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 16     comprehensive search and review of relevant data and information from both published and unpublished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

-------