the concentration-exposure duration relationship and results in a less rapid rate of decrease 13 when extrapolating from shorter to longer exposure periods. Therefore, the extrapolated AEGL 14 values for longer exposure periods are higher and, hence, less conservative in terms of protecting 15 human health. See Figure 2.7-1. 16 17 When data are not available for deriving a value of n, the NAC/AEGL Committee 18 develops tentative AEGL values from shorter to longer exposure durations using n=l in the 19 equation C1 x t = k and evaluates these values with all other supporting data to determine their 20 scientific reasonableness. Therefore a "weight of evidence" test is applied to the tentative 21 AEGLs by comparing these values to the supporting data to determine the most scientifically 22 credible AEGL values. In instances where the supporting data indicate that the tentative AEGL 23 developed using a value of n=l is too low or too high, the AEGL may be adjusted to 24 scientifically accommodate the supporting data. If there are no supporting data indicating that 25 the derived AEGL should be adjusted, a value of n=l is used to account for the uncertainty of the 26 concentration-endpoint relationship at longer exposure durations. 27 28 2.7.6.2 Selection of Values of n When Extrapolating from Longer to Shorter 29 Exposure Periods 30 31 When extrapolating from longer to shorter exposure durations using the equation Cn x t = 32 k and a value of n=l, there is a relatively rapid increase in the extrapolated values (see Figure 33 2.7-1). Under these circumstances, the derived AEGL value represents a relatively high estimate 34 of the toxic endpomt concentration at shorter exposure durations and is, therefore, a less 35 conservative value. When extrapolating from longer to shorter exposure durations using a value 36 of n=3, there is a less rapid rate of increase in the derived AEGL value. As a result, the 37 extrapolated AEGL value is more conservative when selecting a value of n=3. See Figure 2.7-1. 38 39 Under circumstances where the NAC/AEGL Committee selects a value for n to derive 40 AEGL values from empirical data for longer exposure periods, tentative AEGLs are derived 41 using values for n of 3 and then compared to the derived values with all other relevant data. Sop08-02 wpd Printed July 6. 2000 82 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Again, this represents a "weight of evidence" approach to selecting a value of n for the most 2 scientifically credible AEGL values. In instances where the supporting data indicate that the 3 tentative AEGL developed using a value of n=3 is too high or too low, the AEGL may be 4 adjusted to scientifically account for the supporting data. If there are no supporting data 5 indicating that the derived AEGL should be adjusted, a value of n=3 should be used to 6 accommodate for the uncertainty of the concentration-exposure duration relationship for the 7 shorter exposure durations. 8 Sop08-02 wpd Printed July 6, 2000 83 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 Extrapolation from a 60 minute value of 100 ppm to 30 and 480 minutes with different values of n 200 150 -- O E 100 -- CL o. 50 -- 200 126 * * — t = k n=3 n=2 n=1 50 35 12.5 0 60 120 180 240 300 Minutes (t) 360 420 480 1 2 3 4 5 6 7 8 9 10 11 12 FIGURE 2.7-1 EFFECTS OF VARYING n IN THE EQUATION C" x t = k SHORT TO LONG DURATION EXTRAPOLATIONS: Note that when extrapolating from 60 minutes to longer exposure durations, the lower the value of n the lower the extrapolated value Therefore, when extrapolating from short to long exposure durations, a value of n=l yields a more conservative value than any value of n that is >1 LONG TO SHORT DURATION EXTRAPOLATIONS. Conversely, when extrapolating from 60 minutes to shorter exposure durations, the higher the value of n the lower the extrapolated value Therefore when extrapolating from long to short exposure durations a value of n=3 yields a more conservative value than any value of n that is <3 Sop08-02 wpd Printed July 6. 2000 84 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2.7.7 Special Considerations in the Time Scaling of AEGL-1 and AEGL-2 2 Values 3 4 The previous descriptions of approaches to time scaling for toxic end-point 5 concentrations are most applicable to the derivation of AEGL-3 values. This is because 6 unequivocal data relating the concentration required to cause an effect to the time duration of 7 exposure are LC50 data. Lethality is an unambiguous end-point which does not involve 8 gradations of seventy or incidence which are often difficult to quantify (e.g., lung congestion, 9 lung edema, irritation in the respiratory tract involving variations in both degree and area 10 affected). With respect to AEGL-2 values, it is far more difficult to quantify and achieve 11 consensus on gradations in non-lethal toxic effects with respect to severity and incidence in a 12 manner that readily results in a simple, quantitative toxic end-point concentration - exposure 13 duration relationship. Further, the LC50 is a statistically derived value in the mid-point of the 14 dose-response curve which is less subject to the vagaries in response at the extremes of the 15 exposure regimen. For these reasons, the NAC/AEGL Committee primarily has used LC50 data 16 in the derivation of exposure-time scaling relationships. These quantitative relationships have 17 then subsequently been used to derive both the AEGL-2 and the AEGL-3 values, and 18 occasionally the AEGL-1 values. This is believed to be a scientifically credible approach if the 19 mechanism of toxicity for AEGL-2 and AEGL-3 is known or thought to be similar. 20 21 It is recognized that the time scaling relationship observed with a lethality AEGL-3 22 endpoint may not accurately describe the irreversible effects or impairment of escape endpoint 23 used for the AEGL-2 endpoint. However, the NAC/AEGL Committee compares the AEGL-2 24 values against the supporting data to assess the reasonableness of the AEGL-2 determinations. 25 Based on this assessment, adjustments are made to better fit the supporting data If there are data 26 that suggest different lexicological mechanisms for lethal effects and AEGL-2 health effects, 27 selected values of n should be used for the development of the AEGL values. The upper and 28 lower bounds of n=3 and n=l should be used for extrapolation from longer to shorter and from 29 shorter to longer exposure periods, respectively. The resultant AEGL-2 values should then be 30 evaluated using all supporting data and adjusted or maintained accordingly. 31 32 A difficult application of time scaling is encountered when attempting to derive AEGL-1 33 values. The AEGL-1 value defines the air-borne concentration that distinguishes detection from 34 discomfort. As a result, the difficulty in attempting to quantify this often subjective level with 35 respect to severity and incidence in a manner sufficient to derive a concentration-exposure 36 duration relationship is greater than in the case of the AEGL-2. This is further complicated by 37 the nature of the biological end-point that one is attempting to quantify. For example, the 38 concentration level for odor detection in a group of individuals may actually decrease over time 39 due to olfactory fatigue. With respect to mild sensory effects, they generally are not cumulative 40 over a range of exposures of 10 minutes to 8 hours. Hence, the same AEGL-1 value may be 41 assigned to all AEGL-specified exposure periods. In certain instances, where experimental data 42 suggest that the sensory effects may increase due to the cumulative dose over time, the 10 Sop08-02 wpd Printed July 6. 2000 8 5 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 minute, 30 minute and 1 hour values may be constant, yet differ from a lower but constant 2 AEGL-1 value that is established for the 4 hour and 8 hour AEGL exposure durations. 3 4 In the case of certain sensory irritants, the AEGL values may be constant across all AEGL 5 time periods because this end-point is considered a threshold effect and prolonged exposure will 6 not result in an enhanced effect. In fact individuals may adapt to sensory irritation by these 7 chemicals over these exposure periods such that the warning properties are reduced. 8 9 10 2.7.8 Time Scaling - Guidelines for NAC/AEGL Committee Approach 11 12 This section is a compilation of time scaling guidelines which are used when deriving 13 AEGL values for different time penods. 14 15 16 2.7.8.1 Use of Empirical Data to Determine the Exposure Concentration- 17 Exposure Duration Relationship 18 19 THE RATIONALE FOR THE SELECTION OF AN EMPIRICALLY BASED TIME SCALING 20 APPROACH SHOULD INCLUDE: 21 22 1. The health effect used. 23 2. The exposure durations for which data were available. 24 3. Description of the statistical methodology used. If no methodology was used then 25 describe how the value of n was derived. 26 4. Description of the data used, including durations or the concentration/time values used 27 for extrapolation. Include the formula used. 28 5. Description of the different values of n that were used from one or more studies and 29 why a specific derived value of n was used. 30 6. The value of k calculated from C" x t = k after the uncertainty and modifying factors 31 have been applied to C. 32 7. If the value of n is based upon an analysis of the combined data from a number of 33 different studies then provide a description of how the different 34 time/concentration values were combined and why they were used. 35 36 37 2.7.8.2 Estimating the Concentration-Exposure Relationship using a 38 Surrogate Chemical 39 40 THE RATIONALE FOR THE SELECTION OF THIS TIME SCALING APPROACH SHOULD 41 INCLUDE: Sop08-02 wpd Printed July 6. 2000 86 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 1. Description of the structure/activity relationships between the two chemicals. 2 2. The health effect endpomt used. 3 3. The exposure durations for which data were available. 4 4. The statistical methodology used or a statement of how the value of n was derived 5 5. Description of the data from the surrogate chemical used to derive the concentration- 6 exposure duration relationship. If a derived value of n is used, the equation 7 should be included. 8 6. A description of how the different time/concentration values were combined and why 9 they were used if the value of n is based upon an analysis of the combined data 10 from a number of different studies. 11 7. The value of k calculated after uncertainty and modifying factors have been applied. 12 13 2.7.8.3 Estimating the Concentration-Exposure Duration Relationship when 14 Data are not Available to Derive a Value for n and Supporting Data are 15 Available. 16 17 Selection of values for n. In the absence of data to derive a value for n, a value for n of 1 18 is initially selected when extrapolating from shorter to longer exposure durations and a value for 19 n of 3 is initially selected when extrapolating from longer to shorter exposure durations. The 20 values of n are used with the equation C" x t = k to extrapolate from the empirically reported 21 exposure concentrations and exposure durations to the AEGL-specified exposure durations. 22 AEGL values in accord with the supporting data are then selected. 23 24 THE RATIONALE FOR THE SELECTION OF THE TIME SCALING APPROACH SHOULD 25 INCLUDE: 26 27 1. A presentation of the rationale in the TSD text as follows: The relationship between 28 concentration and duration of exposure as related to lethality was examined by ten 29 Berge et al. (1986) for approximately 20 irritant or systemically-acting vapors and 30 gases. The authors subjected the individual animal data sets to probit analysis 31 with exposure duration and exposure concentration as independent variables. An 32 exponential function (C'x t = k), where the value of n ranged from 0.8 to 3.5 for 33 different chemicals was found to be an accurate quantitative descriptor for the 34 chemicals evaluated. Approximately 90 percent of the values of n range between 35 n=l and n=3. Consequently, these values were selected as the reasonable lower 36 and upper bounds of n. A value of n=l is used initially when extrapolating from 37 shorter to longer time periods because the extrapolated values represent the most 38 conservative approach in the absence of other data. Conversely, a value of n=3 is 39 used when extrapolating from longer to shorter time periods because the 40 extrapolated values are more conservative in the absence of other data. If 41 supporting data are available (description and references for data should be 42 included) indicating that the AEGL value initially extrapolated is (too high/too Sop08-02 wpd Printed July 6, 2000 8 7 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 low), the AEGL value has been adjusted to reflect these (data, effects, etc.). 2 2. Presentation of the AEGL values or exposure concentrations extrapolated from data 3 using a value of n=l or n=3 and the adjustments made as a result of supporting 4 data.. 5 3. Discussion of the adjustment(s) made and the rationale for making them. 6 7 2.7.8.4 Determining Concentration-Exposure Relationships when Data are 8 not Available to Derive a Value for n and no Supporting Data are Available. 9 10 In the absence of data to derive a value(s) of n and the absence of supporting data to 11 validate a value of n, the value of n=l will be selected for extrapolating from shorter to longer 12 exposure durations, and the value n=3 will be selected for extrapolating from longer to shorter 13 expoosure durations. 14 15 THE RATIONALE FOR THE SELECTION OF THIS TIME SCALING APPROACH SHOULD 16 INCLUDE- 17 18 1. A presentation of the rationale in the TSD text as follows: The relationship between 19 concentration and duration of exposure as related to lethality was examined by ten 20 Berge et al. (1986) for approximately 20 irritant or systemically-actmg vapors and 21 gases. The authors subjected the individual animal data sets to probit analysis 22 with exposure duration and exposure concentration as independent variables. An 23 exponential function (C1 x t = k), where the value of n ranged from 0.8 to 3.5 for 24 different chemicals was found to be an accurate quantitative descriptor for the 25 chemicals evaluated. Approximately 90 percent of the values of n range between 26 n=l and n=3. Consequently, these values were selected as the reasonable lower 27 and upper bounds of n to use when data are not available to derive a value of n. A 28 value of n=l is used when extrapolating from shorter to longer time periods 29 because the extrapolated values are conservative and therefore, reasonable in the 30 absence of any data to the contrary. Conversely, a value of n=3 is used when 31 extrapolating from longer to shorter time penods because the extrapolated values 32 are conservative and therefore reasonable in the absence of any data to the 33 contrary. 34 35 2.7.8.5 AEGL Exposure Values are Constant Across Time. 36 37 THE RATIONALE FOR THE SELECTION OF THE TIME SCALING APPROACH SHOULD 38 INCLUDE. 39 40 1. The data and mode or mechanism of action of the chemical and its effect on humans 41 that supports the assignment of constant AEGL values across exposure durations. 42 Sop08-02 wpd Printed July 6. 2000 88 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2.8 GUIDELINES/CRITERIA FOR ADDRESSING SHORT TERM 2 EXPOSURE KNOWN AND SUSPECT CARCINOGENS 3 4 Cancer represents a serious adverse health effect. Historically, the concerns for 5 chemically-induced cancers were based on long-term, continuous exposure in controlled animal 6 studies or information derived from clinical or epidemiological studies of continuous or 7 long-term exposures in humans. To conduct quantitative risk assessments for cancer in humans, 8 mathematical (probit-log-dose) models were developed to utilize primarily animal bioassay data 9 and extrapolate from the higher experimental levels to assess the carcinogenic risk to humans at 10 low levels of chemical exposure. The evolution and usefulness of mathematical models to 11 accommodate new understanding or new concepts regarding the mechanisms of carcmogenesis 12 have been summarized in two publications by the National Research Council (NRC), National 13 Academy of Sciences (NAS): Developing Spacecraft Maximum Allowable Concentrations for 14 Space Station Contaminants (NRC, 1992a), and Guidelines for Developing Community 15 Emergency Exposure Levels for Hazardous Substances (NRC, 1993a). 16 17 In the United States, some state and federal regulatory agencies conduct quantitative risk 18 assessments on known or suspect carcinogens for continuous or long-term human exposure by 19 extrapolating downward in linear fashion from an upper confidence limit on theoretical excess 20 nsk (FDA, 1985; U. S. EPA, 1986). The values derived for a specified "acceptable" theoretical 21 excess risk to the U.S. human population, based on a lifetime of exposure to a carcinogenic 22 substance, have been used extensively for regulatory purposes. 23 24 There are no adopted state or federal regulatory methods for deriving such short-term 25 standards on the basis of carcinogenic risk because nearly all carcinogemcity studies in animals 26 and retrospective epidemiologic studies have entailed high-dose, long-term exposures. As a 27 result, there is uncertainty regarding the extrapolation from such studies in animals to the case of 28 brief human exposures. This is particularly problematical because the specific biological 29 mechanisms at the molecular, cell and tissue levels leading to cancer are often not known. It is 30 also possible that the mechanisms of injury that follow brief, high-dose exposures will often 31 differ from those following long-term exposures. To date U.S. federal regulatory agencies have 32 not established regulatory standards based on, or applicable to, less than lifetime exposures to 33 carcinogenic substances. 34 35 36 2.8.1 NRC/NAS Guidance 37 38 Guidance on the development of short-term exposure limits, published by the U. S. 39 National Research Council, National Academy of Sciences identified cancer as one of the 40 potential adverse health effects that may be associated with short-term inhalation exposures to 41 certain chemical substances (NRC, 1993a). This guidance document discusses and recommends 42 specific risk assessment methodologies for known genotoxic carcinogens and for carcinogens Sop08-02 wpd Printed July 6. 2000 89 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 whose mechanisms are not well understood. As a first approximation the general approach 2 involves linear low-dose extrapolation from an upper confidence limit on theoretical excess nsk. 3 Further, the NRC/NAS guidance states that the determination of short-term exposure limits will 4 require the translation of risks estimated from long-term, continuous exposures to risks 5 associated with short-term exposures. Conceptually, the approach recommended for genotoxic 6 carcinogens is the method developed by Crump and Howe (1984) for applying the multistage 7 model to assessing carcinogenic risks based on exposures of short duration. In the case of non- 8 genotoxic chemical carcinogens, the NRC/NAS guidance acknowledges that the approach is less 9 clear because of the many different modes of action and the complexities of non-genotoxic 10 carcinogenic mechanisms and the paucity of data on chemical-specific mode of action. It is 11 acknowledged also that dose thresholds may exist for certain non-genotoxic, carcinogens. The 12 NRC guidance suggests that, in lieu of linear, low-dose extrapolation, approaches involving non- 13 carcinogen nsk assessment techniques or the pure-promoter model from the class of 14 initiation-promotion-progression models be used, provided a known mechanism of action can 15 justify the specific approach. The guidance emphasizes the importance of knowing the 16 underlying biological processes when using any such models. 17 18 19 2.8.2 Precedents for Developing Short-Term Exposure Limits Based on 20 Carcinogenicity 21 22 The NRC/NAS guidance (1993a) for assessing the excess risks of genotoxic carcinogens 23 is based on an adaptation of the work of Crump and Howe (1984) by the NAS1 Committee on 24 Toxicology (COT). The COT's adaptation of the methodology was made for developing 25 Emergency Exposure Guidance Levels (EEGLs) and Short-Term Public Exposure Guidance 26 Levels (SPEGLs) for the Department of Defense (NRC, 1986). EEGLs represent exposure levels 27 intended to be acceptable for the performance of specific tasks by military personnel during 28 emergency conditions lasting 1 to 24 hours. The SPEGLs represent acceptable ceiling 29 concentrations for a single, unpredicted short-term exposure to the public. The exposure periods 30 range from 1 hour or less to 24 hours and the SPEGLs are generally set at 0.1 to 0.5 times the 31 corresponding EEGL value. 32 33 The criteria and methods document prepared by the COT for the development of EEGLs 34 and SPEGLs indicates that theoretical excess carcinogenic risk levels in the range of 10"" to 10"6 35 are acceptable nsk levels (NRC, 1986). However, the document states that "The role of 36 short-term exposures in producing cancer is not clear .... any exposure to a carcinogen has the 37 potential to add to the probability of carcinogenic effects .... (but).... the effects of long or 38 repeated exposures could greatly overshadow brief exposures (up to 24h)." Additionally, the 39 COT states "The assumption that the carcinogenic response is directly proportional to total dose 40 is likely not to hold for all materials and all tissues that these matenals affect." However, these 41 concerns not withstanding, the COT set SPEGL values based on the carcinogenic nsk assessment 42 methodology previously mentioned for hydrazine, methyl hydrazine, and 1,1 -dimethyl hydrazme. Sop08-02 wpd Printed July 6. 2000 90 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 In each case, the excess cancer nsk level used was 10"4 and the derived values were determined to 2 be lower than corresponding airborne concentration levels that were estimated to cause acute 3 toxicity. SPEGL values for exposure periods of less than 24 hours of other known or suspect 4 human carcinogens were not based on carcinogenicity. These chemicals included benzene, 5 tnchloroethylene, ethylene oxide, and lithium chromate. 6 7 The National Aeronautics and Space Administration (NASA) requested that the COT 8 develop spacecraft maximum allowable concentrations (SMACs) for space-station contaminants. 9 The COT published guidelines for the development of short-term and long-term SMACs (NRC, 10 1992a). Short-term SMACs refer to concentrations of airborne substances that will not 11 compromise the performance of specific tasks during emergency conditions lasting up to 24 12 hours. Because of NASA's concern for the health, safety, and functional abilities of space crews, 13 SMACs for exposure from 1 to 24 hours should not cause serious or permanent effects but may 14 cause reversible effects that do not impair judgement or interfere with proper responses to 15 emergencies. The long-term SMACs are designed to prevent deterioration in space crew 16 performance with continuous exposure for up to 180 days. 17 18 The guidelines for determining SMACs for carcinogens recommend the methods 19 proposed by Kodell, et. al., (1987) based on the linear multistage model. The level of excess nsk 20 used in the computation is 10"4. The guidelines suggest extrapolations of long-term (often 21 lifetime) exposures to shorter durations such as 1, 30, or 180 days and refer to a single-day 22 exposure as "the case of near instantaneous exposure." Further, the guidance states "It must be 23 remembered that extrapolation from a daily lifetime exposure level and conversion to an 24 instantaneous exposure level using.... (equations presented).... is an extreme case and is valid 25 only under the assumptions underlying the multistage theory of carcinogenesis." A review of the 26 first three volumes of published SMACs (35 chemicals) including ten (10) known or suspect 27 carcinogens, indicated that an assessment of excess risk for less than a 24 hour exposure period 28 was conducted on only one of the 10 carcinogenic substances. Carcinogenic assessments for 29 excess nsk were conducted on all 10 chemicals for 24 hours, as well as 7, 30, and 180 days. The 30 reasons provided in the COT technical support documents for not undertaking a risk assessment 31 on carcinogenic substances for exposure periods of less than 24 hours included: (1) "Data not 32 considered applicable to the exposure time (1 hr.)", (2) "Extrapolation to one hour exposure 33 duration produces unacceptable uncertainty in the values", and (3) "The COT model was not 34 used to calculate acceptable concentrations for exposures shorter than 24 hours" (NRC, 1992a). 35 36 As stated previously, to date no U.S. federal or state regulatory agency has promulgated 37 or established regulatory limits for single short-term (less than 24 hours) exposures based on 38 carcinogenic properties. 39 40 41 2.8.3 Scientific Basis for Credible Theoretical Excess Carcinogenic Risk 42 Assessments for Single Exposures of 8 Hours or Less Sop08-02 wpd Printed July 6, 2000 91 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2 The NRC/NAS guidance (NRC, 1993a) suggests that AEGLs can be developed using 3 carcinogenic risk assessment methodologies for exposure durations of 1 to 8 hours provided 4 adequate data are available. However, the guidance states that risk assessments on chemical 5 carcmogenicity in humans should be based on all relevant data and embody sound biological and 6 statistical principles. While some of the substances may be considered known human 7 carcinogens, most of the information is based on animal testing information. Additionally, since 8 the mode of action for animal carcinogens are not always the same with respect to biological 9 properties among animal species or strains and humans, a weight-of-evidence evaluation must be 10 carried out on a case-by-case basis. The weight of evidence evaluation considers comparative 11 metabolic disposition, pharmacokinetics parameters, routes of exposure, mechanisms of action, 12 and organ or species differences in response in animals and humans. 13 14 15 Uncertainties regarding lifetime theoretical excess carcinogenic risk assessments increase 16 as shorter durations of a single exposure are considered. Most of these concerns stem from the 17 reliance of both conclusions of carcmogenicity and quantitative assessments on long-term 18 exposures in humans in occupational settings or in test animals. Thus, calculations for 19 short-term risks require substantial extrapolation. At the same time, there are special concerns 20 and unresolved issues regarding short exposures that will require more relevant data before they 21 can be resolved. As evidenced from the actual application of these guidelines, the COT was 22 reluctant in most cases to develop quantitative carcinogenic risk assessments for less than 24 23 hours exposure in the development of SMACs. 24 25 To better understand the empirical data base for single exposures, the U.S. EPA funded a 26 study for the AEGL Program by Dr. Edward Calabrese of the University of Massachusetts to 27 review the published literature and assess the circumstances during which a single exposure of 28 short duration may cause cancer. This effort, referred to as the Single Exposure Carcinogen 29 Database, has been completed and represents a computerized summary that will enable the 30 evaluation of toxicological studies to assist in the NAC/AEGL Committee's assessment as to 31 whether a single exposure to a chemical under consideration for AEGL development could cause 32 tumor development. The data base will contain numerous parameters important to tumor 33 outcome and/or quality of the studies conducted. The database will contain approximately 5,500 34 "studies" or data sets involving approximately 500 chemicals from nearly 2000 references. 35 36 Although a brief overview of the Single Exposure Carcinogen Database has been 37 presented to the NAC/AEGL Committee, at the present time it is not known whether the data 38 available on single exposure of carcinogenic substances will be sufficient to justify their use in 39 the development of AEGL values. First, less than 20 of the 5,500 studies or data sets are based 40 on inhalation exposure. An initial review of the database indicates that only a limited number of 41 short-term cancer studies conducted by the inhalation route are available. Hence, route to route 42 extrapolations would need to be conducted in a manner that would not substantially weaken the Sop08-02 wpd Printed July 6. 2000 92 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 conclusions. This could be done for certain substances using standard U. S. EPA or U. S. NAS 2 procedures if the toxicant is likely to cause tumors at a site other than the port of entry. If the 3 substance causes tumors at the site of application or port of entry in oral or parenteral protocols, 4 extrapolation to the inhalation route of exposure becomes problematic. For this reason the 5 NAC/AEGL Committee in most cases will likely continue to rely on data from long-term human 6 and animal studies as the basis for the quantitative cancer risk assessments it conducts for short- 7 term exposures of 8 hours or less. 8 9 The Single Carcinogen Database may prove to be useful in obtaining some important 10 information for AEGL development. The database shows that single exposure to various 11 chemical classes, using various species and strains of animals, can result in tumor formation. 12 Furthermore, chemicals can be selected from the database for which there is dose-response 13 information. Data and information from positive responses of the chemical in the database could 14 be compared between the single dose study and the long-term study. 15 16 17 2.8.4 Practical Issues of Using Quantitative, Carcinogenic Risk Assessments 18 for Developing AEGLs 19 20 In addition to the important scientific issues regarding carcinogenic risk assessments in 21 the development of AEGL values, there are important practical issues to be considered by 22 emergency planners and responders regarding AEGL values that would be based on possible 23 carcinogenic effects. The acceptable cancer risk for a lifetime exposure to known or suspect 24 human carcinogens ranges from 10"4 to 10"6 for the U.S. EPA and most other U.S. federal 25 regulatory agencies (U. S. EPA, 1991). The AEGL values, however, are designed for emergency 26 planning, response, and prevention to accidental releases from chemical accidents. Thus, 27 theoretical excess cancer risk may be accumulated in 30 minutes or in a few hours. In addition to 28 the individual risk of 10"4 to 10"6 one should also consider a measure of population based nsk. 29 Experts in the chemical accident field indicate that the typical U.S. population at risk during most 30 accidental chemical releases is in the range of 1,000 to 5,000. The actual number of individuals 31 exposed depends on many factors, such as population density, quantity released, release rate, 32 prevailing wind direction and velocity, terrain and ambient temperature to name a few. 33 Therefore, a population-based risk range of 10"4 to 10"6, assuming a credible carcinogenic 34 assessment can be made, appears to be approaching zero for a population of 1,000 to 5,000 or 35 higher. The consideration of population-based risks by using assessment methods designed for 36 individual risks has precedent in U.S. EPA assessments of new industrial chemicals under TSCA 37 (Toxic Substance Control Act) Section 5 and pesticide chemicals under FIFRA (Federal 38 Insecticide Fungicide and Rodenticide Act). 39 40 Implementation of emergency response procedures based on theoretical excess risk values 41 of 10"4 to 10 •* values may be problematical. For example, if such values were used, they would 42 be based on an anticipated increased cancer risk of 10"* to 10"6, a level consistent with the EPA's Sop08-02.wpd Printed July 6. 2000 93 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 acceptable cancer nsk for lifetime exposures to known or suspect human carcinogens. However, 2 the nsks associated with evacuation and other response measures might possibly pose greater 3 risks of injury or perhaps death. Thus, setting AEGL values based on a cancer nsk may lead to 4 response measures that increase actual or total risk for the exposed population. 5 6 2.8.5 Current Approach of the NAC/AEGL Committee to Assessing Potential 7 Single Exposure Carcinogenic Risks 8 9 Based on the discussions and considerations presented in the earlier sections of m this 10 chapter on cancer risk assessment, the NAC/AEGL Committee has developed no AEGL values 11 based on carcinogenicity to date. In view of the great uncertainty of the assumptions used in 12 extrapolating from lifetime exposures to 8 hours or less, the paucity of single, inhalation 13 exposure data, the relatively small populations involved, and the potential risks associated with 14 evacuations and other response measures, the Committee does not believe their use in setting 15 AEGL values is justifiable at the present time. 16 17 However, the NAC/AEGL Committee will continue to identify and evaluate carcinogenic 18 data during the development of AEGLs on a chemical-by-chemical basis. The scientific 19 parameters which are used in this analysis are presented later in this section. In those cases 20 where, in the judgement of the Committee, it is appropriate, risk assessments for 10"4, 10"5, and 21 10"6 levels of cancer risk will be conducted. It is believed that information on known or suspect 22 human carcinogens should be provided to emergency planners and responders and made 23 available to the public at large even when such information is not used to set AEGL values. 24 Therefore, the Committee will continue to provide data and information on the carcinogenic 25 properties of chemicals in the Technical Support Documents, and in instances where the 26 appropriate data are available, develop quantitative cancer risk assessments at risk levels of 10"4, 27 10'5, and 10"6 in accordance with the NAS guidance (NAS, 1993a). The NAC/AEGL Committee 28 will attempt to limit potential cancer risk to 10"" or less where there is scientifically credible data 29 to support the risk based on a single exposure. If at some future date, substantial and convincing 30 scientific data become available that clearly establishes a relationship between a single, short- 31 term inhalation exposure to a chemical and the onset of tumors that are likely to occur in humans, 32 the carcinogenic risk in the development of the appropriate AEGL values will be considered. 33 34 2.8.5.1 Evaluation of Carcinogenicity Data 35 36 The evaluation of the carcinogenicity of a chemical in humans should be based on the 37 analysis of all relevant data, both positive and negative responses. Human epidemiologic and 38 clinical studies, as well as accidental exposure reports are considered and used to evaluate the 39 carcinogenic potential of a substance. In the absence of human data, long-term bioassay data 40 from controlled animal studies are used to derive theoretical excess carcinogenic nsk estimates 41 for exposed humans. The selection of data for estimating risk is based on the species and strain 42 considered most closely resembling the human response to provide the most accurate estimates. Sop08-02 wpd Printed July 6. 2000 94 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Data suggestive of a single exposure inducing a carcinogenic response, including related 2 mechanistic data that support such a possibility, will be considered. If highly convincing data 3 become available the Committee will consider the merits of the information in the development 4 of AEGL values. Weight should be given to those studies most relevant to estimating effects in 5 humans on a case-by-case basis. Data for assessing the strength of conclusions drawn from 6 controlled animal studies should include information on comparative metabolic pathways, 7 pharmacokinetics, routes of exposure, mechanisms of action, and organ or species differences in 8 response. In general, the NAC/AEGL Committee will follow a weight-of-evidence approach in 9 the evaluation of carcinogenicity that is consistent with the availability and biological variability 10 of the data and its relationship to the likelihood of effects in humans. 11 12 2.8.5.2 Methodology Used for Assessing the Carcinogenic Risk of a Single 13 Exposure 14 15 Guidance published in 1993a by the Committee on Toxicology, National Research 16 Council, National Academy of Sciences (NAS) states that the setting of AEGLs (CEELs) should 17 involve linear low-dose extrapolation from an upper confidence limit on excess risk for 18 genotoxic carcinogens and for carcinogens with mechanisms of action that are not well 19 understood. More specifically, the NAS guidance suggests an approach utilizing the methods 20 proposed by Kodell et al. (1987) based on multistage models. Although the NAS guidance states 21 that multistage models could be useful for setting AEGL values, the guidance acknowledges that 22 sufficient information may not be available to postulate the total number of stages in the cancer 23 process and the stage(s) that are dose-related. In these instances, the NAS guidance recommends 24 the use of the time-weighted-average dose where the instantaneous dose D at time t,, is assumed 25 to be the equivalent of the lifetime excess carcinogenic risk as daily dose D up to time t. This 26 equivalence is expressed by the equation D = d x t. As shown by Kodell et al. (1987), the actual 27 risk will not exceed the number of stages in the model (k). In instances where multistage models 28 can be used and prudence dictates conservatism, the NAS guidance suggests reducing the 29 approximation of D by an adjustment factor of 2 to 6, depending on the number of assumed 30 stages in the multistage model employed. 31 32 To date the NAC/AEGL Committee has evaluated excess theoretical risk at levels of 10"4, 33 10"5, and 10"* for a one-time exposure to known or suspect human carcinogens by determining the 34 total cumulative lifetime dose and applying Haber's law (concentration required to produce an 35 effect x time of exposure = constant) for exposure periods ranging from 8 hours to 30 minutes. 36 The resultant doses are then divided by an adjustment factor to account for the multistage nature 37 of carcinogens. See the example below. 38 39 40 2.8.5.2.1 The Determination of an Adjustment Factor Dealing with the Dose- 41 Dependent Stage of Carcinogenesis Sop08-02 wpd Printed July 6, 2000 95 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2 There is an extensive body of literature which deals with the concept of malignant tumor 3 development, progression of an initiated cell through successive stages and quantitative nsk 4 assessment. Two references, Crump and Howe 1984 and Kodell et al. 1987, are cited in the 5 NRC (1993a), publication. The concept has been further discussed in a number of publications 6 (Goddard et al, 1995; Murdoch et al., 1992; Murdoch and Krewski, 1988). This process is 7 referred to as a cell kinetic multistage model. There are several published variations of the basic 8 tenants in the model. If only one or more stages are dose-dependent and exposure is concentrated 9 m the dose-dependent stage, it is possible to underestimate nsk when the risk is based upon 10 lifetime exposure. For example, if the first stage is dose-dependent, and there is a single 11 exposure to an infant, the probability of cancer induction is maximized because the entire 12 lifetime of the individual is available for progression through the remaining stages in the 13 development of the cancer. If the same dose were given to an elderly person, the probability of 14 inducing cancer approaches 0 because there is insufficient time remaining in the life of that 15 individual for the initiated cell to progress through the subsequent stages to a malignant cancer. 16 Kodell et al. (1987) demonstrated that the underestimation of risk which is based upon a lifetime 17 of exposure will not exceed the number of stages in the multistage model. For this reason the 18 NRC (1986) recommends dividing the risk assessment based upon the lifetime exposure by a 19 factor between 2 and 6 to account for the number of stages in the multistage model applicable to 20 the particular chemical of concern. 21 22 In addition to the multistage model there have been a number of publications 23 investigating the two stage birth-death-mutation model (Morrison, 1987; Chen et al., 1988; 24 Murdoch and Krewski, 1988; Moolgavkar and Luebeck, 1990; Murdoch et al., 1992; Goddard et 25 al. 1995). This model is similar to the multistage model in which there are two stages. However, 26 the impact of the number of stem cells at the time of chemical exposure is considered as well as 27 the net growth rate of cells which have undergone the first stage of initiation. If the first stage 28 initiating event creates a cell which has a net growth rate greater than that of the stem cell, then 29 the risk of that initiating event will be greater than if the initiated cell grew at the same relative 30 rate as the stem cell. In this case, exposure early in life will cause a greater nsk than exposure 31 late in life. Conversely, exposure to a completer (effects only the second stage) late in life will 32 be more effective than early exposure because relatively more initiated cells are present. If this is 33 the only stage effected by the chemical, this situation is the same as 2 stages in the multistage 34 model. However, if the net growth of the initiated cells is 10 times the stem cell rate the relative 35 effectiveness of exposure late in life could be 10 fold (Murcoch and Krewski, 1988) Exposure 36 to promoters between the first and second stage event can have an impact by increasing the net 37 growth rate of initiated cells over that of stem cells. However, for maximum effectiveness the 38 exposure to promoters (generally considered to be non-genotoxic exposure) must encompass 39 multiple events (Chen et al., 1988; Murdoch and Krewski, 1988). Thus, the cancer risk 40 associated with a single exposure to a promoter should not be greater than predicted from 41 multiple exposures and no correction to the estimated risk need be made in this case. 42 Sop08-02 wpd Printed July 6. 2000 96 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 A major impact upon the risk assessment of the two stage model comes from carcinogen 2 exposure in the first stage in which the initiation event creates a cell with a greater net growth 3 rate than the stem cells. Modelers have considered a number of scenarios in which the net 4 growth rate of initiated cells varies from -10 to +10. The greatest increase in risk in the two stage 5 model comes about when the first stage is dose-dependent and the initiating event creates a cell 6 with a net growth rate of+10. In this case the increased risk is 10 fold (Murdoch and Krewski, 7 1988; Murdoch et al., 1992; Goddard et al., 1995) 8 9 Unfortunately, data on the biological plausibility of the maximum value for the net 10 growth rate of initiated cells is lacking (Murdoch et al, 1992). Major data needs for the two stage 11 birth-death-mutation model include the number of stem cells at different times of the life cycle, 12 how fast they divide and differentiate and how they respond to chemical exposure in terms of cell 13 division and mutation rate. This information is also needed for the initiated cell populations 14 (Moolgavkar and Luebeck, 1990). Because of this major uncertainty, the projections made for 15 the two stage model remain more speculative than for the multistage model in which there is 16 general agreement that the number of stages should not exceed 6. 17 18 For the above reasons, unless there is evidence to the contrary, the multistage model is 19 used when estimating risks for short-term exposures from lifetime exposure studies. In all of the 20 above referenced publications on the multistage model, the maximum number of stages modeled 21 is 6. 22 23 AEGL values are applicable to humans in all stages of life so the maximum risk to an 24 infant must be considered. In this case, the concentration based upon a lifetime exposure study is 25 divided by 6 unless there is evidence that the chemical is a later stage carcinogen or operates by a 26 mechanism different from the multistage model. The NAC/AEGL Committee will use the 27 divisor of 6 in agreement with the 1993a NAS guidance on the development of short-term 28 exposure limits which states that a factor of 6 represents a conservative adjustment factor for a 29 near-instantaneous exposure. 30 31 2.8.5.3 Summary of Cancer Assessment Methodology used by the 32 NAC/AEGL Committee 33 34 The U.S. EPA ql * values that are listed on the Integrated Risk Information System (IRIS) 35 or the GLOBAL86 generated slope factor values (Howe et al., 1986) are used to compute lifetime 36 risk levels. These values are based upon the guidance in U. S. EPA 1986. The U.S. EPA 37 (1996a) proposed methodology will be considered in the future. These values are used to 38 compute the concentration for a single exposure to the time penods of interest. As discussed in 39 the beginning of this section, these values are typically divided by 6 to account for early exposure 40 to a carcinogen in which the first stage is dose-dependent or late exposure to a carcinogen in 41 which the last stage is dose-dependent. If there is information about the number of stages 42 required for development of the cancer or the stage which is dose-dependent, the divisor will be Sop08-02 wpd Printed July 6, 2000 97 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 modified accordingly. An example of a Carcinogenicity Assessment is given in Appendix I. 2 3 The cancer evaluation includes a weight of evidence discussion which considers the 4 following factors: 5 6 • Less evidence of carcmogenicity from a short-term exposure 7 -No evidence for human carcmogenicity (may or may not lend 8 support of cancer induction from a single exposure but an 9 important consideration) 10 • Lifetime or long-term exposure necessary to elicit cancer 11 • Positive response only at very high doses 12 • Neoplasia appears reversible (when treatment is discontinued) 13 • Appears to be a "threshold" carcinogen 14 15 • Greater evidence of carcinogenicity from a short-term exposure 16 • Proven human carcinogen (may or may not lend support of cancer 17 induction from a single exposure but an important consideration) 18 • Short time-to-tumor 19 • Evidence for cancer from one to a few exposures 20 • Positive response at low doses 21 • Complete carcinogen 22 • Irreversible (when treatment is discontinued) 23 • Strongly mutagenic 24 25 Sop08-02 wpd Printed July 6. 2000 98 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30. 2000 1 2.9 GUIDELINES/CRITERIA FOR MISCELLANEOUS PROCEDURES 2 AND METHODS 3 4 2.9.1 Mathematical Rounding of AEGL Values 5 6 Given the uncertainties involved in generating AEGL values it could be argued that only 7 one significant figure should be used. However, because of a number of considerations 8 discussed below, numbers will be rounded to 2 significant figures. For example, 1.5, or 23, or 9 0.35. The value 7.35 would be rounded to 7.4. 10 11 Trivial differences in numbers can give large differences if only one significant figure is 12 used. For example, values of 14.9 and 15.1 would yield AEGL values of 10 and 20 respectively. 13 This is a two fold difference for a very small difference in computed AEGL values. Values of 14 18,14, 11, and 6 ppm for 30 minute, 1,4, and 8 hours would give values of 10, 10,10, and 20 15 ppm for the time points. It would not give the appearance of a logical progression. These 16 numbers will be used in exposure models to make decisions. The use of 2 significant figures will 17 allow for a more reasonable progression when different exposure scenarios are considered. 18 19 Two significant figures may seem overly precise when values less than 1 ppm are 20 presented since those levels may be difficult to measure. However, the AEGL-2 values will 21 often be used to compare with ambient air dispersion modeling projections for planning 22 purposes. In this case the use of 2 vs 1 significant figure could have an impact. Other rounding 23 off schemes may be used on a case by case basis with a justification. 24 25 26 2.9.2 Multiplication of Uncertainty Factors 27 28 When uncertainty factors are multiplied together the NAC/AEGL Committee often 29 multiplies two uncertainty factors of 3. Since the value 3 represents the geometric mean of 10 30 and 1, the actual number is 3.16. Therefore, the product of two different uncertainty factors is 31 not 3 times 3 but 3.16 times 3.16, which equals 10. For simplicities sake 3 times 10 is 32 represented by 30. 33 34 Sop08-02 wpd Printed July 6, 2000 99 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 i 3. FORMAT AND CONTENT OF TECHNICAL 2 SUPPORT DOCUMENTS 3 4 The Technical Support Document (TSD) is the compilation of all relevant data and 5 information from all key studies/references and the most important supporting studies/references 6 for both human exposures and laboratory animals. Additionally, this support document 7 addresses all methodologies employed in the derivation of the AEGL values in question and the 8 rationale and justifications for why certain data were used in the derivation and why certain 9 studies or data were not selected, why specific methodologies and adjustment factors were or 10 were not used, the scientific evidence supporting the rationale and justification, and the 11 appropnate references to the published scientific literature or sources of unpublished data and 12 information. 13 14 Major components to the TSD include 1) the Preface which includes definitions of the 15 AEGL tiers; 2) an Executive Summary which includes a concise summary of toxicity 16 information on the chemical, rationales used for time scaling and selection of uncertainty factors, 17 and a table of AEGL values for the three tiers as well as key references; 3) the main body of the 18 TSD which includes a detailed discussion of the items in listed in 2) and; 4) a Denvation 19 Summary Table which includes a list and discussion of the key data elements and rationale used 20 to derive the AEGL values. 21 22 EDITORIAL CONVENTIONS 23 24 • Concentrations will be expressed in the units used in the publication. If the data in the 25 publication or other data sources, were expressed in ppm, enter only ppm values. If data, 26 were expressed in mg/m3 or other units, then state the concentration as expressed in the 27 data source and add ppm in parentheses. 28 29 • References to footnotes should be superscript and lower case. 30 31 3.1 FORMAT AND CONTENT OF THE TECHNICAL SUPPORT 32 DOCUMENT (TSD) 33 34 PREFACE 35 36 The AEGL tiers are defined in the Preface of each TSD. See Chapter 2.1 for definitions 37 of AEGL-1, 2, and 3. 38 39 TABLE OF CONTENTS 40 Major headings in the text, tables and figures should be marked with the word processor Sop08-02 wpd Printed July 6, 2000 100 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 indexing tool so that the Table of Contents can be generated by the computer. A sample Table of 2 Contents is presented in Appendix E. 3 4 EXECUTIVE SUMMARY 5 6 The Executive Summary should include: 7 8 The name and CAS number of the chemical being reviewed. 9 10 A brief description of the substance, its physical properties, and uses. 11 12 A brief statement or overview of the toxicology, including the extent of the 13 data/information retrieved and reviewed, highlights of the most important research and strengths 14 and weaknesses of the database. Discuss data on human exposures and data on laboratory 15 animals. 16 17 A brief summary (1 paragraph for each AEGL tier) of the key study (with references), the 18 data used, and the denvation of the AEGL values. Each summary will include: 19 20 Information on the toxic endpoints and exposure levels used as the basis for 21 deriving the AEGL values. 22 Exposure level (If the data m the publication are expressed in ppm enter only ppm 23 values. If data were expressed in mg/m3 or other units then state the 24 concentration as expressed in the publication and add ppm in parentheses). 25 Exposure period. 26 Why this time-concentration point was selected (include effects observed or not 27 observed, relate to the AEGL level, etc.). 28 The species and number of animals used. 29 Consistency with human data if appropnate. 30 The reference to the key study. 31 A statement of uncertainty factors and modifying factors used or not used and why 32 a specific value was chosen. 33 A statement of the time scaling method used and why it was selected (include the 34 rationale for the value of n in the time scaling equation). 35 36 A brief statement regarding carcinogenicity, if appropnate. 37 38 A brief statement on the adequacy of the data (see Section 2.3.3 of this SOP Manual). 39 40 A summary table of draft/proposed AEGL values with: 41 Values presented in ppm with mg/m in parentheses. 42 A rationale and reference for AEGL-1, -2, and -3. Sop08-02 wpd Printed July 6. 2000 101 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Reasons for no AEGL value. 2 3 References 4 5 A sample Executive Summary is presented in Appendix F. Sop08-02 wpd Printed July 6, 2000 102 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 OUTLINE OF THE MAIN BODY OF THE TECHNICAL SUPPORT 2 DOCUMENT 3 4 1. INTRODUCTION 5 • General information regarding occurrence, production/use, physical/chemical data 6 (table for physical chemical data) 7 8 2. HUMAN TOXICITY DATA 9 2.1 Acute Lethality - include anecdotal case reports if pertinent 10 2.2 Nonlethal Toxicity 11 2.2.1 Acute Studies - include anecdotal case reports if pertinent 12 2.2.2 Epidemiologic Studies 13 2.3 Developmental/Reproductive Toxicity 14 2.4 Genotoxicity 15 2.5 Carcinogemcity - include EPA and IARC classifications 16 2.6 Summary - weight-of-evidence approach 17 18 As appropriate, data are tabulated within sections and/or in summary 19 20 3. ANIMAL TOXICITY DATA 21 3.1 Acute Lethality - include species/strain, number of animals, exposure 22 concentrations/durations, mortality rates/ratios, time to death. (The order of 23 animals shown should be used. If no data are available for a species, the number 24 should be used for the next species discussed.) 25 3.1.1 Nonhuman Primates 26 3.1.2 Dogs 27 3.1.3 Rats 28 3.1.4 Mice 29 3.1.5 Guinea Pigs 30 3.1.6 Rabbits 31 3.1.7 Other Species 32 33 • Sections to include relevant studies (potential key studies and supporting data) or 34 provide overall picture of toxicity data as appropriate 35 • Third-level headers to vary dependent upon available data; exclusion of header to 36 imply no data 37 38 3.2 Nonlethal Toxicity - include species/strain, no. of animals, exposure 39 concentrations/durations, critical effects, time course data, etc. (The order of 40 animals shown should be used. If no data are available for a species, the number 41 should be used for the next species discussed.) 42 3.2.1 Nonhuman Primates Sop08-02 wpd Printed July 6. 2000 103 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 3.2.2 Dogs 2 3.2.3 Rats 3 3.2.4 Mice 4 3.2.5 Guinea Pigs 5 3.2.6 Rabbits 6 3.2.7 Other Species 7 8 • Sections to include relevant studies (potential key studies and supporting data) or 9 provide overall picture of toxicity data as appropriate 10 • Third-level headers to vary dependent upon available data; exclusion of header to 11 imply no data 12 13 3.3 Developmental/Reproductive Toxicity 14 3.4 Genotoxicity 15 3.5 Carcinogenicity 16 3.6 Summary - weight-of-evidence approach 17 18 Tabulation of data as appropriate within sections and/or in summary 19 20 4. SPECIAL CONSIDERATIONS 21 4.1 Metabolism and Disposition - general background; interspecies and 22 individual variabilities especially as they pertain to AEGL derivation 23 4.2 Mechanism of Toxicity - general background; interspecies and individual 24 variabilities especially as they pertain to AEGL derivation 25 4.3 Structure-Activity Relationships - data relevant to filling data gaps on the 26 chemical 27 4.4 Other Relevant Information 28 4.4.1 Species Variability 29 4.4.2 Concurrent Exposure Issues (potentiation, etc) 30 31 • Third-level headers to vary dependent upon available data; exclusion of header 32 implies no data 33 34 35 5. DATA ANALYSIS FOR PROPOSED AEGL-1 36 5.1 Summary of Human Data Relevant to AEGL-1 - general summary 37 description of selected key and supporting study(ies) if available 38 5.2 Summary of Animal Data Relevant to AEGL-1 - general summary 39 description of selected key and supporting study(ies) if available 40 5.3 Derivation of AEGL-1 - key study, critical effect, dose/exposure, uncertainty 41 factor application/justification, temporal extrapolation, assumptions, confidence, 42 consistency with human data if appropriate Sop08-02 wpd Printed July 6, 2000 104 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 6. DATA ANALYSIS FOR PROPOSED AEGL-2 2 6.1 Summary of Human Data Relevant to AEGL-2 - general summary 3 description of selected key and supporting study(ies) if available 4 6.2 Summary of Animal Data Relevant to AEGL-2-general summary description of 5 selected key and supporting study(ies) if available 6 6.3 Derivation of AEGL-2 - key study, critical effect, dose/exposure, uncertainty 7 factor application/justification, temporal extrapolation, assumptions, confidence, 8 consistency with human data if appropriate 9 10 7. DATA ANALYSIS FOR PROPOSED AEGL-3 11 7.1 Summary of Human Data Relevant to AEGL-3 - general summary 12 description of selected key and supporting study(ies) if available 13 7.2 Summary of Animal Data Relevant to AEGL-3 - general summary 14 description of selected key and supporting study(ies) if available 15 7.3 Derivation of AEGL-3 - key study, critical effect, dose/exposure, 16 uncertainty factor application/justification, temporal extrapolation, 17 assumptions, confidence, consistency with human data if appropriate 18 19 8. SUMMARY OF PROPOSED AEGLS 20 8.1 AEGL Values and Toxicity Endpoints 21 8.2 Comparison with Other Standards and Criteria (summarized in text and presented 22 in a table - see SOP Appendix K for an example) 23 8.3 Data Adequacy and Research Needs (for content see Section 2.3.3 of this 24 SOP Manual) 25 26 9. REFERENCES CITED 27 28 10. APPENDICES 29 30 APPENDDC A (Derivation of AEGL Values) See SOP Appendix G for an example 31 APPENDDC B (Time Scaling Calculations) See SOP Appendix H for an example 32 APPENDDC C (Carcinogenicity Assessment) See SOP Appendix I for an example 33 APPENDDC D (Derivation Summary) See SOP Appendix J for specific format and an example Sop08-02 wpd Printed July 6. 2000 105 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 APPENDIX D: Format for Derivation Summary DERIVATION SUMMARY (CAS NUMBER; CHEMICAL NAME) 10 minutes ppm AEGL-1(OR 2 OR 3) VALUES 30 minutes ppm 1 hour ppm 4 hours ppm 8 hours ppm Reference: Test Species/Strain/Number: Exposure Route/Concentrations/Durations: Effects: Endpoint/Concentration/Rationale: Uncertainty Factors/Rationale: Modifying Factor: Animal to Human Dosimetric Adjustment: Time Scaling: Data Adequacy3: a Elements that should be included in the Data Adequacy Section are discussed in Section 2.3.3 of this SOP Manual. If an AEGL-1 value is not recommended, there should be a short discussion of the rationale for that choice. The rationale should include as appropriate a discussion that numeric values for AEGL-1 are not recommended because (1) relevant data are lacking, (2) the margin of safety between the derived AEGL-1 and AEGL-2 values is inadequate, or (3) the derived AEGL-1 is greater than the AEGL-2. Absence of an AEGL-1 does not imply that exposure below the AEGL-2 is without adverse effects. Sop08-02 wpd Printed July 6. 2000 106 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 3.2 Potential Inclusion of Graphic Descriptions of Data 2 3 Graphic descriptions of important and relevant data can be helpful in identifying, 4 understanding and comparing data in terms of similarities and differences, degree of variation, 5 and trends among the values cited. Well prepared graphs provide the reader with a rapid 6 overview of dose-response relationships in terms of both airborne concentrations and exposure 7 periods among vanous studies and various species. The graphs should supplement the data 8 tables but not replace them. They can be placed in the body of the document or in an appendix. 9 Below are examples of presentations of graphic data. 10 11 It is very difficult to keep different times and the toxicity values for those times in one's 12 head when reading the Technical Support Document. Comparisons are difficult to make between 13 times because the values vary according to the time. The old adage "A picture is worth a 14 thousand words" is especially appropriate when analyzing inhalation data. A particularly useful 15 way to present the data is presented in Table 3.2-1 and Figure 3.2-1. It is based upon the concept 16 of placing the toxic response into severity categories (Hertzberg and Miller, 1985; Hertzberg and 17 Wymer, 1991; and Guth et al., 1991). In Table 3.2-1 the seventy categories are chosen to fit into 18 definitions of the AEGL level health effects. In the table the category severity definitions for the 19 column headings are 0 = No effect; 1 = Discomfort; 2 = Disabling; 3 = Lethal; NL = Did not die 20 at a lethal cone (at an experimental concentration in which some of the animals died and some 21 did not, the NL label refers to the animals which did not die); AEGL or C = AEGL or censored 22 (severity category could not be established). The effects which will place an experimental result 23 into a particular category will vary according to the spectrum of data available on a specific 24 chemical and the effects from exposure to that chemical. When the exposure concentration is 25 placed into the appropriate column, the graph in Figure 3.2-1 is generated. The doses often span 26 a number of orders of magnitude, especially when human data exist. Therefore the concentration 27 is placed on a log scale. Note that the AEGL values are designated as a triangle without an 28 indication to their level. The AEGL-3 is higher than the AEGL-2, which is higher than the 29 AEGL-1. 30 31 This type of plot is useful for a number of reasons and can be used to address the 32 following questions: 33 34 • Are the AEGL levels protective? 35 36 • Are the AEGL-3 levels below the concentration causing death in experimental 37 animals? If the answer is no then the question should be raised about the 38 appropriateness of the AEGL-3 value. Is the AEGL-3 level appropriate and the data 39 point anomalous, or should the AEGL-3 value be lowered? 40 41 • Similar questions should be asked about the AEGL-1 and AEGL-2 values. 42 Sop08-02 wpd Printed July 6, 2000 107 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 • Are there data points which appear to be outliers? Why are they outliers? Should they be 2 considered in the development of AEGL values or discarded because of faulty 3 experimental technique. 4 5 • Does the spread of data points for a particular seventy category indicate major differences 6 between species or are the results from different species congruent. 7 8 • Is the time scaling algorithm reasonable consistent with the data? For example, does the 9 plot of the AEGL-3 values using the derived or chosen value of n in the equation Cn x t = 10 k parallel the slope of the lethality data. Similar questions can be asked about the AEGL- 11 1 and AEGL-2 plots. 12 13 • Is there evidence that a different time scaling factor should be used for the AEGL-2? 14 15 • What are the most appropnate data points to use for the time scaling? 16 17 Sop08-02 wpd Printed July 6. 2000 108 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 TABLE 3.2-1 GROUPING DATA INTO CATEGORIES FOR PLOTTING Reference chemical NAC/AEGL-i NAC/AEGL-1 NAC/AEGL-1 NAC/AEGL-1 NAC/AEGL-2 NAC/AEGL-2 NAC/AEGL-2 NAC/AEGl-2 NAC/AEGL-3 NAC/AEGL-3 NAC/AEGL-3 NAC/AEGL-3 Baelurn et a).. 1965 : Wilson 1943 Wilson 1943 Wilson 1943 Ukalelal., 1993 Lee at al.. 1988 Gamberale and Huttengren, 1972 Gamberale and Hultengren, 1972 von Oettingen etal., 1942 von Oettingen el at., 1942 yon Oettlngen el al., 1942 . Baetum et al.. 1990 : Echeverriaelel.. 1991 Andersen et al., 1963 RahiH et at.. 1996 Dfcketal., 1984 Cherry el al., 1983 : Carpenter etal., 1976 Piyoretal.. 1978 Pryor etal., 1978 Cameron et al.. 1938 Cameron el al.. 1938 Kojima and Kobayashl. 1973 Kpjirna and Kobayashl, 1973 Cameron el al.. 1938 Carpenter el al.. 1976 Carpenter el al., 1976 Smyth etal., 1969 ' Smyth et a!.. 1969 Bonnet et at.. 1979 Bonnet et el.. 1979 Svirbely etal.. 1943 Svkbely etal., 1943 Moser and Batster, 1985 Moser and Balster, 1985 Moser and Balster. 1985 Moser and Batster. 1985 . Moser and Balster. 1985 , Moser and Balster, 1985 Exp : Grp Species Sex hu ,hu 4rat mouse ' mouse \ mouse mouse mouse 1 mouse , mouse mouse ; mouse 897 634 317 224 100 200 200 500 500 100 100 300 700 200 200 600 800 100 150 40 100 100 100 80 220 26700 26700 24400 24400 15000 15000 12200 8800 8800 4000 4000 6940 6940 5320 5320 38465 38465 21872 21872 19018 19018 GpSize sensory Imtatton, steeplness, Intoxication, manual dexterity, color discrimin headache, lassitude, anorexia ' headache, nausea, (ncoordination, reaction lime headache, nausea, Incoodination, reaction time headache, nausea. incoordinaUon. reaction time and palpitation, extreme wi weight toss, dizziness, headache, tightness In chest dimmed vision weight toss, dizziness, headache. reaction time : perceptual speed ! muscular weakness, confusion. Impaired coordination, end dilated pupils severe incoodination, confusion, dilated pupils, nausea, and extreme fatigue severe (ncoordination, confusion, dilated pupils, nausea, and extrerne fatigue loss of self-control, rnuscuiar weakness, extreme fatigue, nausea, and bone i sensory irritation, altered temp, perception, headache, dizziness, and score: performance on spatial and neurobehavtoral tasks, headache, eye Irritation. no effect/sensory Irritation, odor [ no effect/sensory Irritation, odor : latency on a neurobehavioral task [not a btotoglcalr/ relevant neufobehavioi accuracy on visual-vigilance test (not a biologically no impairment on neurobehavioral tasks , sensory threshold LC50 ILC50 60% mortality , 6O% mortality 80% mortality 80% mortality 100% mortality iLCSO UC50 , 16% mortality 16% mortality : LCSO iLCSO LCSO jLCSO LCSO LCSO ; LCSO LC50 ILC50 ,LC50 Sop08-02.wpd Printed July 6, 2000 109 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 Q. Q. 100.0 - 10.0 Chemical Toxicity - TSD All Data Toluene 60 120 180 240 Minutes 300 360 420 480 o no or minimal effect discomfort disabling did not die @ lethal cone AEGL or censored 1 FIGURE 3.2-1 PLOT OF CATEGORIES OF DATA Sop08-02.wpd Printed July 6, 2000 110 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 i 4. CURRENT ADMINISTRATIVE PROCESSES AND 2 PROCEDURES FOR THE DEVELOPMENT OF AEGL 3 VALUES 4 5 The primary purpose of the AEGL Program and the NAC/AEGL Committee is to develop 6 guideline levels for short-term exposures to airborne concentrations of acutely toxic, high priority 7 chemicals. These Acute Exposure Guideline Levels (AEGLs) are needed for a wide range of 8 planning, response, and prevention applications. These applications may include many U. S. 9 initiatives such as the EPA's SARA Title ni Section 302-304 emergency planning program, the 10 CAAA Section 112(r) accident prevention program, and the remediation of Superfund sites 11 program; the DOE environmental restoration, waste management, waste transport, and fixed 12 facility programs; the DOT emergency waste response program; the DOD environmental 13 restoration, waste management, and fixed facility programs; ATSDR health consultation and nsk 14 assessment programs; NIOSH/OSHA regulations and guidelines for workplace exposure; State 15 CAA Section 112(b) programs and other state programs; the U. S. Chemical Manufacturer's 16 Association (CMA) Chemtrec program; and other chemical emergency programs in the U. S. 17 private sector. From an international perspective, it is anticipated that the AEGLs will find a 18 wide range of applications in chemical emergency planning, response, and prevention programs 19 in both the public and private sectors of member-countries of the Organization for Economic and 20 Cooperation Development (OECD). It is hoped that the AEGLs also will be used by other 21 countries in the international community 22 23 A principal objective of the NAC/AEGL Committee is to develop the most scientifically 24 credible, acute (short-term) exposure guideline levels possible within the constraints of data 25 availability, resources and time. This includes highly effective and efficient efforts in data 26 gathering, data evaluation and data summarization, fostering the participation of a large cross- 27 section of the relevant scientific community, both nationally and internationally, and the adoption 28 of procedures and methods that facilitate consensus-building for AEGL values within the 29 NAC/AEGL Committee. 30 31 Another principal objective of the NAC/AEGL Committee is to develop AEGL values for 32 approximately 400 to 500 acutely hazardous substances within the next ten (10) years. 33 Therefore, the near-term objective is to increase the level of production of AEGL development to 34 approximately forty (40) to fifty (50) chemicals per year without exceeding budgetary limitations 35 or compromising the scientific credibility of the values developed. 36 37 To reach these objectives, the NAC/AEGL Committee must adopt and adhere to specific 38 processes and procedures both scientifically and administratively. This is accomplished through 39 the development and maintenance of a comprehensive "Standing Operating Procedures" Manual 40 (SOP Manual) that addresses both the scientific and administrative procedures required to 41 achieve the objectives of the NAC/AEGL Committee previously mentioned. This section is Sop08-02 wpd Printed July 6, 2000 111 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 devoted to those administrative processes and procedures deemed necessary to achieve the 2 AEGL Program objectives. 3 4 4.1 COMMITTEE MEMBERSHIP AND ORGANIZATIONAL 5 STRUCTURE 6 7 The NAC/AEGL Committee is comprised of representatives of U. S. federal, state and 8 local agencies, and organizations in the pnvate sector that derive programmatic or operational 9 benefits from the AEGL values. This includes federal representatives from the Environmental 10 Protection Agency (EPA), the Department of Energy (DOE), the Agency for Toxic Substances 11 and Disease Registry (ATSDR), the National Institute for Occupational Safety and Health 12 (NIOSH), Occupational Safety and Health Administration (OSHA), the Department of 13 Transportation (DOT), the Department of Defense (DOD), the Center for Disease Control 14 (CDC), the Food and Drug Administration (FDA), and the Federal Emergency Management 15 Agency (FEMA). States providing committee representatives include New York, New Jersey, 16 Texas, California, Minnesota, Illinois, Connecticut, and Vermont. Private companies with 17 representatives include Allied Signal Corporation, Exxon Corporation, and Olm Chemical 18 Company. Other organizations with representatives include the American Industrial Hygiene 19 Association (AIHA), American College of Occupational and Environmental Medicine 20 (ACOEM), American Association of Poison Control Centers (AAPCC), and the American 21 Federation of Labor - Congress of Industrial Organizations (AFL-CIO). In addition, the 22 committee membership includes individuals from academia, a representative of environmental 23 justice, and other organizations in the private sector. A current list of the NAC/AEGL 24 Committee members and their affiliations is shown in Appendix A of this SOP manual. At 25 present, the Committee is comprised of 32 members. 26 27 Recently, the Organization of Economic and Cooperation Development (OECD) and 28 various OECD member countries have expressed an interest in the AEGL Program. Several 29 OECD member countries such as Germany and the Netherlands have been participating in the 30 Committee's activities and actively pursuing formal membership on the NAC/AEGL Committee. 31 It is envisioned that the Committee and the AEGL Program in general will progressively expand 32 its scope and participation to include the international community. 33 34 The Director of the AEGL Program has the overall responsibility for the entire AEGL 35 Program and the NAC/AEGL Committee and its activities. A Designated Federal Officer (DFO) 36 is responsible for all administrative matters related to the Committee to insure that it functions 37 properly and efficiently. These individuals are not voting members of the Committee. The 38 NAC/AEGL Committee Chair is appointed by EPA and is selected from among the committee 39 members. In concert with the Program Director and the DFO, the Chair coordinates the activities 40 of the Committee and also directs all formal meetings of the Committee. From time to time, the 41 members of the Committee serve as Chemical Managers and Chemical Reviewers in a 42 collaborative effort with assigned scientist-authors (non-Committee members) to develop AEGLs Sop08-02 wpd Printed July 6, 2000 112 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 for a specific chemical. These groups of individuals are referred to as the AEGL Development 2 Teams and their function is discussed in Section 4.8 of this manual.. 3 4 4.2 THE AEGL DEVELOPMENT AND PEER REVIEW PROCESS 5 6 The process that has been established for the development of the AEGL values is the 7 most comprehensive ever employed for the determination of short-term exposure limits for 8 acutely toxic chemicals. A summary of the overall process is presented in diagram form in 9 Figure 4.2-1. The process consists of four basic stages in the development and status of the 10 AEGLs and they are identified according to the review level and concurrent status of the AEGL 11 values. They include (1) "Draft" AEGLs, (2) "Proposed" AEGLs, (3) "Interim" AEGLs and (4) 12 "Final" AEGLs. The entire development process can be descnbed by individually describing the 13 four basic stages in the development of AEGL values. 14 15 16 Stage 1: "Draft" AEGLs 17 18 This first stage begins with a comprehensive search of the published scientific literature. 19 Attempts are made to mobilize all relevant, non-published data through industry trade 20 associations and from individual companies in the private sector. A more detailed description of 21 the published and unpublished sources of data and information utilized is provided in Section 2.3 22 of this document which addresses search strategies. The data are evaluated following the 23 guidelines published in the NRC/NAS guidance document and this SOP manual and selected 24 data are used as the basis for the derivation of the AEGL values and the supporting scientific 25 rationale. Data evaluation, data selection, and the development of a technical support document 26 are all performed as a collaborative effort among the Staff Scientist at the organization which 27 drafts Technical Support Documents, the Chemical Manager, and two Chemical Reviewers. 28 This group is referred to as an "AEGL Development Team". NAC/AEGL Committee members 29 are specifically assigned this responsibility for each chemical under review. Hence, a separate 30 team comprised of different Committee members is formed for each chemical under review. The 31 product of this effort is a technical support document (TSD) that contains "Draft" AEGLs. The 32 Draft TSD is subsequently circulated to all other NAC/AEGL Committee members for review 33 and comment prior to a formal meeting of the Committee. Revisions to the initial TSD and the 34 "Draft" AEGLs are made up to the time of the NAC/AEGL Committee meeting scheduled for 35 formal presentation and discussion of the AEGL values and the documents. Following 36 deliberations during the committee meeting, an attempt is made to reach consensus, or the 37 minimum of a two-thirds majority of a quorum present, to elevate the AEGLs to "Proposed" 38 status. If agreement cannot be reached, the Committee conveys its issues and concerns to the 39 AEGL Development Team and further work is conducted by this group. After completion of 40 additional work, the chemical is resubmitted for consideration at a future meeting. If a consensus 41 or two-thirds majority vote of the Committee cannot be achieved because of inadequate data 42 unrelated to the completeness of the data search, the chemical becomes a candidate for Sop08-02 wpd Printed July 6.2000 113 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30,2000 1 appropriate toxicity studies. 2 3 4 Stage 2: "Proposed" AEGLs 5 6 Once the NAC/AEGL Committee has reached a consensus, or the minimum two-thirds 7 majority vote, on the AEGL values and supporting rationale, they are referred to as "Proposed" 8 AEGLs and are published in the Federal Register for a thirty (30) day review and comment 9 penod. Following publication of the "Proposed" AEGLs in the Federal Register, the Committee 10 reviews the public comments, addresses and resolves relevant issues and seeks a consensus or 11 minimum two-thirds majority of those present on the Committee on the original or modified 12 AEGL values and the accompanying scientific rationale. 13 14 15 Stage 3: "Interim" AEGLs 16 17 Following resolution of relevant issues raised through public review and comment and 18 subsequent approval of the Committee, the AEGL values are classified as "Interim". The 19 "Interim" AEGL status represents the best efforts of the NAC/AEGL Committee to establish 20 exposure limits and the values are available for use as deemed appropnate on an interim basis by 21 federal and state regulatory agencies and the private sector. The "Interim" AEGLs, the supporting 22 scientific rationale, and the TSD are subsequently presented to the U. S. National Academy of 23 Sciences (NAS/AEGL Subcommittee) for review and concurrence. If concurrence cannot be 24 achieved, the NAS/AEGL Subcommittee will submit its issues and concerns to the NAC/AEGL 25 Committee for further work and resolution. 26 27 28 Stage 4: "Final" AEGLs 29 30 When concurrence by the NAS/AEGL Subcommittee is achieved, the AEGL values are 31 considered "Final" and published by the U. S. NAS. Final AEGLs may be used on a permanent 32 basis by all federal, state and local agencies and private sector organizations. It is possible that 33 from time to time new data will become available that challenges the scientific credibility of 34 "Final" AEGLs. If this occurs, the chemical will be resubmitted to the NAC/AEGL Committee 35 and recycled through the review process. 36 Sop08-02 wpd Printed July 6. 2000 114 ------- 1 2 Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 FIGURE 4.2-1 THE AEGL DEVELOPMENT PROCESS AEGL Development Process Non-published. Non-peer Reviewed Industry Data Published Literature Search Other Data / Information Sources Special Toxiaty Studies AEGL Dei Team- Saentist. Manager. Revie /elopment ORNL Chemical Chemical iwers Technical Support Documents (TSDs) Distribute Draft or Proposed TSDs / AEGLs to Committee Members 4- NAC/AEGL Committee Meeting to Discuss Draft or Proposed AEGLs -NO- NAC/AEGL Committee Consensus on Proposed AEGLs Major Changes FR Publication I NAS-NRC AEGL Subcommittee NAS-NRC Publication of Final AEGLs Sop08-02 wpd Printed July 6. 2000 115 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 4.3 OPERATION OF THE COMMITTEE 2 3 The NAC/AEGL Committee meets formally four (4) times each year for two and one-half 4 (2-1/2) days. The meetings are scheduled for each quarter of the calendar year and are generally 5 held in the months of March, June, September, and December. Based on overall cost 6 considerations, the meetings are generally held in Washington, D.C. However, from time to 7 time, committee meetings may be held at other locations for justifiable reasons. 8 9 At least 15 days prior to the committee meetings, a notice of the meeting is published in 10 the Federal Register together with a list of chemicals and other matters to be addressed by the 11 Committee and provides dates, times and location of the meetings. The agenda is finalized and 12 distributed to committee members approximately one week prior to the meeting. The agenda 13 also is available to other interested parties at that time, upon request, through the Designated 14 Federal Officer (DFO). 15 16 All NAC/AEGL Committee meetings are open to the public and interested parties may 17 schedule individual presentations of relevant data and information by contacting the DFO to 18 establish a date and time. Relevant data and information from interested parties also may be 19 provided to the Committee through the DFO during the period of development of the Draft 20 AEGLs so that it can be considered during the early stage of development. Data and information 21 also may be submitted during the Proposed and Interim stages of AEGL development as well. 22 23 The NAC/AEGL Committee meetings are conducted by the Chair who is appointed by 24 the U.S. Environmental Protection Agency in accordance with the Federal Advisory Committee 25 Act (FACA). At the time of the meeting, both the Chair and all other committee members will 26 have received the initial draft and one or more revisions of the Technical Support Document 27 (TSD) and "Draft", "Proposed", or "Interim" AEGL values for each chemical on the agenda. 28 Reviews, comments, and revisions are continuous up to the time of the meeting and committee 29 members are expected to be familiar with the "Draft", "Proposed", or "Interim" AEGLs, 30 supporting rationale, and other data and information in each TSD and to participate in the 31 resolution of residual issues at the meeting. Procedures for the AEGL Development Teams and 32 the other Committee members regarding work on AEGLs in the Proposed or Interim status are 3 3 similar to those for Draft AEGLs. 34 35 All decisions of the NAC/AEGL Committee related to the development of Draft, 36 Proposed, Interim, and Final AEGLs and their supporting rationale are made by consensus or a 37 minimum of two-thirds (2/3) majority of a quorum of committee members. A quorum of the 38 NAC/AEGL Committee is defined as fifty-one percent (51%) or more of the total NAC/AEGL 39 Committee membership m attendance. 40 41 The highlights of each meeting are recorded by the scientists who draft the Technical 42 Support Documents and written minutes are prepared, ratified and maintained in the Sop08-02 wpd Printed J uly 6, 2000 116 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Committee's permanent records. Deliberations of each meeting also are tape-recorded and stored 2 in the Committee's permanent records by the Designated Federal Officer (DFO) for future 3 reference as necessary. 4 5 All Proposed AEGL values and supporting scientific rationale are published in the 6 Federal Register. Review and comment by interested parties and the general public are requested 7 and encouraged. The Committee's response to official comments on Federal Register notices on 8 Proposed AEGL values consists of the discussions and deliberations that take place during the 9 Committee meetings for elevating the AEGLs from "Proposed" to "Interim" status. This 10 information is reflected on the tapes and in the minutes of the meetings and will be maintained 11 for future reference. Changes in the Proposed AEGL values and the supporting rationale that are 12 considered appropriate by the NAC/AEGL Committee based on Federal Register Comments will 13 be made prior to elevating the AEGLs to Interim status. 14 15 As previously mentioned a "Standing Operating Procedures" Workgroup (SOP 16 Workgroup) was established in March, 1997 to document, summarize, and evaluate the various 17 procedures, methodologies, and guidelines employed by the Committee in the gathering and 18 evaluation of scientific data and information and the development of the AEGL values. The SOP 19 Workgroup performs a critical function by continually providing the Committee with detailed 20 information on the Committee's interpretation of the NAS guidelines and the approaches the 21 Committee has taken in the derivation of each AEGL value for each chemical addressed. This 22 documentation enables the Committee to continually assess the basis for its decision-making, 23 insure consistency with the NAS guidelines, and maintain the scientific credibility of the AEGL 24 values and accompanying scientific rationale. This ongoing effort is continuously documented 25 and is identified as the "SOP Manual". 26 27 4.4 ROLE OF THE DIRECTOR OF THE AEGL PROGRAM 28 29 The Director has the overall responsibility for the AEGL Program, including the 30 NAC/AEGL Committee and its interface with other programs and organizations in the public and 31 private sectors nationally and internationally. More specifically he is responsible for the overall 32 management of the AEGL Program as it relates to: 33 34 • NAC/AEGL Committee and AEGL Program objectives of scientific credibility, quality, 35 productivity and cost effectiveness. 36 37 • AEGL Program resource needs. 38 39 • Fostering a collaborative spirit among Committee members, Staff Scientists of the 40 organization which drafts Technical Support Documents, and interested parties from all 41 participating organizations in the public and private sectors. 42 Sop08-02 wpd Printed July 6, 2000 117 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 • Matters related to the U. S National Academy of Sciences. 2 3 • Expanding the scope of the AEGL Program, including international participation. 4 5 4.5 ROLE OF THE DESIGNATED FEDERAL OFFICER 6 7 The Designated Federal Officer (DFO) serves as the administrative officer of the 8 Committee to insure that all operations, processes, and general precedures function properly and 9 efficiently. The DFO serves as an executive secretariat to the NAC/AEGL Committee and has 10 the responsibility for: 11 12 • Effective communication/coordination with NAC/AEGL Committee members, the 13 Committee Chair, the organization which drafts Technical Support Documents, and 14 interested parties in the public and private sector. 15 16 • Day-to-day administrative management of the NAC/AEGL Committee with respect to the 17 agenda for future meetings, distribution of Technical Support Documents and other 18 correspondence with Committee members, maintenance of meeting minutes, tapes of 19 meetings and other important Committee records, funding and other financial matters and 20 Committee membership matters. 21 22 • Administrative management of quarterly meetings including responsibility for all Federal 23 Register Notices related to NAC/AEGL Committee activities, minutes and decision 24 making records, meeting venues, facilities, and equipment, as well as the assurance that 25 the meetings are held in compliance of the Federal Advisory Committee Act (FACA). 26 27 • Ensuring compliance with the FACA on all matters that extend beyond the quarterly 28 meetings such as the submission of appropriate reports to the U.S. Office of Management 29 and Budget (OMB) and the Library of Congress. 30 31 4.6 ROLE OF THE NAC/AEGL COMMITTEE CHAIR 32 33 The NAC/AEGL Committee Chair is appointed by EPA as specified in the Federal 34 Advisory Committee Act (FACA) and is selected from the Committee membership. The Chair's 35 responsibilities include conducting and directing specific activities to insure the effective and 36 efficient conduct of business by the Committee: 37 38 • Support in the planning and preparation of upcoming meetings by collaborating with the 39 AEGL Program Director, the DFO and the organization which drafts Technical Support 40 Documents, including the review of the meeting agenda. 41 42 • Manage the NAC/AEGL Committee meetings in an effective and efficient manner to Sop08-02.wpd Printed July 6, 2000 118 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 insure completion of the agenda for each meeting. 2 3 • Attempt to reach a consensus of the Committee by msunng adequate time for 4 presentation of differing opinions and focusing on the major issues to break deadlocks or 5 stalemates. 6 7 • Participation in scientific matters on AEGLs as related to the U. S. National Academy of 8 Sciences. 9 10 • Participate with the AEGL Progam Director and the DFO in evaluating and improving 11 Committee activities and expanding the scope of the AEGL Program. 12 13 4.7 CLASSIFICATION OF THE STATUS OF AEGL VALUES 14 15 Draft AEGL Values are AEGL values proposed by the AEGL Development Team (see 16 section 4.8) prior to the full NAC/AEGL Committee discussion and approval. 17 18 Proposed AEGL Values are AEGL values which have been formally approved and 19 elevated to "Proposed" status by a consensus or two-thirds majority of a quorum of the 20 NAC/AEGL Committee. 21 22 Interim AEGL Values are AEGL values formally approved by the NAC/AEGL 23 Committee and elevated to "Interim" status after publication in the Federal Register, response to 24 comments, and appropriate adjustments made by the Committee. These "Interim" AEGLs are 25 forwarded to the Committee on Toxicology, National Research Council, National Academy of 26 Sciences for review and comment by the Subcommittee on Acute Exposure Guideline Levels 27 (NAS/AEGL Subcommittee). 28 29 Final AEGL Values are AEGL values which have been reviewed and finalized by the U. 30 S. National Academy of Sciences (NRC NAS) and are published inder the auspices of the NAS. 31 32 Sop08-02 wpd Printed July 6.2000 119 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 4.8 ROLE OF AEGL DEVELOPMENT TEAMS 2 3 Each AEGL Development Team consists of a Staff Scientist from the organization which 4 drafts Technical Support Documents and a Chemical Manager and two Chemical Reviewers, 5 who are members of the NAC/AEGL Committee. The primary function of the NAC/AEGL 6 Development Team is to provide the NAC/AEGL Committee with Draft AEGL values and a 7 Technical Support Document (TSD) containing relevant data and information on the chemical 8 and the derivation of the Draft AEGLs. The Staff Scientist provides the initial effort by 9 identifying and preliminarily evaluating available data from varied resources including on-line 10 literature databases, other databases, journal reviews, secondary source reviews, unpublished 11 data, federal and state documents and other sources, including accounts of accidents in the 12 workplace or in the community (see Section 2.3). Interaction takes place among the Chemical 13 Manager, the Chemical Reviewers, and the Staff Scientist during the development of the TSD 14 and the Draft AEGL values. The resulting document is then distributed and reviewed by 15 Committee Members prior to a formal meeting and attempts are made to resolve issues of 16 concern expressed by Committee Members prior to distribution of the TSD to the NAC/AEGL 17 Committee and formal presentation and discussion at a Committee meeting. 18 19 4.8.1 Role of a Chemical Manager 20 21 The Chemical Manager has the overall responsibility for the development of the "Draft", 22 "Proposed", and "Interim" AEGL values and their presentation to the rest of the NAC/AEGL 23 Committee and to the NAS Committee for evaluation of "Final" AEGLs. The Chemical Mangers 24 serve on a rotating basis as the Committee's principal representative on the AEGL Development 25 Team for a specific chemical. The Chemical Manager in turn selects two Committee members to 26 serve as Chemical Reviewers. 27 28 The Chemical Manager collaborates with the Staff Scientist and the Chemical Reviewers 29 on the development of the AEGLs, the supporting rationale, and the Technical Support 30 Documents. In instances where the Chemical Manager has accepted the responsibilities, taken 31 ownership for the AEGL values, resolved scientific issues, and led the discussions with 32 Committee members, the Committee has moved rapidly toward the development of a consensus. 33 Where the Chemical Manager's role has been less decisive, the Committee's deliberations have 34 been more protracted, less focused, and highly inefficient. Implicit in the description of the 35 Chemical Manager's role is the expectation that he/she will work with the Staff Scientist, the 36 Chemical Reviewers, and the rest of the Committee members to develop exposure guidance 37 levels that are appropriate and scientifically credible. It is expected that the Chemical Manager 38 will achieve a consensus within the AEGL Development Team on the issues related to the 39 development of the AEGL values prior to the meeting of the full Committee. Further, as time 40 permits, the Chemical Manager will attempt to resolve issues raised by individual Committee 41 Members prior to the scheduled Committee meeting. 42 Sop08-02 wpd Printed July 6, 2000 120 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 The following is a summary outline of specific activities and responsibilities of the 2 Chemical Manager within the NAC/AEGL Committee: 3 4 • To participate as the leader of the ad hoc AEGL Development Team. 5 6 • To select and utilize two Chemical Reviewers as technical support. 7 8 • To provide direct support to the Staff Scientist assigned to the chemical in the 9 development of the Technical Support Documents (TSD), the "Draft" AEGL values, 10 and the supporting rationale. 11 12 • To serve as liaison among Committee members and the Staff Scientist during the 13 development of draft AEGL values and the Technical Support Document. 14 15 • To resolve scientific issues prior to the Committee meetings such as: 16 Completeness of data gathering (published/unpublished). 17 Selection of key and supporting data (following guidelines). 18 Interpretation of data. 19 Credibility of AEGL values (use of appropriate methodology). 20 Validity of scientific rationale for AEGLs. 21 Other (as necessary for development of scientifically credible AEGL values). 22 23 • To seek consensus of Committee members by resolving issues with individual Committee 24 members prior to the Committee meeting. 25 26 • To frame important scientific issues related to the chemical and the AEGLs for 27 presentation at the Committee meeting (i.e. significant issues that cannot be resolved 2 8 before the meeting). 29 30 • To participate in the presentation of AEGL values, supporting rationale and important 31 issues at the Committee meeting in collaboration with the Staff Scientist. 32 33 • To oversee appropriate follow-up activities: 34 Revisions as appropriate (AEGL values, TSD, rationales). 35 Toxicity testing. 36 FR Notice comments (conversion of "Proposed" to "Interim" values). 37 Preparation of AEGL proposal to NAS. 38 39 4.8.2 Role of a Chemical Reviewer 40 41 42 • To participate as a member of the ad hoc AEGL Development Team Sop08-02 wpd Printed July 6, 2000 121 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2 • To conduct a detailed review of the assigned document and key references. 3 4 • To assist the Chemical Manager and Staff Scientist in evaluating the data, the candidate 5 AEGLs, and the scientific rationale for their support. 6 7 • To participate actively in discussions of the document during AEGL Committee 8 meetings. 9 10 • To stand in for the Chemical Manager if and when he/she is unable to perform his/her 11 duties. 12 13 4.8.3 Role of an Staff Scientist at the Organization which Drafts Technical 14 Support Documents 15 16 The Staff Scientist has the primary responsibility for data gathering, data evaluation, 17 identification of potential key data and supporting data, identification of potential methodologies, 18 calculations, and extrapolations, and the preparation of the Technical Support Document. This 19 includes the following tasks: 20 21 • To participate as a member of the ad hoc AEGL Development Team 22 23 • To participate with the others on the AEGL Development Team in the development of 24 "Draft" AEGL values and their presentation at the NAC/AEGL Committee meetings 25 26 • To prepare Technical Support Documents (TSD) in a timely manner and make 27 appropriate revisions based upon discussions and decisions of the AEGL 28 Development Team and later based upon the discussions and decisions of the 29 NAC/AEGL Committee. 30 31 • To develop and maintain a data file on the chemical substance. 32 33 • To present a summary of the data and information on the substance in collaboration with 34 the Chemical Manager at the AEGL Committee meetings. 35 36 • To provide continuing support to an assigned chemical through the "Draft," "Proposed," 37 "Interim," and "Final" stages of AEGL development, including preparation for, and 38 response to, Federal Register Notice review and comment. 39 40 4.9 ROLE OF NAC/AEGL COMMITTEE MEMBERS 41 42 Sop08-02 wpd Printed July 6. 2000 122 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 • To review all Technical Support Documents in advance of meetings and to work out 2 issues with the Chemical Manager at the earliest possible date. The importance of 3 resolving issues before Committee meetings is greatly emphasized to increase the 4 efficiency and productivity of the meetings. 5 6 • To circulate Technical Support Documents to other qualified scientists within their 7 respective organizations or other organizations as appropriate to broaden the 8 evaluation by the scientific community. 9 10 • To serve as experts in specific areas or on specific scientific issues (e.g. sensitive human 11 sub-populations, etc.) as a member of an ad hoc task force under the SOP Workgroup 12 chair. 13 14 • To volunteer as a Chemical Manager at least once a year and to select chemicals where a 15 significant contribution to the development of credible AEGL values can be made 16 based on special knowledge, expertise, or past experience. 17 18 • To assist in the application of AEGLs in appropriate programs within the organization the 19 Committee member represents. 20 21 • To make suggestions for modification or expansion of the Chemical Priority List by 22 providing lists of chemicals and supporting rationale for their priority to the 23 Designated Federal Officer (DFO). 24 25 • To attend all scheduled NAC/AEGL Committee meetings and to participate in the 26 discussions and decision making of all AEGL values. AEGL values are approved or 27 disapproved by a vote of 2/3 majority of a quorum, with a quorum defined as the 28 presence of more than 50 percent of the total NAC/AEGL Committee membership.. 29 30 4.10 ROLE OF THE ORGANIZATION THAT DRAFTS TECHNICAL 31 SUPPORT DOCUMENTS 32 33 The role of the organization that drafts the TSDs is to provide the principal technical 34 support in gathering and evaluating the relevant scientific data and information from all sources, 35 including preparation and/or revision of the Technical Support Documents (TSDs) following the 36 guidance provided in this SOP Guidance Manual. As a member of the AEGL Development 37 Team, to collaborate with the Chemical Manager and Chemical Reviewers in the preparation and 38 distribution of "Draft" AEGLs, the supporting rationale, and the TSDs for the NAC/AEGL 39 Committee members. Provide continuing technical and administrative support to assigned 40 chemicals through the "Draft," "Proposed," "Interim," and "Final" stages of AEGL development, 41 with revisions based upon the consensus or majority opinion of the NAC/AEGL Committee and 42 the NAS/AEGL Subcommittee. SopOS-02 wpd Printed July 6. 2000 123 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 To provide the Staff Scientists, the administrative personnel, and the facilities and 2 equipment necessary for data gathering, maintenance of databases, dissemination of relevant 3 information to Committee members, presentations or co-presentations (with Chemical Managers) 4 at the NAC/AEGL Committee meetings, development and revisions of TSDs, preparation of 5 submissions to the Federal Register, summarization of Federal Register (F.R.) comments and 6 identification of important scientific issues, presentations to the Committee on F.R. comments, 7 and preparation of technical information to be entered on the Internet. 8 9 To distribute the TSDs to companies and other interested parties as directed by the DFO 10 after review and comment by the NAC/AEGL Committee. This distribution to interested parties 11 will be only by request through the Designated Federal Officer (DFO). This initial distributed 12 version will be without the AEGL values and the rationale used to derive them and will occur 13 between 1-14 days prior to the Committee meeting. 14 15 Sop08-02.wpd Printed July 6. 2000 124 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 i 5. REFERENCES 2 3 40 CFR (Code of Federal Regulations). 2000. Protection of human subjects. Office of the 4 Federal Register National Archives and Records Administration, Part 26, Source- 56 FR 5 28012, 28022, June 18, 1991, unless otherwise noted. 6 7 Adams, E.M., Spencer, H.C., Rowe, V.K., McCollister, D.D., Irish, D.D. 1952. Vapour toxicity 8 of carbon tetrachloride determined by experiments on laboratory animals. A.M. A. Arch 9 Ind. Hyg. Occup. Med. 6: 50-66. 10 11 Alder, H., Roessler, R. 1968. Introduction to Probability and Statistics. Fourth Edition. W. H. 12 Freeman and Company. 333p. 13 14 Aleksieva, Z. 1983. Sulphur compounds. In: Encyclopedia of Occupational Health and Safety, 15 Prameggiani L. (Ed). Geneva: Internaitonal Labour Office, pp. 2122-2124. 16 17 Allen, B.C., Kavlock, R.J., Kimmel, C.A., Faustman, E.M. 1994 Dose-response assessment for 18 developmental toxicity. II. Comparison of generic benchmark dose estimates with no 19 observed adverse effect levels. Fundamental and Appl. Toxicol. 23:487-495. 20 21 AAAI (American Academy of Allergy and Immunology). 1986. American Academy of Allergy 22 and Clinical Immunology. Executive Committee. Position Statement: Clinical Ecology. 23 J. Allergy Clin. Immunol. 78:269-71. 24 25 AAAAI (American Academy of Allergy Asthma and Immunology). 1999. Physician Reference 26 Materials: Position Statement 35: Idiopathic Environmental Tolerances. AAAAI Board 27 of Directors. JACI, Vol. 103. January. Pp. 36-40. Available at 28 http://www.aaaai.org/professional/physicianreference/positionstatements/ps35.stm. 29 Accesses March 15, 1999. 30 31 Armitage, P., Doll, R. 1961. Stochastic models for carcinogenesis. Pp. 19-38 in Proceedings of 32 the Fourth Berkeley Symposium on Mathematical Statistics and Probability, Vol. 4, J. 33 Neyman, ed. Berkeley, Calif.: University of California Press. 34 35 Ashford, N. 1999. Low-level chemical sensitivity: implications for research and social policy. 36 Toxicol. and Ind. Health, 15:421-427. 37 38 Ashford, N., Miller, C. 1998. Chemical Exposures. Low Levels and High Stakes. vanNostrand 39 Reinhold, 2nd Ed. 40 41 Atherley, G. 1985. A critical review of time-weighted average as an index of exposure and dose, 42 and of its key elements. Am. Ind. Hyg. Assoc. J. 46:481-487. Sop08-02 wpd Printed July 6, 2000 125 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Balmes, J.R., Fine, J.M., Sheppard, D. 1987. Symptomatic bronchoconstriction after short-term 2 inhalation of sulfur dioxide. Amer. Rev. Respir. Disease 136-1117-1121. 3 4 Barnes, G.B., Daston, G.P., Evans, J.S., Jarabek, A.M., Kavlock, R.J., Kimmel, C.A., Park, C., 5 Spitzer, H.L. 1995. Benchmark dose workshop: Criteria for use of a benchmark dose to 6 estimate a reference dose. Regulatory Toxicol. and Pharmacol. 21:296-306. 7 8 Barnes, J.M., Denz, F.A. 1954. Experimental methods used in determining chronic toxicity. 9 Pharmacol. Rev. 6:191-242. 10 11 Barnes, D.G. and ML. Dourson. 1988. Reference dose (RfD): Description and use in health risk 12 assessments. Reg. Toxicol. Pharmacol. 8:471-486. 13 14 Bigwood, E.J. 1973. The Acceptable Daily Intake of Food Additives. CRC Crit. Rev. Toxicol. 15 June, 41-93. 16 17 Bliss, C.I. 1935a. The Calculation of the Dosage-Mortality Curve. Ann. Appl. Bio. 22: 18 134-167. 19 20 Bliss, C.I. 1935b. The Comparison of the Dosage-Mortality Data. Ann. Appl. Bio. 22: 307-333. 21 22 Bogdanffy, M.S., Sarangapani, R., Plowchalk, D.R., Jarabek, A., Andersen, M.E. 1999. A 23 biologically based risk assessment for vinyl acetate-induced cancer and noncancer 24 inhalation toxicity. Toxicol. Sci. 51:19-35. 25 26 Brown, N. A., Fabro, S. 1983. The value of animal teratogenicity testing for predicting human 27 risk. Clin. Obst. Gynecol. 26(2): 467-477. 28 29 Brown-DeGagne, A-M, McClone, J. 1999. Multiple Chemical Sensitivity: A Test of the 30 Olfactory-Limbic Model. J. Occ. Environ. Med. 41(5):366-377. 31 32 Bruckner, J.V., Weil, W.B. 1999. Biological factors which may influence an older child's or 33 adolescent's responses to toxic chemicals. Regul. Toxicol. Pharmacol. 29(2 Pt 1):158- 34 164. 35 36 Calabrese, E.J. 1983. Principles of Animal Extrapolation. Wiley, New York. (Repnnted 1991) 37 38 Calabrese, E.J. 1985. Uncertainty factors and mterindividual variation. Regulator. Toxicol. 39 Pharmacol. 5:190-196. 40 41 Chen, J.J., Kodell, R.L., Gaylor, D.W. 1988. Using the biological two-stage model to assess nsk 42 from short-term exposures. Risk Analysis. 8:223-230. Sop08-02 wpd Printed July 6. 2000 126 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Crump, K.S., Howe, R.B.. 1984. The multistage model with a time-dependent dose pattern: 2 Applications to carcinogenic risk assessment. Risk Analysis 4:163-176. 3 4 Cullen, M.R. 1987. The worker with multiple chemical sensitivities-an overview. Occup. Med 5 2(4):655-661. 6 7 David, A., Frantik, E., Holusa, R., Novakova, O. 1981. Role of time and concentration on 8 carbon tetrachloride in rats. Int. Arch. Occup. Environ. Health. 48: 49-60. 9 10 Done, A.K. 1964. Developmental pharmacology. Clin. Pharmacol. Therap. 5:432-479. (as cited 11 inNRC, 1993b). 12 13 Dourson, M. 1996. Editorial: Uncertainty factors in noncancer risk assessment Regul. Toxicol. 14 Pharmacol. 24(2): 107. 15 16 Dourson, M,L. and J.F. Stara, 1983. Regulatory history and experimental support of uncertainty 17 (safety) factors. Reg. ToxicoL Pharmacol. 3: 224-238. 18 19 Dourson, M.L, S.P. Felter, and D. Robinson. 1996. Evolution of science-based uncertainty 20 factors in noncancer risk assessment. Regulatory Toxicol. Pharmacol. 24:108-120. 21 22 Dourson, M.L., Knauf, L.A., Swartout, J.C. 1992. On reference dose (RfD) and its underlying 23 toxicity data base. Toxicol. and Indust. Health 8:171-189. 24 25 ECETOC (European Chemical Industry Ecology and Toxicology Centre) 1991. Emergency 26 exposure indices for industrial chemicals. Tech. Report no. 43, Brussels, pp.28-29. 27 28 Evans, R.D., Harris, R.S, Bunker, J.W.M. 1944. Radium metabolism in rats, and the production 29 of osteogenic sarcoma by experimental radium poisoning. Amer. J. Roentgenol. 52:353- 30 373. 31 32 Faustman, E.M., Allen, B.C.,Kavlick, R.J., Kimmel, C.A. 1994. Dose-response assessment for 33 developmental toxicity. I. Characterization of database and determination of no 34 observed adverse effect levels. Fund, and Appl. Toxicol. 23:478-486. 35 36 FDA (U.S. Food and Drug Administration). 1985. Sponsored compounds in food-producing 37 animals: Proposed rule and notice. Fed. Regist. 50:45530-45556. 38 39 Finney, D.J. 1971. Probit Analysis. Third Edition. Cambridge University Press, London. 40 41 Flury, F 1921. Uber kampfgasvergiftungen. I. Uber reizgase. Z. Ges. Exp. Med. 13: 1. 42 Sop08-02 wpd Printed July 6. 2000 127 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Food Safety Council 1982. A Proposed Food Safety Evaluation Process. The Nutrition 2 Foundation, Inc., Washington, D. C. 3 4 Fowles, J.R., Alexeeff, G.V., Dodge, D. 1999. The use of benchmark dose methodology with 5 acute inhalation lethality data. Regulator. Toxicol. and Pharmacol. 29:262-278. 6 7 Frank, R. 1980. SO2-particulate interactions: recent observations. Amer. J. Ind. Med. 1:427- 8 434. 9 10 Frederick C.B., Bush, M.L., Lomax, L.G., Black, K.A., Finch, L., Kimbell, J.S., Morgan, K.T., 11 Subramaniam, R.P., Morris, J.B., Ultman, J.S. 1998. Application of a hybrid 12 computational fluid dynamics and physiologically based inhalation model for interspecies 13 dosimetry extrapolation of acidic vapors in the upper airways. Toxicol. Appl. Pharmacol. 14 152:211-31. 15 16 Gardner, D.E., Miller, F.J., Blommer, E.J., Coffin, D.L. 1979. Influence of exposure mode on the 17 toxicity of NO2. Environ. Health Persp. 30: 23-29. 18 19 Gardner, D.E., Coffin, D.L., Pinigm, M.A., Siderenko, G.I. 1977. Role of time as a factor in the 20 toxicity of chemical compounds in intermittent and continuous exposures. Part I.: Effects 21 of continuous exposure. Part II: Effects of intermittent exposure. J.Toxicol. Environ. 22 HealthS: 811-282. 23 24 Gaylor, D.W. 1996. Quantalization of continuous data for benchmark dose estimation. 25 Regulatory Toxicol. and Pharmacol. 24:246-250. 26 27 Gaylor, D.W., Ryan, L., Krewski, D., Zhu, Y. 1998. Procedures for calculating benchmark 28 doses for health risk assessment. Regulatory Toxicol. and Pharmacol. 28:150-164. 29 30 Gaylor, D.W., Kodell, R.L., Chen, J.J., Krewski, D. 1999. A unified approach to risk 31 assessment for cancer and noncancer endpoints based On benchmark doses and 32 uncertainty/safety factors. Regulatory Toxicol. and Pharmacol. 29:151-157. 33 34 Glaubiger, D.L., von Hoff, D.D., Holcenbern, J.S., Kamen, B., Pratt, C., Ungerleider, R.S. 1982 35 The relative tolerance of children and adults to anticancer drugs. Front. Radiat. Ther. 36 Oncol. 16: 42-49. 37 38 Goddard, M.J., Murdoch, M.J., Krewski, D. 1995. Temporal aspects of risk characterization. 39 Inhal. Toxicol. 7: 1005-1018. 40 41 Goldenthal, E.I. 1971. A compilation of LD50 values in newborn and adult animals. Toxicol 42 Appl. Pharmacol. 18:185-207. Sop08-02 wpd Printed July 6, 2000 128 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Graveling R.A., Pilkington, A., George, J.P.K., Butler, M.P., Tannahill, S.N. 1999. A review of 2 multiple chemical sensitivities. Occup. Environ. Med. 56:73-85 3 4 Gregory, G.A., Eger, E.I., Munson, E.S. 1969. The relationship between age and halomethane 5 requirements in man. Anesthesiology. 30: 488-491. 6 7 Guth, D.J., Jarabek, A.M., Wymer, L., Hertzberg, R.C. 1991. Evaluation of risk assessment 8 methods for short-term inhalation exposure. Paper presented at the 84th Annual Meeting 9 of the Air and Waste Management Association, Paper No. 91 -173.2, June 16-21, 1991, 10 Vancouver, B.C. 11 12 Haber, F. 1924. Zur Geschichte des Gaskrieges. In: Fiinf Vortrage aus den Jahren 1920-1923. 13 Springer-Verlag, Berlin, pp. 76-92. 14 15 Hattis, D., E. Anderson, E. 1999. What should be the implications of uncertainty, variability, 16 and inherent "biases'V'conservatism" for risk management decision-making? Risk Anal. 17 19(1):95-107. 18 19 Hattis, D., Barlow, K. 1996. Human interindividual variability in cancer risks-technical and 20 management challenges. Human Ecol. Risk Assesses. 2(1): 194-220. 21 22 Hattis, D., D.E. Burmaster, D.E. 1994. Assessment of variability and uncertainty distributions 23 for practical risk analyses. Risk Anal. 14(5):713-730. 24 25 Hattis, D., Erdreich, L., Ballew, M. 1987. Human variability in susceptibility to toxic 26 chemicals: A preliminary analysis of pharmacokinetic data from normal volunteers. Risk 27 Anal. 7(4):415-426. 28 29 Hattis, D., Banati, P., Goble, R. 1999. Distributions of individual susceptibility among humans 30 for toxic effects—for what fraction of which kinds of chemicals and effects does the 31 traditional 10-fold factor provide how much protection? Annals of the New York 32 Academy of Sciences. 895:286-316. 33 34 Hayes, W.J. 1967. Toxicity of pesticides toman: Risks from present levels. Proc. Roy. Soc. 35 London 167 (1007): 101-127. 36 37 Henderson, R. 1992. Short-term exposure guidelines for emergency response: The approach of 38 the Committee on Toxicology. Conference on Chemical Risk Assessment in the 39 Department of Defense (DOD): Science Policy and Practice. Ed. Harvey J. Clewell III. 40 American Conference of Governmental Industrial Hygienists Inc. Cincinnati, Ohio. 89- 41 92. 42 Sop08-02 wpd Printed July 6. 2000 129 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Henschler, D. 1984. Exposure limits: History, philosophy, future developments. Ann. Occup 2 Hyg. 28: 59-72. 3 4 Hertzberg, R.C., Miller, M. 1985. A statistical model for species extrapolation using categoncal 5 response dat. Toxicol. Ind. Health 1:43-57. 6 7 Hertzberg, R.C., Wymer, L. 1991. Modeling the severity of toxic effects. Paper presented at the 8 84* Annual Meeting of the Air and Waste Management Association, Paper No. 91-173.4, 9 June 16-21,1991, Vancouver, B.C. 10 11 Homan, E.R. 1972. Quantitative relationships between toxic doses of anti-tumor 12 chemotherapeutic agents in animals and man. Cancer Chemother. Rep. 3:13-19. 13 14 Horstman, D.H., Seal E., Folinsbee, L. J., Ives, P., Roger, J. 1988. The relationship between 15 exposure duration and sulfur dioxide-induced bronchoconstnction in asthmatic subjects. 16 Am. Ind. Hyg. J. 49:38-47. 17 18 Howe, R.B. Crump, K.S. Van Landingham, C. 1986. GLOBAL86: A Computer Program to 19 Extrapolate Quantal Animal Toxicity Data to Low Doses. Prepared for U.S. 20 Environmental Protection Agency, Washington, D.C.. Subcontract No. 2-251U-2745. 21 22 IPCS (International Programme on Chemical Safety) 1994. Environmental Health Criteria No. 23 170: Assessing human health risks of chemicals: Derivation of guidance values for 24 health-based exposure limits. World Health Organization, Geneva 25 26 IPCS (International Programme on Chemical Safety) 1996. Report of multiple chemical 27 sensitivities workshop. February 21-23, 1996. Berlin. Germany. 28 29 Kalberlah, F., Schneider, K. 1998. Quantification of extrapolation factors. Final report of the 30 research project No. 116 06 113 of the Federal Environmental Agency. Bundesanstalt fur 31 Arbeitsschutz und Arbeitsmedizin, Hauptsitz Dortmund: ISBN 3-89701-107-7. 206 32 pages. 33 34 Katoh, T., Ikeda, K. 1992. Minimum alveolar concentration of seroflurane in children. Brit. J. 35 Anaesth. 68: 139-141. 36 37 Kepler, G.M., Richardson, R.B., Morgan, K.T., Kimbell, J.S. 1998. Computer simulation of 38 mspiratory nasal airflow and inhaled gas uptake in a rhesus monkey. Toxicol. Appl. 39 Pharmacol. 150:1-11. 40 41 Keplinger, M.L., Suissa, L.W.. 1968. Toxicity of fluorine short-term inhalation. Am. Ind. Hyg. 42 Assoc.J. 29:10-18. Sop08-02 wpd Printed July 6, 2000 130 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Kimbell, J.S., Gross, E.A., Joyner, D.R., Godo, M.N., Morgan, K.T. 1993. Application of 2 computational fluid dynamics to regional dosimetry of inhaled chemicals in the upper 3 respiratory tract of the rat. Toxicol. Appl. Pharmacol. 121: 253-263. 4 5 Kimbell, J.S., Gross, E.A., Richardson, R.B., Conolly, R.B., Morgan, K.T. 1997a. Correlation 6 of regional formaldehyde flux predictions with the distribution of formaldehyde-induced 7 squamous metaplasia in F344 rat nasal passages. Mutation Research 380:143-154. 8 9 Kimball, J.S., Godo, M.N., Gross, E.A., Joyner, D.R., Richardson, R.B., Morgan, K.T 1997b. 10 Computer simulation of inspiratory airflow in all regions of the F344 rat nasal passages. 11 Toxicol. Appl. Pharmacol. 145:388-398. 12 13 Kipen, H.M., Fiedler, N. 1999. Invited Commentary: Sensitivities to Chemicals-Context and 14 Implications. Amer. J. Epi. 150(1):13-16. 15 16 Klaassen, C.D., Doull, J. 1980. Evaluation of Safety: Toxicological Evaluation. In Toxicology 17 (C. D. Klaassen and M.O. Amdur, eds.), MacMillan, New York:p. 26. 18 19 Kodell, R.L., Gaylor, D.W., Chen, J.J. 1987. Using average life-time dose rate for intermittent 20 exposure to carcinogens. Risk Anal. 7:339-345. 21 22 Koenig, J.Q., Pierson, W. E., Horike M., et al. 1981. Effects of SO2 plus NaCl aerosol combined 23 with moderate exercise on pulmonary functions in asthmatic adolescents. Environ Res. 24 25:340-348. 25 26 Koenig, J.Q., Dumler, K., Reboddedo, V., Williams, P.V., Pierson, W.E. 1993. Respiratory 27 effects of inhaled sulfunc acid on senior asthmatics and nonasthmatics. Archiv. Environ 28 Health. 48:171-175. 29 30 Kokoski.CJ. 1976. Written testimony of Charles J. Kokoski, Docket No. 76N-0070. DHEW, 31 Food and Drug Administration, Washington, D.C. (as cited in Dourson and Stara, 1983). 32 33 34 Kreutzer, R., Neutra, R.R., Lashuay, N. 1999. Prevalence of people reporting sensitivities to 35 chemicals in a population-based survey. Amer. J. of Epi. 150(1):1-12. 36 37 LeDez, K. M., Lerman, J. 1987. The minimum alveolar concentration (MAC) of isoflurane in 38 preterm neonates. Anesthesiology. 67: 301-307. 39 40 Lehman, A.J., Fitzhugh, O.G. 1954. 100-Fold Margin of Safety. Assoc. Food Drug Off. U.S.Q. 41 Bull. 18:33-35. 42 Sop08-02 wpd Printed July 6, 2000 13 1 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Leisenring, W. Ryan, L. 1992. Statistical properties of the NOAEL. Regulatory Toxicol. and 2 Pharmacol. 15:161-171. 3 4 Lerman, J., Robinson, S., Willis, M.M., Gregory, G.A. 1983. Anesthetic requirements for 5 halothane in young children 0-1 months and 1-6 months of age. Anesthesiology. 59: 6 421-424. 7 8 Linn.W.S., Avol E.L., Peng R. C., Shamoo, D.A., Hackney, J.D. 1987. Replicated dose-response 9 study of sulfur dioxide effects in normal, atopic, and asthmatic volunteers. Am. Rev. 10 Respir. Dis. 136: 1127-1134. 11 12 Litchfield Jr., J.T., Wilcoxon, F. 1948. A simplified method of evaluating dose-effect 13 experiments. J. Pharmacol. Exp. Ther. 96: 99-113. 14 15 Lu, F.C. 1979. Assessments at an international level of health hazards to man of chemicals 16 shown to be carcinogenic in laboratory animals. In Regulatory Aspects of Carcinogenesis 17 (F. Coulston, ed.), Academic Press, New York :pp. 315-328. 18 19 Lu, F.C. 1988. Acceptable daily intake: Inception, evolution, and application. Regul. Toxicol. 20 Pharmacol. 8:45-60. 21 22 Lu, F.C., Sielken, R.L. 1991. Assessment of safety/nsk of chemicals: Inception and evolution of 23 the ADI and dose-response modeling procedures. Toxicol. Lett. 59:5-40. 24 25 Marsoni, S., Ungerleider, R.S., Hurson, S.B., Simon, R.M., Hammershaimb, L.D. 1985. 26 Tolerance to antineoplastic agents in children and adults. Cancer Treat. Rep. 69: 27 1263-1269. 28 29 Matthew, M.T.V., Mainland, P., Gin, T. 1996. Minimum alveolar concentration of halothane 30 and enflurane are decreased in early pregnancy. 85: 782-786. 31 32 McLellan, R.K., Becker, C.E., Borak, J.B., Coplein, C., Ducatman, A.M., Favata, E.A., Green, 33 J.F., Herzstein, J., Holly, A.T., Kalnas, J., Kulig, K., Kipen, H.M., Logan, D.C., Mitchell, 34 F.L., McKinnon, H.W., Roberts, M.A., Russi, M., Sawyer, H.J., Sepulveda, M.J., Upfal, 35 M.J., Zepeda, M.C. 1999. ACOEM Position statement. Multiple chemical sensitivities: 36 Idiopathic environmental intolerance. J. Occup. Environ. Med. 41(11):940-942. 37 38 Meek, M.E., Newhook, R., Liteplo, R.G., Armstrong, V.C. 1994 Approach to assessment of 39 risk to human health for priority substances under the Canadian Environmental Protection 40 Act. Environ. Carcinogen. Ecotoxicol. Rev. C12 (2). 105-134. 41 42 Melton, C.E. 1982. Effect of long-term exposure to low levels of ozone: a review. Aviat. Space Sop08-02 wpd Printed July S. 2000 132 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Environ. Med. 53: 105-111. 2 3 Moolgavkar,S.H., Luebeck.G. 1990. Two-event model for carcmogenesis: Biological, 4 mathematical, and statistical considerations. Risk Analysis. 10:323-341 5 6 Momson. P.P. 1987. Effects of time-variant exposure on toxic substance response. Environ. 7 Health Perspect. 76-133-140. 8 9 Murdoch, D.J., Krewski, D. 1988. Carcinogenic risk assessment with time-dependent exposure 10 patterns. Risk Analysis. 8. 521-530. 11 12 Murdoch, D.J., Krewski, D., Wargo, J. 1992. Cancer risk assessment with intermittent 13 exposure. Risk Analysis 12-569-577. 14 15 NAS (National Academy of Sciences). 1977. Drinking Water and Health. Washington, D.C., 16 National Academy Press. 17 18 NRC (National Research Council). 1983. Risk Assessment in the Federal Government. 19 Washington, D.C., National Academy Press. 20 21 NRC (National Research Council). 1986. Criteria and Methods for Preparing Emergency 22 Exposure Guidance Level (EEGL), Short-Term Public Emergency Guidance Level 23 (SPEGL), and Continuous Exposure Guidance Level (CEGL) Documents. Washington, 24 D.C., National Academy Press. Appendix F. 25 26 NRC (National Research Council). 1992a. Guidelines for Developing Spacecraft, Maximum 27 Allowable Concentrations for Space Station Contaminants. Washington, D.C., National 28 Academy Press. 29 30 NRC (National Research Council). 1992b. Multiple Chemical Sensitivities: A Workshop. 31 Washington, D.C., National Academy Press. 32 33 NRC (National Research Council). 1992c. Biologic Markers in Immunotoxicology. 34 Washington, D.C., National Academy Press. 35 36 NRC (National Research Council), Committee on Toxicology. 1993a. Guidelines for 37 Developing Community Emergency Exposure Levels for Hazardous Substances. 38 Washington, DC., National Academy Press. 39 40 NRC (National Research Council). 1993b. Pesticides in the Diets of Infants and Children. 41 Washington, D.C., National Academy Press. 42 Sop08-02 wpd Printed July 6, 2000 13 3 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 NRC (National Research Council). 1994. Science and Judgement in risk Assessment. 2 Washington, D.C., National Academy Press. 3 4 NTP (National Toxicology Program). 1987. Toxicology and carcinogenesis studies of ethylene 5 oxide (CAS No. 75-21-8) in B6C3F, Mice (Inhalation Studies). NTP TR 326, NIH Publ. 6 No. 88-2582, U.S. Department of Health and Human Services, Research Triangle Park, 7 NC. 8 9 OECD. OECD's Guidelines for the Testing of Chemicals. 10 http://www.oecd.org/ehs/test/testlist.htm 11 12 P/CC (President/Congressional Commission on Risk Management). 1997. Final Report. Vol 2. 13 U.S. Government Printing Office, Washington, D.C. 14 15 Pieters, M.N., Kramer, H.J. 1994. Concentration • time = constant ? The validity of Haber's Law 16 in the extrapolation of discontinuous to continuous exposition. Rapportnummer 659101 17 002, National Institute for Public Health and Environmental Protection, The Netherlands. 18 19 Plummer, J.L., Hall, P., de la Ilsey, A.H., Jenner, M.A., Cousins, M J. 1990. Influence of 20 enzyme induction and exposure profile on liver injury due to chlorinated hydrocarbon 21 inhalation. Pharmacol. Toxicol. 67:329-335. 22 23 Pohl, H.R., Abdin, H.G. 1995. Utilizing uncertainty factors in minimal risk levels derivation. 24 Regul. Toxicol. Pharmacol. 22:180-188. 25 26 Rail, D.P. 1969. Difficulties in extrapolating the results of toxicity studies in laboratory animals 27 to man. Environ. Res. 2:360-367. 28 29 Renwick, A.G. 1993. Data derived safety factors for the evaluation of food additives and 30 environmental contaminants. Food Additives and Contaminants. 10(3):275-305. 31 32 Rhomberg, L.R., Wolff, S.K. 1998. Empirical scaling of single oral lethal doses across 33 mammalian species based on a large database. Risk Analysis. 18: 741-753. 34 35 Rinehart, W.E., Hatch, T. 1964. Concentration-time product (CT) as an expression of dose in 36 sublethal exposures to phosgene. Ind. Hyg. J. 25: 545-553. 37 38 Rondinelli, R.C.A., Koemg, J.Q., Marshall.S.G. 1987. The effects of sulfur dioxide on 39 pulmonary function in healthy nonsmoking male subjects aged 55 years and older. Am. 40 Ind. Hyg. Assoc. J. 48:299-303. 41 42 Roger, L. J., Kehrl H.R., Hazucha M., Horstman, D.H. 1985. Bronchoconstriction in asthmatics SopOB-02 wpd Printed July 6, 2000 134 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 exposed to sulfur dioxide during repeated exercise. J. Appl. Physiol. 59: 784-791 2 3 Savolamen, H., Kurppa, K., Pfaffli, P., Kivisto, H. 1981. Dose-related effects of dichloromethane 4 on rat brain in short-term inhalation exposure. Chem. Biol.Interact. 34: 315-322 5 6 Sheenan, D.M., Gaylor, D.W. 1990. Divisions of reproductive & developmental toxicology and 7 biometry. Teratology 41:590-591 (abstract). 8 9 Simon, R.A. 1986. Sulfite sensitivity. Ann. Allergy 56:281-288. 10 11 Spiegel, M.R. 1996. Schaum's Outline of Theory and Problems of Statistics. Second Edition. 12 McGraw-Hill:519 pages. 13 14 Stevens, W.C., Dolan, W.M., Gibbons, R.T., White, A., Eger, E.I., Miller, R., deJong, R.H., 15 Elashoff, R.M. 1975. Minimum alveolar concentration (MAC) of isoflurane with and 16 without nitrons oxide in patients of various ages. Anesthesiology. 42: 196-200 17 18 ten Berge, W.F. Zwart, A., Appelman, L.M. 1986. Concentration-time mortality response 19 relationship of irritant and systemically acting vapours and gases. J. Hazard. Materials. 20 13:301-309. 21 22 Truhaut, R. 1991. The concept of the acceptable daily intake: an historical review Food 23 Additives and Contaminants. 8:151-162. 24 25 U.S. DHHS (Department of Health and Human Services). 1995. Multiple Chemical Sensitivity- 26 A Scientific Overview. ATSDR (Agency for Toxic Substances and Disease Registry). F. 27 Mitchell, Ed. National Academy Press. 28 29 U. S. EPA (Environmental Protection Agency). Health Effects Test Guidelines. 30 http://www.epa.gov/docs/OPPTS_HarmonizeoV 31 32 U. S. EPA (Environmental Protection Agency). 1980. Guidelines and methodology used in the 33 preparation of health assessment chapters of the consent decree water quality criteria. 34 Fed. Regist. 45:79347-79357. 35 36 U. S. EPA (Environmental Protection Agency). 1986. Guidelines for carcinogen nsk 37 assessment. Federal Register 51(185): 33992-34003. 38 39 U. S. EPA (Environmental Protection Agency). 1987. Technical Guidance for Hazards 40 Analysis. Emergency Planning for Extremely Hazardous Substances. U. S. 41 Environmental Protection Agency in conjunction with the Federal Emergency 42 Management Agency and the Department of Transportation. Sop08-02 wpd Printed July 6, 2000 13 5 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 U. S. EPA (Environmental Protection Agency). 1991. National primary drinking water 2 regulations; Final rule. 40 CFR Parts 141,142, and 143. Federal Register 56(20) January 3 30, 1991:3526-3597. 4 5 US. EPA (Environmental Protection Agency). 1992. Draft report: A cross-species scaling 6 factor for carcinogen risk assessment based on equivalence of mg/kg^Vday. Federal 7 Register 57(109): 24152-24173. 8 9 U. S. EPA (Environmental Protection Agency). 1993. Styrene. The Integrated Risk Information 10 System (IRIS), http://epa.gov/iris/. 11 12 U. S. EPA (Environmental Protection Agency). 1994a. Glossary of Risk Assessment Related 13 Terms, February 1, 1994. Integrated Risk Information System (IRIS). 14 http://www.epa.gov/ngispgm3/iris/glossary.htm 15 16 U.S. EPA (Environmental Protection Agency). 1994b. Methods for Derivation of Inhalation 17 Reference Concentrations and Application of Inhalation Dosimetry. EPA/600/8-90/066F. 18 October 1994. 19 20 U. S. EPA (Environmental Protection Agency). 1995a. The Use of the Benchmark Dose 21 Approach in Health Risk Assessment. Risk Assessment Forum. EPA/630/R-94/007. February 22 1995. 23 24 U. S. EPA (Environmental Protection Agency). 1995b. Methyl mercury. The Integrated Risk 25 Information System (IRIS), http://epa.gov/iris/. 26 27 U. S. EPA (Environmental Protection Agency). 1996a. Proposed Guidelines for Carcinogen 28 Risk Assessment. EPA/600/P-92/003C. April 1996. 29 30 U. S. EPA (Environmental Protection Agency). 1996b. Aroclor 1016. The Integrated Risk 31 Information System (IRIS), http://epa.gov/ins/. 32 33 U. S. EPA (Environmental Protection Agency). 2000. National Center for Environmental 34 Assessment. Benchmark Dose Software Draft Beta Version 1.2. 35 http://www.epa.gov/ncea/bmds.htm. 36 37 U S. PHS (Public Health Service). 1998. A report on multiple chemical sensitivity (MCS). A 38 predecisional draft. The Environmental Health Policy Committee. (The Interagency 39 Workgroup on Multiple Chemical Sensitivity). U.S. Public Health Service. August 24. 40 41 van Stee, E.W., Boorman, G. A., Moorman, M.P., Sloane, R.A. 1982. Time varying 42 concentration profile as a determinant of the inhalation toxicity of carbon tetrachlonde. Sc.p08-02.wpd Printed July 6. 2000 136 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 J.Toxicol. Environ. Health 10: 785-795. 2 3 Vettorazzi, G. 1976. Safety factors and their application in the lexicological evaluation. In the 4 evaluation of Toxicological Data for the Protection of Public Health. Pergamon, 5 Oxford:pp. 207-223. 6 7 Vettorazzi, G. (1980) Handbook of International Food Regulatory Toxicology. Evaluations, 8 Spectrum, New York, Vol. I:pp. 66-68. 9 10 Vocci, F., Farber, T. 1988. Extrapolation of animal toxicity data toman. Regulatory Toxicol. 11 and Pharmacol. 8: 389-198. 12 13 Weil, C. 1972. Statistics vs safety factors and scientific judgment in the evaluation of safety for 14 man. Toxicol. Appl. Pharmacol. 21:454-463. Sop08-02 wpd Printed July 6, 2000 13 7 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 i APPENDIX A. NAC/AEGL PROGRAM PERSONNEL. 2 3 ADMINISTRATIVE MANAGEMENT: 4 Roger Garrett Director, AEGL Program, U.S. Environmental Protection Agency 5 Paul S. Tobin Designated Federal Officer, NAC/AEGL Committee, 6 U.S. Environmental Protection Agency 7 8 NAC/AEGL COMMITTEE MEMBERS AND THEIR AFFILIATIONS 9 George Rusch Chair, Honeywell, Inc. 10 Ernest Falke Chair, SOP Workgroup, U S. Environmental Protection Agency 11 12 George AlexeefF Air Toxicology & Epidemiology California EPA 13 Steven Barbee Olin Corporation 14 Lynn Beasley U.S. Environmental Protection Agency (5204G) 15 David Belluck Minnesota Pollution Control Agency 16 Robert Benson U.S. Environmental Protection Agency Region VIE 17 Jonathan Borak American College of Occupational and Environmental Medicine 18 (ACOEM) 19 William Bress Vermont Department of Health 20 George Cushmac Department of Transportation 21 Larry Gephart Exxon Mobil Biomedical Sciences, Inc. 22 Doan Hanson Brookhaven National Laboratory (DOE Alternate) 23 John P. Hinz Armstrong Laboratory/Occupational and Environmental Health Directorate (AF) 24 Jim Holler Agency for Toxic Substances Disease Registry 25 Thomas C. Homshaw Illinois Environmental Protection Agency 26 Nancy K. Kim New York State Department of Health 27 Loren Koller College of Veterinary Medicine, Oregon State University 28 Dr. Glenn Leach U.S. Army Center for Health Promotion & Preventive Medicine 29 Mark A. McClanahan Centers for Disease Control & Prevention 30 John Morawetz International Chemical Workers Union 31 Richard Niemeier National Institute for Occupational Safety and Health 32 Marinelle Payton Harvard Medical School 33 Zarena Post Texas Natural Resource Conservation Commission 34 George Rodgers American Association of Poison Control Centers (AAPCC) 35 Michelle Schaper Mine Safety and Health Administration 36 Robert Snyder Environmental and Occupational Health Sciences Institute 37 Thomas J. Sobotka Food and Drug Administration HFS-507 38 Kenneth Still Medical Service Corp/U.S. Navy 39 Judy Strickland U.S. Environmental Protection Agency (Pending) 40 Richard Thomas International Ctr for Environmental Hlth 41 Thomas Tuccinardi U.S. Department of Energy 42 Sop08-02 wpd Printed July 6. 2000 A 1 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30,2000 1 PAST COMMITTEE MEMBERS 2 Kyle Blackman Federal Emergency Management Agency 3 Luz Claudio Mt. Sinai Medical Center 4 Benjamin Jackson Consultant 5 William Pepelko U.S. Environmental Protection Agency 6 Patricia Talcott University of Idaho, Dept of Food Science & Toxicology 7 Sop08-02.wpd Printed July 6. 2000 A 2 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 i APPENDIX B. PRIORITY LISTS OF CHEMICALS 2 3 A master list of approximately 1,000 acutely toxic chemicals was initially compiled 4 through the integration of individual priority lists of chemicals submitted by each U. S. federal 5 agency placing a representative on the Committee. The master list was subsequently reviewed by 6 individuals from certain state agencies and representatives from organizations in the private 7 sector and modified as a result of comments and suggestions received. The various priority 8 chemical lists were compiled separately by each federal agency based on their individual 9 assessments of the hazards, potential exposure, risk, and relevance of a chemical to their 10 programmatic needs. 11 12 On May 21, 1997, a list of 85 chemicals was published in the Federal Register. This list 13 identified those chemicals to be of highest priority across all U. S. federal agencies and 14 represented the selection of chemicals for AEGL development by the NAC/AEGL Committee for 15 the first two to three years of the program. The Committee has now addressed these chemicals 16 and they are presently in the Proposed, Interim, or Final stages of development. Certain 17 chemicals did not contain an adequate database for AEGL development and, consequently, are on 18 hold pending decisions regarding further toxicity testing. This initial "highest" priority list of 85 19 chemicals is shown below. 20 21 A second "working list" of approximately 100 priority chemicals is being selected from 22 the original master list, or from new, high priority candidate chemicals submitted by U. S. 23 Agencies and organizations and by OECD member countries that are planning to participate in 24 the AEGL Program. Although "working lists" will be published in the U. S. Federal Register 25 and elsewhere from time-to-time to indicate the NAC/AEGL Committee's agenda, the priority of 26 chemicals addressed, and , hence, the "working list" is subject to modification if priorities of the 27 NAC/AEGL Committee or individual stakeholder organizations, including international 28 members, change during that period. 29 30 31 Initial List of 85 Priority Chemicals for Acute Exposure Guideline Level 32 (AEGL) Development* 33 34 35 ORGANIZATION LISTS USED TO COMPILE THE MASTER LIST AND THE INITIAL 36 LIST OF 85 PRIORITY CHEMICALS 37 38 'ATSDR Medical Managment Agency for Toxic Substances and Disease Registry 39 M = Chemicals with an ATSDR Medical 40 Management Guideline 41 T = Chemicals with an ATSDR Toxicology Profile 42 Sop08-02 wpd Printed July 6. 2000 B 1 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 2DOD 3DOE SCAPA "DOT ERP 5EPACAA112b 6EPACAA112r 7EPA Superfund 8OSHA PSM 9OSHA STEL IONIOSH IDLH "Seveso Annex in Department of Defense A = Army Toxicity Summary Chemical C = Chemical Weapons Convention Schedule 3.A Toxic Chemical Cs = Chemical Stockpile Emergency Preparedness Program (CSEPP) Chemical I = Air Force Installation Restoration Program Chemical N = Navy Chemical S = Strategic Environmental Research and Development Program (SERDP) Chemical DOE Subcommittee for Consequence Assessment and Protective Action Chemical Department of Transportation Emergency Response Guidebook P = Priority DOT ERG Chemical O = Other ERG Chemical Environmental Protection Agency Clean Air Act 112b Chemical Environmental Protection Agency Clean Air Act 112b Chemical (+ = SARA s.302 also) Environmental Protection Agency Superfund Chemical OSHA Process Safety Management Chemical OSHA Short-term Exposure Limit Chemical NIOSH Immediately Dangerous to Life or Health Chemical International Seveso Convention List * The initial list of 85 priority chemicals shown below has been created by identifying the highest priority hazardous chemicals from the Master List. This initial list is a starting point for the development of AEGL values by the National Advisory Committee for Acute Exposure Guideline Levels for Hazardous Chemicals (NAC/AEGL). However, the list of chemicals is subject to modification, pending changes in priorities recommended by the various stakeholders that make up the NAC/AEGL. While it is anticipated that most of these chemicals will remain as Sop08-02 wpd Printed July 6. 2000 B2 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 high priority for AEGL development, changes to the list could occur. The NAC/AEGL hopes to 2 select 30 to 40 chemicals per year to address in the AEGL development process. Consequently, 3 the initial list will expand as the NAC/AEGL continues to address chemicals of interest to its 4 member organizations. 5 Sop08-02.wpd Printed July 6. 2000 B 3 ------- 1 2 Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 TABLE B-l. PRIORITY LIST OF CHEMICALS CAS NO. 56-23-5 57-14-7 60-34-4 62-53-3 67-66-3 68-12-2 71-43-2 71-55-6 74-90-8 74-93-1 75-09-2 75-21-8 75-44-5 75-55-8 75-56-9 CHEMICAL Carbon tetrachloride 1,1-Dimethyl hydrazine Methyl hydrazine Aniline Chloroform Dimethylformamide Benzene 1,1,1 -Trichloroethane Hydrogen cyanide Methyl mercaptan Methylene chloride Ethylene oxide Phosgene Propyleneimine Propylene oxide 'ATSDR T M T X T M T MT MT M 2DOD: AIS AIS AIS X c AIS C 3DOE SCAPA' X X X X 4DOT ERG P P P P P P P 5EPA CAA 112b X X X X X X X X X X X X X X 6EPA CAA 112r X+ x+ + x+ x+ x+ x+ x+ x+ x+ 7EPA Super fund X X X X X X X •OSHA ,PSM X X X X X X Seves 0 Annex III X X X X X 'OSHA STEL "N10SH IDLH X X X X X X X X X X X 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Sop08-02 wpd Printed July 6, 2000 B4 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30. 2000 CAS NO. 75-74-1 75-77-4 75-78-5 75-79-6 78-82-0 79-01-6 79-21-0 79-22-1 91-08-7 106-89-8 107-02-8 107-11-9 107-12-0 107-15-3 107-18-6 107-30-2 CHEMICAL Tetramethyllead Trimethychlorosilane Dimethyldichlorosilane Methyltrichlorosilane Isobutyronitrile Trichloroethylene Peracetic acid Methy chloroformate Toluene 2,6-diisocyanate Epichlorohydrin Acrolein Allyl amine Propionitrile Ethylenediamine Allyl alcohol Chloromcthyl methyl ether 'ATSDR MT M T 2DOD AIS 3E»OE SCAPA X X 4DOT ERG P P P O 5EPA CAA 112b X X X X X 6EP-A CAA 112r X+ x+ x+ x+ x+ x+ x+ x+ x+ x+ x+ x+ x+ x+ x+ 7EPA Super fund • X X 'OSHA PSM X X X X X X X Seves 0 Annex HI X X X X X X 'OSHA STEL X X "NIOSH IDLH X X X X X 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Sop08-02 wpd Printed July 6. 2000 B5 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 CAS NO. 108-23-6 108-88-3 108-91-8 109-61-5 110-00-9 1 10-89-4 123-73-9 126-98-7 127-18-4 151-56-4 302-01-2 353-42-4 506-77-7 509-14-8 540-59-0 540-73-8 CHEMICAL Isopropyl chloroformatc Toluene Cyclohexylamine Propyl chloroformate Furan Piperidine Crotonaldehyde, (E) Methacrylonitrile Tetrachloroethylene Ethyleneimine Hydrazine Boron triflounde compound with methyl ether (1 I) Cyanogen chloride Tetranitromethane 1 ,2-Dichloroethylene 1 ,2-Dimethylhydrazine 'ATSDR MT T T T 2DOD AINS AIS I '3DOE SCAPA X X X "DOT ERG P O O P P 5EPA CAA 112b X X X X X 6EPA CAA 112r- X+ X+ x+ x+ x+ x+ x+ x+ x+ x+ .x+ x+ x+ 7EPA Super fund X X X 'OSHA PSM X X X X Seves 0 Annex III X 'OSHA STEL X "NIOSH IDLH X X X X X X X 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Sop08-02 wpd Printed July 6, 2000 B6 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 CAS NO. 584-84-9 594-42-3 624-83-9 811-97-2 814-68-6 1330-20-7 1717-00-6 4170-30-3 6423-43-4 7446-09-5 7446-11-9 7647-01-0 7647-01-0 7664-39-3 CHEMICAL Toluene 2,4-diisocyanate Perchloromcthylmcrcaptan Methyl isocyanate HFC 134A (1,1,1,2- Tetrafluoroethane) Acrylyl chloride Xylenes (mixed) HCFC 141b(l,l- Dichloro- 1 -fluoroethane) Crotonaldehyde cis & trans mixture Propylene glycol dinitrate (Otto Fuel II) Sulfur dioxide Sulfur trioxide Hydrogen chloride Hydrochloric acid Hydrogen fluoride 'ATSDR M X T M 2DOD N AIN N Navy 3DOE SCAPA "DOT ERG P P P P P P P 5EPA CAA 112b X X X X X X 6EPA CAA 112r X+ X+ x+ x+ x+ x+ x+ x+ x+ x+ 7EPA Super fund X X X X 'OSHA PSM X X X X X X X X Seves 0 Annex III X X X X X X 'OSHA STEL X X X X X "NIOSH IDLH X X X X X X X X 1 2 3 4 5 6 7 10 11 12 13 14 Sop08-02 wpd Printed July 6, 2000 B7 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 CAS NO. 7664-41-7 7664-93-9 7697-37-2 7719-12-2 7726-95-6 7782-41-4 7782-50-5 7783-06-4 7783-60-0 7783-81-5 7784-34-1 7784-42-1 7790-91-2 7803-51-2 8014-95-7 10025-87-3 CHEMICAL Ammonia Sulfuric acid Nitric acid Phosphorus trichloride Bromine Fluorine Chlorine Hydrogen sulfide Sulfur tetrafluoride Uranium hexafluoride Arsenous trichloride Arsine Chlorine trifluoride Phosphine Oleum Phosphorus oxychlonde 'ATSDR MT M M M M 2DOD 3DOE SCAPA X X X X X 4DOT ERG P P P P P P P P P 0 P P o SEPA CAA 112b X X X X 6EPA CAA I12r X+ + X+ x+ x+ x+ x+ x+ x+ x+ x+ x+ x+ x+ 7EPA Super fund X X X X 'OSHA PSM X X X X X X X X X X X X Seves 0 Annex III X X X X X 'OSHA STEL X X X X X IONIOSH IDLH X X X X X X X X X X 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Sop08-02 wpd Printed July 6, 2000 B8 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 CAS NO. 10049-04-4 10102-43-9 10102-44-0 10294-34-5 13463-39-3 13463-40-6 19287-45-7 25323-89-1 70892-10-3 163702-07-6 163702-08-7 CHEMICAL Chlorine dioxide Nitric oxide Nitrogen dioxide Boron trichloride Nickel carbonyl Iron, pentacarbonyl- Diborane Trichloroethane Jet fuels (JP-5 and JP-8) Methyl nonafluorobutyl ether (HFE 7 100 component) Methyl nonafluorobutyl ether (HFE 7 100 component) 'ATSDR T ,2DOD AS N N N 3DOR SCAPA X X X 'DOT ERG P X P P P P 5EPA CAA 112b X X 6EPA CAA 112r X X+ X x+ x+ x+ x+ 7EPA Super fund X 'OSHA PSM X X X X X X X Seves o Annex III X 'OSHA STEL X X X '•NIOSH IDLH X X X 1 2 3 4 5 6 7 8 9 10 11 12 Sop08-02 wpd Printed July 6. 2000 B9 ------- 2 3 4 5 6 Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 APPENDIX C. DIAGRAM OF THE AEGL DEVELOPMENT PROCESS FIGURE C-l THE AEGL DEVELOPMENT PROCESS AEGL Development Process Non-published. NoTHpeer Reviewed IndusdyData Published Literature Search Older Data / Information Sources Speoal Toxraty Studies AEGL Dm Team- Saentis, Manager. Rene Blopni6nt ORNL •herncal rfherracel were Technical Support Documents (TSDs) Oistnbute Draft or Proposed TSDs/ AEGLs to COCTTlttOG Members + NAC/AEGL Draft or Proposed AEGLs !S Sop08-02 wpd Pnnted July 6. 2000 C 1 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 1 AAPCC ACGIH ACOEM ADI AEGL AFL-CIO AIHA ATSDR BMC BMC05 BMC.o CAAA CAER CAS CDC CEEL CEL CMA CORR COT Ct or Cxt CURE DFO DOD DOE DOT DTIC ECETOC EEGL EEL Einsatztoleranzwert EPA ERP ERPG APPENDIX D. GLOSSARY - ACRONYMS, ABBREVIATIONS, AND SYMBOLS American Association of Poison Control Centers American Conference of Government Industrial Hygienist American College of Occupational and Environmental Medicine Acceptable Daily Intake National Advisory Committee for Acute Exposure Guidelines Levels for Hazardous Substances (AEGL Committee) American Federation of Labor - Congress of Industrial Organizations American Industrial Hygienist Association Agency for Toxic Substances and Disease Registry (U. S.) Benchmark Concentration Benchmark Concentration, 5% response Benchmark Concentration, 10% response Clean Air Act Amendments (U. S. EPA) Community Awareness and Emergency Response Chemical Abstract Service (U. S.) Centers for Disease Control and Prevention (U. S. HHS) Community Emergency Exposure Levels (U. S. NAS) Emergency Exposure Limits (U. S. NAS) Chemical Manufacturers Association (U. S.) Chemicals on Reporting Rules Committee on Toxicology (U. S. NAS) Measure of cumulative exposure Chemical Unit Record Estimates Designated Federal Official Department of Defense (U. S.) Department of Energy (U. S.) Department of Transportation (U. S.) Defense Technical Information Center (U. S.) European Chemical Industry Ecology and Toxicology Centre Emergency Exposure Guideline Levels (U. S NAS) Emergency Exposure Limits (U. S. NAS) [Action Tolerance Levels] Federation for the Advancement of German Fire Prevention (Germany) Environmental Protection Agency (U. S.) Emergency Response Planning, (U. S.) American Industrial Hygiene Association (AIHA) Emergency Response and Planning Guidelines, (U. S.) American Sop08-02 wpd Printed July 6. 2000 D2 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 FACA FDA FEDRIP FEMA FEV, FR FYI GLP GSA HEAST HSDB HUD IARC IDLH IPCS IRIS LC01 LCSO LCL LOAEL MAC MAC MAK MAKS. MCS MF MLE MLEOI MTD N/A NAAQS NAC NAC/AEGL NAS NAS/AEGL NASA Industrial Hygiene Association (AIHA) Federal Advisory Committee Act (U. S.) Food and Drug Administration (U. S.) Federal Research in Progress Federal Emergency Management Agency (U. S.) Forced Expiratory Volume Federal Register (U. S.) For Your Information Good Laboratory Practice Standards General Services Administration (U. S.) Health Effects Assessment Tables Hazardous Substances Data Base Department of Housing and Urban Development (U. S.) International Agency for Research on Cancer Immediately Dangerous to Life and Health (U. S. NIOSH) International Programme for Chemical Safety Integrated Risk Information System Lethal Concentration, 1 % kill Lethal Concentration, 50 % kill Lower Confidence Limit Lowest-observe-adverse effect level Mean Alveolar Concentration Maximum Acceptable Concentration (The Netherlands) [Maximale Arbeitsplatzkonzentration] Maximum Workplace Concentration, 8 hour time weighted average German Research Association (Germany) Spitzenbegrenzung (Kategorie 11,2) [Peak Limit 11,2] 30 minute x 2 per day (Germany) Multiple Chemical Sensitivity Modifying Factor Maximum Likelihood Estimate Maximum Likelihood Estimate, 1% response Maximum Tolerated Dose Not Applicable National Ambient Air Quality Standards, U.S. National Advisory Committee National Advisory Committee for Acute Exposure Guideline Levels for Hazardous Substances (NAC/AEGL Committee) National Academy of Sciences (U. S.) National Academy of Sciences Subcommittee on Acute Exposure Guideline Levels (NAS/AEGL Subcommittee) (U. S.) National Aeronautical and Space Administration (U. S.) Sop08-02 wpd Pnnted July 6. 2000 D3 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 NCI NIOSH NOAEL NRC NSF NTIS NTP OECD ORNL OSHA OSWER PEL-TWA PEL-STEL QA QC QSARs REL-STEL REL-TWA RfC RfD RTECS SARA SMAC SOP SPEGL STPL TARA TLV-STEL TLV-TWA TSD UF WEELS > > < National Cancer Institute (U. S.) National Institute for Occupational Safety and Health (U. S.) No Observed-Adverse-Effect Level National Resource Council (U. S.) National Science Foundation (U. S.) National Technical Information Services (U. S.) National Toxicology Program (U. S.) Organization for Economic Cooperation and Development Oak Ridge National Laboratories (U. S.) Occupational Safety and Health Administration (U. S.) Office of Solid Waste and Emergency Response Permissible Exposure Limits - Time Weighted Average (U. S. OSHA) Permissible Exposure Limits - Short Term Exposure Limit (U. S. OSHA) Quality Assurance Quality Control Quantitative Structure Activity Relationships Recommended Exposure Limits-Short Term Exposure Limit (U. S. NIOSH) Recommended Exposure Limits-Time Weighted Average (U. S. NIOSH) Reference Concentration (U. S. EPA) Reference Dose (U. S. EPA) Registry of Toxic Effects of Chemical Substances Superfund Amendments and Reauthorization Act (CERCLA) Spacecraft Maximum Allowable Concentrations Standing Operating Procedures Manual Short-term Public Exposure Guideline Levels (U. S. NRC, NAS) Short Term Public Limits (U. S. NAS) Toxicology And Risk Assessment Document List (ORNL) Threshold Limit Value - Short Term Exposure Limit (U. S. ACGIH) Threshold Limit Value - Time Weighted Average (U. S. ACGIH) Technical Support Document Uncertainty Factor Workplace Environmental Exposure Levels (AIHA) Greater than Greater than or equal to Less than Sop08-02 wpd Printed July 6, 2000 D4 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30,2000 1 < Less than or equal to 2 % Percent 3 4 dl - Deciliter 5 gorgm -- Gram 6 hr. — Hour 7 urn — Micrometer 8 ug — Microgram 9 mg ~ Milligram 10 min — Minute 11 mL - Milliliter 12 mm — Millimeter 13 ppb — Parts per billion 14 ppm — Parts per million 15 ppt — Parts per trillion 16 Sop08-02 wpd Printed July 6. 2000 D 5 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 i APPENDIX E. EXAMPLE OF A TABLE OF CONTENTS 2 IN A TECHNICAL SUPPORT DOCUMENT 3 4 TABLE OF CONTENTS 5 6 PREFACE 2 7 8 LIST OF TABLES 5 9 10 EXECUTIVE SUMMARY 6 11 12 1. INTRODUCTION 9 13 14 2. HUMAN TOXICITY DATA 10 15 2.1. Acute Lethality 10 16 2.2. Nonlethal Toxicity 10 17 2.2.1. Acute Studies 10 18 2.2.2. Epidemiologic Studies 11 19 2.3. Developmental/Reproductive Toxicity 11 20 2.4. Genotoxicity 11 21 2.5. Carcinogenicity 11 22 2.6. Summary 11 23 24 3. ANIMAL TOXICITY DATA 11 25 3.1. Acute Lethality 12 26 3.1.1. Nonhuman Primates 12 27 3.1.2. Dogs 12 28 3.1.3. Rats 12 29 3.1.4. Mice 13 30 3.1.5. Hamsters 14 31 3.2. Nonlethal Toxicity 14 32 3.2.1. Nonhuman Primates 14 33 3.2.2. Dogs 14 34 3.2.3. Rats 15 35 3.2.4. Mice 15 36 3.3. Developmental/Reproductive Toxicity 15 37 3.4. Genotoxicity 18 38 3.5. Carcinogenicity 19 39 3.6. Summary 19 40 41 4. SPECIAL CONSIDERATIONS 20 SopOS-02 wpd Printed July 6. 2000 E 1 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 4.1. Metabolism and Disposition 20 2 4.2. Mechanism of Toxicity 20 3 4.3. Structure-Activity Relationships 21 4 4.4. Other Relevant Information 21 5 4.4.1. Species Variability 21 6 4.4.2. Unique Physicochemical Properties 22 7 4.4.3. Concurrent Exposure Issues 22 8 9 5. DATA ANALYSIS FOR AEGL-1 22 10 5.1. Summary of Human Data Relevant to AEGL-1 22 11 5.2. Summary of Animal Data Relevant to AEGL-1 22 12 5.3. Derivation of AEGL-1 22 13 14 6. DATA ANALYSIS FOR AEGL-2 23 15 6.1. Summary of Human Data Relevant to AEGL-2 23 16 6.2. Summary of Animal Data Relevant to AEGL-2 23 17 6.3. Derivation of AEGL-2 23 18 19 7. DATA ANALYSIS FOR AEGL-3 24 20 7.1. Summary of Human Data Relevant to AEGL-3 24 21 7.2. Summary of Animal Data Relevant to AEGL-3 24 22 7.3. Derivation of AEGL-3 24 23 24 8. SUMMARY OF AEGLS 25 25 8.1. AEGL Values and Toxicity Endpoints 25 26 8.2. Comparison with Other Standards and Criteria 26 27 8.3. Data Adequacy and Research Needs 27 28 29 9. REFERENCES CITED 29 30 31 Appendix A (Derivation of AEGL Values) 32 32 Appendix B (Time Scaling Calculations for Dimethylhydrazme AEGLs) 36 33 Appendix C (Carcinogenicity Assessment for Dimethylhydrazme) 39 34 Appendix D (Derivation Summary for Dimethylhydrazme AEGLs) 41 35 36 Sop08-02 wpd Printed July 6, 2000 £ 2 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 i APPENDIX F. EXAMPLE OF AN EXECUTIVE 2 SUMMARY IN A TECHNICAL SUPPORT DOCUMENT 3 4 EXECUTIVE SUMMARY 5 6 Dimethylhydrazine occurs as a symmetrical (1,2-dimethylhydrazine) and unsymmetrical 7 (1,1-dimethylhydrazine) isomer. Unless otherwise specified, dimethylhydrazine refers to 8 unsymmetrical dimethylhydrazine in this document. Both compounds are clear, colorless liquids. 9 Unsymmetrical dimethylhydrazine (1,1-dimethylhydrazine) is a component of rocket fuels and is 10 also used as an absorbent for acid gas, as a plant growth control agent, and in chemical synthesis. 11 Although it has been evaluated as a high-energy rocket fuel, commercial use of the symmetrical 12 isomer (1,2-dimethylhydrazine) is limited to small quantities and it is usually considered to be a 13 research chemical. Because data are limited for 1,2-dimethylhydrazine (symmetrical 14 dimethylhydrazine), the AEGL values for both isomers are based upon 1,1-dimethylhydrazine 15 (unsymmetrical). Limited data suggest that 1,1-dimethylhydrazine may be somewhat more toxic 16 than 1,2-dimethylhydrazine. 17 18 Data on acute exposures of humans to both isomers of dimethylhydrazine are limited to 19 case reports of accidental exposures. Signs and symptoms of exposure include respiratory 20 irritation, pulmonary edema, nausea, vomiting, and neurological effects. However, definitive 21 exposure data (concentration and duration) were unavailable for these exposures. The limited 22 data in humans suggest that the nonlethal toxic response to acute inhalation of dimethylhydrazine 23 is qualitatively similar to that observed in animals. No information was available regarding 24 lethal responses in humans. In the absence of quantitative data in humans, the use of animal data 25 is considered a credible approach for developing AEGL values. 26 27 Toxicity data of varying degrees of completeness are available for several laboratory 28 species, including, rhesus monkeys, dogs, rats, mice, and hamsters (Weeks et al., 1963). Most of 29 the animal studies were conducted using 1,1-dimethylhydrazine, although limited data suggest 30 that 1,2-dimethylhydrazine exerts similar toxic effects. Minor nonlethal effects such as 31 respiratory tract irritation appear to occur at cumulative exposures of < 100 ppm-hrs. At 32 cumulative exposures of 100 ppm-hrs, or slightly greater than this level more notable effects 33 have been reported, including, muscle fasciculation, behavioral changes, tremors, and 34 convulsions. Lethality has been demonstrated when cumulative exposures exceed these levels 35 only slightly. The available data suggest that there is a very narrow margin between exposures 36 resulting in no significant toxicity and those causing substantial lethality (LC50 = 900-2,000 ppm- 37 hrs). 38 39 Developmental toxicity of dimethylhydrazmes has been demonstrated in rats following 40 parenteral administration of maternally toxic doses. 41 Sop08-02 wpd Printed July 6, 2000 F 1 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 Both isomers of dimethylhydrazine have been shown to be carcinogenic in rodents 2 following oral exposure and 6-month inhalation exposure to 1,1-dimethylhydrazine. Increased 3 tumor incidence was observed in mice, although these findings are compromised by the 4 contaminant exposure to dimethylmtrosarnine. An increased incidence of lung rumors and 5 hepatocellular carcinomas was also seen in rats but not in similarly exposed hamsters. Inhalation 6 slope factors are currently unavailable. 7 8 AEGL-1 values for dimethylhydrazine are not recommended. This is due to inadequate 9 data to develop health-based criteria, and because the concentration-response relationship for 10 dimethylhydrazine indicated a very narrow margin exists between exposures producing no toxic 11 response and those resulting in significant toxicity. 12 13 Behavioral changes and muscle fasciculations in dogs exposed for 15 minutes to 360 ppm 14 1,1-dimethylhydrazine (Weeks et al., 1963) served as the basis for deriving AEGL-2 values. 15 Available lethality data in dogs and rats indicated a near linear temporal relationship (n=0.84 and 16 0.80 for dogs and rats, respectively). For temporal scaling (C1 x t = K) to derive values for 17 AEGL-specific exposure durations a linear concentration-response relationship; n=l was used. 18 This value was adjusted by an uncertainty factor of 30. An uncertainty factor of 3 for 19 interspecies variability was applied because the toxic response to dimethylhydrazine was similar 20 across the species tested. This was especially true for lethality responses among rats, mice, dogs, 21 and hamsters with LCSO values for time periods ranging from 5 minutes to 4 hours. A 22 comparison of LC50 values for the same exposure durations in these species did not vary more 23 than 3-fold. An uncertainty factor of 10 was used for mtraspecies variability. This was based 24 primarily on the variability in the toxic response observed in dogs where responses varied from 25 one of extreme severity (vomiting, tremors, convulsions, and death) to no observable effects. 26 Additionally, experiments by Weeks et al. (1963) indicated that dogs previously stressed by 27 auditory stimuli may have potentiated their response to dimethylhydrazine. Based on these data, 28 it was assumed that humans may be equally divergent in their response to dimethylhydrazine as a 29 result of similar stresses. 30 31 The AEGL-3 values were derived from the 1 -hr LC50 (981 ppm) for 1,1 - 32 dimethylhydrazine in dogs (Weeks et al., 1963). Because of the steep slope of the dose-response 33 curve of 1,1-dimethyl hydrazine, the 1 hour LC50 of 981 ppm was adjusted downward to estimate 34 the lethality threshold of 327 ppm. An uncertainty factor of 3-fold for interspecies variability 35 was applied for several reasons. The 4-hr LC50 values for mouse, rat, and hamster differ by a 36 factor of approximately 2 and were consistent with the dog data when extrapolated from 1 hr 37 using n=l. The more sensitive species, the dog, was used to derive the AEGL-3 values. An 38 uncertainty factor of 10 for intraspecies variability was used since a broad spectrum of effects 39 were seen including behavioral effects, hyperactivity, fasciculations, tremors, convulsions, and 40 vomiting. The mechanism of toxicity is uncertain and sensitivity among individuals may vary. 41 Following identical exposures, the responses of the dogs varied from one of extreme seventy 42 (vomiting, tremors, convulsions, and death) to no observable effects. Temporal scaling as Sop08-02 wpd Printed July 6. 2000 F 2 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 previously described was applied to obtain exposure values for AEGL-specific exposure periods. Verified inhalation and oral slope factors were unavailable for dimethylhydrazine. A cancer assessment based upon the carcinogenic potential (withdrawn cancer slope factors) of dimethylhydrazine revealed that AEGL values for a 10"4 carcinogenic risk exceeded the AEGL- 2 values that were based on noncancer endpoints. Because the cancer risk for dimethylhydrazine was estimated from nonverified cancer estimates, and because AEGLs are applicable to rare events or single once-in-a-lifetime exposures to a limited geographic area and small population, the AEGL values based on noncarcinogenic endpoints were considered to be more appropriate. SUMMARY OF AEGL VALUES FOR 1,1- and 1,2-DIMETHYLHYDRAZINES Classification AEGL-1 (Nondisablmg) AEGL-2 (Disabling) AEGL-3 (Lethal) 30-mm NR 6 pptn 14.7 mg/m3 22ppm 54 mg/m3 1-hour NR 3ppm 7 4 mg/m3 11 ppm 27 mg/m3 4-hour MR 0 75 ppm 2 mg/m1 2 7 ppm 6 6 mg/m3 8-hour MR 0 38 ppm 1 mg/m3 1 4 ppm 3 4 mg/m3 Endpomt(Reference) Not recommended due to insufficient data, concentration-response relationships suggest little margin between exposures causing minor effects and those resulting in senous toxicity * Behavioral changes and muscle fasciculations in dogs exposed to 360 ppm for 1 5 minutes (Weeks ct al , 1963) Lethality threshold of 327 ppm for 1 hr estimated from 1 -hr LCM in dogs (Weeks et al . 1 963) NR: Not Recommended. Analysis of dimethylhydrazine toxicity data in total revealed that significant toxicity may occur at or below the odor threshold. Furthermore, the available data indicate that there is there an almost nonexistent margin between exposures resulting in no response and those causing lethality. Therefore, AEGL-1 values for dimethylhydrazine are not recommended (NR) Absence of an AEGL-1 does not imply that exposure below the AEGL-2 is without adverse effects "Refer to AEGL-1 for hydrazine if hydrazme is also present. References Weeks, M.H., Maxey, G.C., Sicks, Greene, E.A. 1963. Vapor toxicity of UDMH in rats and dogs from short exposures. American Industrial Hygiene Association Journal 24: 137-143. Sop08-02 wpd Printed July 6, 2000 F3 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 i APPENDIX G. EXAMPLE OF THE DERIVATION OF 2 AEGL VALUES APPENDIX IN A TECHNICAL 3 SUPPORT DOCUMENT 4 5 6 7 DERIVATION OF AEGL-1 VALUES 8 9 10 Key study None. An AEGL-1 was not recommended due to inadequate data for developing 11 health-based criteria and because exposure-response relationships suggest little 12 margin between exposures resulting in no observable adverse effects and those 13 producing significant toxicity. The absence of an AEGL-l does not imply that 14 exposure below the AEGL-2 is without adverse effects. In situations where 15 hydrazine may also be present, the AEGL-1 values (0.1 ppm for all exposure 16 periods) for hydrazine should be used. 17 18 Sop08-02 wpd Printed July 6, 2000 G 1 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 DERIVATION OF AEGL-2 VALUES Key study: Toxicity endpoint: Uncertainty factors: Calculations: Time scaling: 30-min AEGL-2 1-hr AEGL-2 4-hr AEGL-2 8-hr AEGL-2 Weeks etal., 1963 Dogs exposed to 360 ppm 1,1-dimethylhydrazine for 15 minutes exhibited behavioral changes and muscle fasciculations An uncertainty factor of 3 for interspecies variability was applied because the toxic response to dimethylhydrazine was similar across the species tested. This was especially true for lethality responses (LC50 values for varying time periods ranging from 5 minutes to 4 hours) among rats, mice, dogs, and hamsters. A comparison of LCSO values for the same exposure durations in these species did not vary more than 3-fold. An uncertainty factor of 10 was retained for intraspecies variability (protection of sensitive populations). A broad spectrum of effects were seen which included behavioral effects, hyperactivity, fasciculations, tremors, convulsions, and vomiting. The mechanism of toxiciry is uncertain and sensitivity among individuals regarding these effects may vary. Following identical exposures, the responses of the dogs varied from one of extreme seventy (vomiting, tremors, convulsions, and death) to no observable effects. A factor of 10 was also retained because experiments by Weeks et al. (1963) indicated that dogs that had been previously stressed (auditory stimuli) were more sensitive to the adverse effects of dimethylhydrazine. 12 ppm 360ppm/30 = 12 C1 x t = k 12 ppm x 15 mm = 180 ppm-min C1 \t = Jt(tenBerge, 1986) (12 ppm)1 x 15 min =180 ppm-min LC50 data were available for 5, 15, 30, 60, and 240-minute exposures in rats and 5, 15, and 60 minutes for the dog. Exposure-response data indicated a near linear concentration-response relationship (n=0 84 for rats, n=0.80 for dogs) For time- scaling, a linear relationship was assumed and a value of n=l was selected C1 x 30 min = 180 ppm-min C = 6 ppm C1 x 60 min = 180 ppm-min C = 3 ppm C1 x 240 min = 180 ppm-min C = 0.75 ppm C1 x 480 min =180 ppm-min C = 0.38 ppm Sop08-02 wpd Printed July 6, 2000 G2 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 DERIVATION OF AEGL-3 VALUES Key study Toxicity endpomt Uncertainty factors: Calculations: Time scaling. 30-mm AEGL-2 l-hrAEGL-2 4-hr AEGL-2 8-hr AEGL-2 Weeks etal., 1963 1 -hr LCjo of 98 1 ppm in dogs reduced by a factor of three to 327 ppm as an estimate of a lethality threshold. Weeks et al. (1963) provided data showing that 15-minute exposure of dogs to 36-400 ppm produced only minor, reversible effects (behavioral changes and mild muscle fasciculations) An uncertainty factor of 3 for mterspecies variability was applied because the toxic response to dimethylhydrazme was similar across the species tested. This was especially true for lethality responses (LC50 values for varying time penods ranging from 5 minutes to 4 hours) among rats, mice, dogs, and hamsters. A companson of LCjo values for the same exposure durations in these species did not vary more than 3-fold. An uncertainty factor of 10 was retained for intraspecies variability (protection of sensitive populations). A broad spectrum of effects were seen which included behavioral effects, hyperactivity, fasciculations, tremors, convulsions, and vomiting. The mechanism of toxicity is uncertain and sensitivity among individuals regarding these effects may vary. Following identical exposures, the responses of the dogs vaned from one of extreme seventy (vomiting, tremors, convulsions, and death) to no observable effects. A factor of 10 was also retained because experiments by Weeks et al. (1963) indicated that dogs that had been previously stressed (auditory stimuli) were more sensitive to the adverse effects of dimethylhydrazme. 327ppm/30 = 10.9 ppm C'xt = k 1 1 .9 ppm x 60 mm = 654 ppm-min C'x/ = it(tenBerge, 1986) 1 1 .9 ppm' x 60 min = 654 ppm-min LCj0 data were available for 5, 15, 30, 60, and 240-mmute exposures in rats and 5, 15, and 60 minutes for the dog. Exposure-response data indicated a near linear concentration-response relationship (n=0.84 for rats, n=0 80 for dogs). For time- scaling, a linear relationship was assumed and a value of n=l was selected. C1 x 30 mm = 654 ppm-min C = 22 ppm C1 x 60 mm = 654 ppm-min C = 11 ppm C1 x 240 mm = 654 ppm-min C = 2.7 ppm C1 x 480 mm = 654 ppm-min C = 1 .4 ppm Sop08-02 wpd Printed July 6. 2000 G3 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 i APPENDIX H. EXAMPLE OF A TIME SCALING 2 CALCULATIONS APPENDIX IN A TECHNICAL 3 SUPPORT DOCUMENT 4 5 6 7 APPENDIX B 8 9 TIME SCALING CALCULATIONS FOR 10 DIMETHYLHYDRAZINE AEGLS 11 Sop08-02 wpd Printed July 6. 2000 H 1 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 The relationship between dose and time for any given chemical is a function of the 2 physical and chemical properties of the substance and the unique lexicological and 3 pharmacological properties of the individual substance. Historically, the relationship according 4 to Haber (1924), commonly called Haber's Law (NRC, 1993a) or Haber's Rule (i.e., C x t = k, 5 where C = exposure concentration, t = exposure duration, and k = a constant) has been used to 6 relate exposure concentration and duration to effect (Rinehart and Hatch, 1964). This concept 7 states that exposure concentration and exposure duration may be reciprocally adjusted to 8 maintain a cumulative exposure constant (k) and that this cumulative exposure constant will 9 always reflect a specific quantitative and qualitative response. This inverse relationship of 10 concentration and time may be valid when the toxic response to a chemical is equally dependent 11 upon the concentration and the exposure duration. However, an assessment by ten Berge et al. 12 (1986) of LC50 data for certain chemicals revealed chemical-specific relationships between 13 exposure concentration and exposure duration that were often exponential. This relationship can 14 be expressed by the equation C" x t = k, where n represents a chemical specific, and even a toxic 15 endpoint specific, exponent. The relationship described by this equation is basically the form of 16 a linear regression analysis of the log-log transformation of a plot of C vs t. Ten Berge et al. 17 (1986) examined the airborne concentration (C) and short-term exposure duration (t) relationship 18 relative to death for approximately 20 chemicals and found that the empirically derived value of 19 n ranged from 0.8 to 3.5 among this group of chemicals. Hence, these workers showed that the 20 value of the exponent («) in the equation C" x t = k quantitatively defines the relationship 21 between exposure concentration and exposure duration for a given chemical and for a specific 22 health effect endpoint. Haber's Rule is the special case where n = 1. As the value of n increases, 23 the plot of concentration vs time yields a progressive decrease in the slope of the curve. 24 25 Two data sets of LC50 values for different time periods of exposure were analyzed using a 26 linear regression analysis of the log-log transformation of a plot of C vs t to derive values of n for 27 dimethylhydrazine. 28 29 SopOS-02 wpd Printed July 6. 2000 H 2 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30,2000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Dog data from Weeks et al.,1963 The LC50 values for 5, 15, respectively. Log Time Cone. Time 5 22300 0.6990 15 3580 1.1761 60 981 1.7782 n= 0.8 Calculated LC50 values: Minutes Cone. 30 2036.15 60 860.12 240 153.48 480 64.83 and 60-minute exposures were 22,300, 3580, and 981 ppm, Log Cone. 4.3483 3.5539 2.9917 at 13 Sop08-02 wpd Printed July 6. 2000 H3 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 Rat data from Weeks et al., 1963 For the 5, 15, 30, 60, and 240-minute exposure periods, LC50 values of 24,500, 8,230, 4,010, 1,410, and 252 ppm were reported by the study authors. Time Cone. 5 15 30 60 240 24500 8230 4010 1410 252 Log Time 0.6990 1.1761 1.4771 1.7782 2.3802 Log Cone. 4.3892 3.9154 3.6031 3.1492 2.4014 n = 0.84 Calculated LC50 values: Minutes 30 60 240 480 Cone. 3323.28 1449.93 276.00 120.42 45 Sop08-02 wpd Printed July 6, 2000 H4 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 i APPENDIX I. EXAMPLE OF A CARCINOGENICITY 2 ASSESSMENT APPENDIX IN A TECHNICAL SUPPORT 3 DOCUMENT 4 5 CARCINOGENICITY ASSESSMENT OF DIMETHYLHYDRAZINE 6 7 Slope factors for 1,1 -dimethylhydrazine and 1,2-dimethylhydrazme were available but have been 8 withdrawn from the U.S. EPA Integrated Risk Information System (IRISJ. For a preliminary 9 carcinogenicity assessment, the withdrawn inhalation slope factor for 1,1-dimethylhydrazine 10 (cited in ATSDR, 1994) will be used. The assessment follows previously described 11 methodologies (NRC, 1985; Henderson, 1992). 12 13 The withdrawn slope factor for 1,1-dimethylhydrazine was 3.5 (mg/kg-day)"1 which, based upon 14 a human inhalation rate of 20 mVday and a body weight of 70 Kg, is equivalent to 1 (mg/m3)"'. 15 16 To convert to a level of monomethylhydrazine that would cause a theoretical excess cancer risk 17 oflO-": 18 19 Risk of 1 x 10"4 = (1 x lO-4/!) x 1 mg/m3 = 1 x 10"4 mg/m3 20 (virtually safe dose) 21 22 To convert a 70-year exposure to a 24-hour exposure: 23 24 24-hr exposure = d x 25,600 25 = (1 x 10^ mg/m3) x 25,600 days 26 = 2.56 mg/m3 27 28 To account for uncertainty regarding the variability in the stage of the cancer process at which 29 monomethylhydrazine or its metabolites may act, a multistage factor of 6 is applied (Crump and 30 Howe, 1984): 31 32 (2.56mg/m3)/6 = 0.43 mg/m3 (0.18 ppm) 33 34 Therefore, based upon the potential carcinogenicity of monomethylhydrazine, an acceptable 24- 35 hr exposure would be 0.9 mg/m3 (0.5 ppm). 36 37 If the exposure is limited to a fraction (f) of a 24-hr period, the fractional exposure becomes 1/f x 38 24 hrs (NRC, 1985). 39 40 24-hr exposure = 0.43 mg/m3 (0.18 ppm) 41 8-hr = 1.3 mg/m3 (0.5 ppm) SopO&-02 wpd Printed July 6.2000 I 1 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30,2000 1 4-hr = 2.6mg/m3 (l.lppm) 2 1-hr 10.3mg/m3(4.2ppm) 3 0.5 hr = 20.6 mg/m3 (8.5 ppm) 4 5 Because the AEGL-2 values based upon acute toxicity were equivalent to or lower than the 1O*4 6 risk values derived based on potential carcinogenicity, the acute toxicity data were used for the 7 AEGLs for dimethylhydrazine. For 10's and 10"6 risk levels, the 10"* values are reduced by 10- 8 fold or 100-fold, respectively. Sop08-02 wpd Printed July 6.2000 I 2 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 APPENDIX J. EXAMPLE OF THE DERIVATION SUMMARY APPENDIX IN A TECHNICAL SUPPORT DOCUMENT DERIVATION SUMMARY (CAS NO. 57-14-7; 1,1-DIMETHYLHYDRAZINE) (CAS NO. 540-73-8; 1,2-DIMETHYLHYDRAZINE) AEGL-1 VALUES 30 minutes Not recommended 1 hour Not recommended 4 hours Not recommended 8 hours Not recommended Reference: Not applicable. Test Species/Strain/Number: Not applicable Exposure Route/Concentrations/Durations- Not applicable Effects: Not applicable Endpomt/Concentration/Rationale: Not applicable Uncertainty Factors/Rationale: Not applicable Modifying Factor: Not applicable Animal to Human Dosimetric Adjustment: Not applicable Time Scaling: Not applicable Data Adequacy: Analysis of dimethylhydrazine toxicity data in total revealed that significant toxicity may occur at or below the odor threshold. Furthermore, the available data indicate that there is there an almost nonexistent margin between exposures resulting in no response and those causing lethality. Therefore, AEGL-1 values for dimethylhydrazine are not recommended (NR). Absence of an AEGL-1 does not imply that exposure below the AEGL- 2 is without adverse effects. NOTE: If an AEGL-1 value is not recommended, there should be a short discussion of the rationale for that choice. The rationale should include as appropriate a discussion that numeric values for AEGL-1 are not recommended because (1) relevant data are lacking, (2) the margin of safety between the derived AEGL-1 and AEGL-2 values is inadequate, or (3) the derived AEGL- 1 is greater than the AEGL-2. Absence of an AEGL-1 does not imply that exposure below the Sop08-02 wpd Printed July 6, 2000 J 1 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30. 2000 1 AEGL-2 is without adverse effects. 19 Sop08-02 wpd Printed July 6, 2000 J * ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2 3 4 5 DERIVATION SUMMARY (CAS NO. 57-14-7; 1,1-DIMETHYLHYDRAZINE) (CAS NO. 540-73-8; 1,2-DIMETHYLHYDRAZINE) 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 AEGL-2 VALUES 30 minutes 6.0 ppm 1 hour 3.0 ppm 4 hours 0.75 ppm 8 hours 0.38 ppm Reference: Weeks, M.H., G.C. Maxey, M.E. Sicks, E.A. Greene. 1963. Vapor toxicity on UDMH in rats and dogs from short exposures. Am. Ind. Hyg. Assoc. J. 24: 137- 143. Test Species/Strain/Sex/Number: mongrel dogs, 2-4/group, sex not specified Exposure Route/Concentrations/Durations: Inhalation; 1,200-4,230 ppm for 5 minutes; 360,400 or 1,530 ppm for 15 minutes; 80-250 ppm for 60 minutes Effects: Exposure (15 min) Effect 360 ppm muscle fasciculations in 1 of 4 dogs (determinant for AEGL-2) 400 ppm behavioral changes in 2 of 4 dogs 1,530 ppm tremors, convulsions, vomiting in 2 of 2 dogs Endpoint/Concentration/Rationale: 15-min exposure to 360 ppm considered a threshold for potentially irreversible effects or effects that would impair escape. At this exposure, muscle fasciculations were observed in 1 of 4 exposed dogs and at 400 ppm behavioral changes were observed. Sop03-02 wpd Printed July 6, 2000 J3 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Uncertainty Factors/Rationale: Total uncertainty factor: 30 Interspecies: 3 - The toxic response to dimethylhydrazine (LCSO values) was similar across species. The 4-hr LC50 values for mouse, rat, and hamster differ by a factor of approximately 2 and were consistent with the dog data when extrapolated from 1 hr using n=l. The more sensitive species, the dog, was used to derive the AEGL-2 values. Intraspecies: 10 - A broad spectrum of effects were seen which included behavioral effects, hyperactivity, fasciculations, tremors, convulsions, and vomiting. The mechanism of toxicity is uncertain and sensitivity among individuals regarding these effects may vary. This variability was especially demonstrated in dogs wherein responses varied from one of extreme severity (vomiting, tremors, convulsions, and death) to no observable effects. Therefore, a factor of 10 was retained. A factor of 10 was also retained because experiments by Weeks et al. (1963) indicated that dogs that had been previously stressed (auditory stimuli) which may have affected their response to dimethylhydrazine. Based upon these data, it was assumed that humans may be equally divergent in their response to dimethylhydrazine. Modifying Factor: None Animal to Human Dosimetric Adjustment: None applied, insufficient data Time Scaling: C" x t = k where n = 1 and k = 180 ppm-min; LC50 data were available for 5, 15, 30, 60, and 240-minute exposures in rats and 5,15, and 60 minutes for the dog. Exposure-response data indicated a near linear concentration-response relationship (n=0.84 for rats, n=0.80 for dogs). For time-scaling, a linear relationship was assumed and a value where n=l was selected. Data Adequacy: Information regarding the human experience for acute inhalation exposure to dimethylhydrazine are limited to qualitatively case reports indicating nasal and respiratory tract irritation, breathing difficulties, and nausea.. Data in animals have shown concentration- dependent effects ranging from respiratory tract irritation, pulmonary edema and neurological effects to lethality. Because the nonlethal effects in humans and annuals are qualitatively similar, the animal data were considered relevant and appropriate for development of AEGL values. The AEGL values for dimethylhydrazine reflect the steep exposure-response relationship suggested by available data. Sop08-02 wpd Printed July 6, 2000 J4 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 DERIVATION SUMMARY (CAS NO. 57-14-7; 1,1-DIMETHYLHYDRAZINE) (CAS NO. 540-73-8; 1,2-DIMETHYLHYDRAZINE) AEGL-3 30 minutes 1 hour 22 ppm 1 1 ppm VALUES 4 hours 8 hours 2.7 ppm 1 .4 ppm Reference: Weeks, M.H., G.C. Maxey, M.E. Sicks, E.A. Greene. 1963. Vapor toxicity of UDMH in rats and dogs from short exposures. Am. Ind. Hyg. Assoc. J. 24: 137-143. Test Species/Strain/Sex/Number: mongrel dogs, 3-4/group; sex not specified Exposure Route/Concentrations/Durations: Inhalation; exposure to various concentrations (80-22,300 ppm) for 5, 15, or 60 minutes Effects: 1-hr LC50 981 PPm (reduction by 1/3 was basis for AEGL-3 derivation) 15-minLC50 3,580 ppm 5-min LC50 22,300 ppm Endpoint/Concentration/Rationale: 1-hr LC50 (981 ppm) reduced by 1/3 was considered an estimate of the lethality threshold (327 ppm). Based on the available exposure-response data for this chemical (Jacobson et al., 1955) a three fold reduction in LC50 values results in exposures which would not be associated with lethality. Sop08-02 wpd Printed July 6, 2000 J5 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Uncertainty Factors/Rationale: Total uncertainty factor: 30 Interspecies: 3 -The toxic response to dimethylhydrazine (LC50 values) was similar across species. The 4-hr LC50 values for mouse, rat, and hamster differ by a factor of approximately 2 and were consistent with the dog data when extrapolated from 1 hr using n=l. The more sensitive species, the dog, was used to derive the AEGL-3 values. Intraspecies: 10 - A broad spectrum of effects were seen which included behavioral effects, hyperactivity, fasciculations, tremors, convulsions, and vomiting. The mechanism of toxicity is uncertain and sensitivity among individuals regarding these effects may vary. This variability was especially demonstrated in dogs wherein responses varied from one of extreme seventy (vomiting, tremors, convulsions, and death) to no observable effects. Therefore, a factor of 10 was used. A factor of 10-fold was also used because experiments by Weeks et al. (1963) indicated that dogs previously stressed by auditory stimuli may have a potentiated response to dimethylhydrazine. Based upon these data, it was assumed that humans may be equally divergent in their response to dimethylhydrazine subsequent to similar stresses. Modifying Factor: None Animal to Human Dosimetnc Adjustment: None applied, insufficient data Time Scaling: Cn x t = k where n = 1 and k = 654 ppm-min; LC50 data were available for 5, 15, 30,60, and 240-minute exposures in rats and 5, 15, and 60 minutes for the dog. Exposure-response data indicated a near linear concentration-response relationship (n=0.84 for rats, n=0.80 for dogs). For time-scaling, a linear relationship was assumed and a value where n=l selected by the National Advisory Committee. Data Adequacy: Information regarding the lethality of dimethylhydrazine in humans were not available. Lethality data for several animal species allowed for a defensible development of the AEGL-3 values but uncertainties remain regarding individual variability in the toxic response to dimethylhydrazines. SopOS-02 wpd Printed July 6, 2000 J6 ------- Standing Operating Procedures of the NAC/AEGL FACA Committee Version 08-02 - June 30, 2000 i APPENDIX K. LIST OF EXTANT STANDARDS AND 2 GUIDELINES IN A TECHNICAL SUPPORT 3 DOCUMENT 4 5 Section 8.2 of the Technical Support Document compares the AEGL values for a chemical 6 with other standards and guidelines previously published for exposure durations ranging from 10 7 minutes to 8 hours. A summary discussion of important comparisons should be presented in the 8 text and the values for recognized standards and guidelines, if available, should be presented on 9 the table. The statement "All currently available standards and guidelines are shown in Table ..." 10 should be included in the text to affirm completeness of the table. Only those standards or 11 guidelines with published values for a given chemical should be included in the table. In cases 12 where the exposure duration of a published standard or guideline differs from those designated 13 for AEGLs (e.g., 15 minute PEL-STEL), the value should be placed in parentheses in the column 14 of the closest AEGL exposure duration category and footnoted to indicate its true exposure 15 duration. A list of recognized standards and guidelines and the order in which they should 16 appear in the table is shown below. 17 18 List and Order of Presentation of Extant Standards and Guidelines in the TSD Table. 19 20 AEGL-1 21 AEGL-2 22 AEGL-3 23 ERPG-1 (AIHA) 24 ERPG-2 (AIHA) 25 ERPG-3 (AIHA) 26 SPEGL(NRC) 27 EEL (NRC) 28 STPL (NRC) 29 GEL (NRC) 30 EEGL (NRC) 31 SMAC(NRC) 32 PEL-STEL (OSHA) 33 PEL-TWA (OSHA) 34 IDLH (NIOSH) 35 REL-STEL (NIOSH) 36 TLV-STEL (ACGffl) 37 TLV-TWA (ACGffl) 38 MAC (THE NETHERLANDS) 39 MAK (GERMANY) 40 MAK S. (GERMANY) 41 BINSATZTOLERANZWERT (ACTION TOLERANCE LEVELS - GERMANY) 42 Sop08-02 wpd Printed July 6, 2000 K 1 ------- |