vvEPA
         United States
         Environmental Protection
         Agency
             Office of Pesticides
             and Toxic Substances
             Washington, DC 20460
Toxic Substances
         Chemical Selection
         Methods:
         An Annotated
         Bibliography

         Second Edition
         Toxics Integration
         Information Series
EPA560/TIIS-83-003
March, 1983

-------
Toxics Integration Information Series                       EPA  560/TIIS-83-003
                                                           March,  1983
                         Chemical Selection Methods:
                          An Annotated Bibliography
                                Second  Edition
                               Daniel E. Meyer
                               Verna L. Halpin
                               Project Manager
                               Regina  Origoni
                           Contract No. 68-01-6651
                              Work  Assignment  05
                              Technical Manager
                                Robert Janney
                         Office  of  Toxics  Integration
                               Project Officer
                              Douglas  W.  Sellers
                         Management Support Division
                          Office  of  Toxic  Substances
                  Office of Pesticides and Toxic Substances
                    U. S. Environmental Protection Agency
                            Washington, DC  20460

-------
                                  DISCLAIMER

This report  has  been reviewed and  approved  for publication by  the Office  of
Toxics   Integration,   Office   of   Pesticides   and   Toxic   Substances,   U.S.
Environmental Protection Agency.  Approval  does  not  signify that the  contents
necessarily  reflect  the  views  and   policies   of   the  U.S.  Environmental
Protection  Agency,  nor  does  mention  of  trade  names or  commercial  products
constitute an endorsement or  recommendation  for  use.

-------
                                   FOREWORD

The selection, ordering,  and  ranking of  chemical  substances  stands as  one  of
the most  critical,  yet  difficult,  tasks  facing those who  are  engaged  in  the
testing,  assessment,  and  regulation  of  substances as  well as  enforcement  of
those regulations.  The  Office of  Toxics  Integration has  discovered  a  number
of  resource  materials that might  serve  the  needs of  those  in other  federal
agencies  as well  as  those in  states,  industry, and  academia.   This  annotated
bibliography is intended  to  assist those  faced with  decisions  related  to  the
selection  of  chemicals  to  become"  aware  of  the  state-of-the-art  for  those
decision-making processes.  Hopefully, duplicative  and  overlapping  efforts  can
be avoided through such awareness.

We anticipate further updating of  this  bibliography.  Your  comments,  suggested
revisions,  and   additions   should   be   addressed   to:    Office   of   Toxics
Integration, TS-777,  U.S.  Environmental Protection Agency,  401 M  Street,  SW,
Washington,  DC    20460,  telephone  (202)   382-3393.   Copies  of all  materials
referenced in this bibliography are available for examination.
                                            Walter W. Kovalick, Jr.
                                            Acting Director
                                            Office of Toxics Integration

-------
                              Table of Contents

                                                                          Page

Introduction                                                                 1

Chemical Selection Methods                                                    6

       Federal Government

       — Environmental Protection Agency                                 7-19

       — National Institute of  Occupational
             Safety and Health                                              20

       — National Cancer  Institute                                      21-22

       — Consumer Product  Safety Commission                                 23

       — National Science  Foundation                                     24-27

       — Department of Transportation                                      28

       — Department of the Army                             "               29

       State Governments

       — California                                                        30

       — Michigan                                                          31

       International

       — United Nations Environmental Programme                             32

       — European Economic Community                                        33

       — International Agency for Research on Cancer                        34

       — Canada                                                            35

    -  Industry                                                          36-38

       Other (academia, etc.)                                            39-51

Author Index                                                                52
                                      11

-------
                      Acknowledgements

This bibliography was prepared by the U.S. Environmental  Protection
Agency's Office of Pesticides and Toxic Substances  (OPTS).   Special
thanks are due  to  Judy  Hushon,  who provided many system  references
for this report, and  to Ginny Shreve for  her  efforts  in lha  first
edition.   Document   identification  and   retrieval,   as  well   as
preparation  of  original  abstracts,  were  conducted  primarily   by
Tracor Jitco, Incorporated,  under  contract to  the  Office of  Toxic
Substances.   We gratefully  acknowledge  John  Gevertz  and  Elaine
Bild,  Office  of  Toxics  Integration, for  their  direction and  input
in the first edition of  this  bibliography.
                          ill

-------
          OTHER  PUBLICATIONS  IN THE TOXICS INTEGRATION POLICY SERIES:

State Administrative Models for Toxic Substances  Management  (July  1980)
  EPA-560/13-80-018: PB81-147373

State Integrated Toxics Management:   Fact  and Challenge  (July  1981)
  EPA-560/TIPS-81-001; PB81-242406

Chemical Substances Designation (December  1981)
  EPA-560/TIIS-82; PB83-130294
       OTHER PUBLICATIONS IN THE TOXICS INTEGRATION INFORMATION SERIES:

EPA Chemical Activities Status Report - 1st  Edition (June  1979)
  EPA-560/13-79-003

EPA Chemical Activities Status Report - 2nd  Edition (December  1980)
  EPA-560/13-80-040(a), PB81-176414;  EPA-13-80-040(b),  PB81-176422

EPA Chemical Activities Status Report - 3rd  Edition (June  1982)
  EPA-560/TIIS-82-002a, -002b

Chemical Substances Designation (December 1981)
  EPA-560/TIIS-82-003,   PB83-130294;   -004,   PB83-130302;  -005,  PB83-130310;
  -006; PB83-130328

Directory of Federal Coordinating Groups for Toxic  Substances  -  1st  Edition
  (June 1979) EPA-560/13-80-008; PB80-137870

Directory of Federal Coordinating Groups for Toxic  Substances  -  2nd  Edition
  (March 1980) PB80-177314

Perspectives on the Top 50 Production Volume Chemicals  (July 1980)
  EPA-560/13-80-27; PB80-221682

Federal Activites in Toxic Substances (May 1980)
  EPA-560/13-80-015  Revised May 1983; PB-157638

TSCA Status Report for Existing Chemicals Volume  1, Issue  2  (July 1980)
  EPA-560/13-80-033

TSCA Status Report for Existing Chemicals Volume  2, Issue  2  (July 1981)
  EPA-560/TIIS-81-004

Perspectives on State-EPA Grant Activities (September  1980)
  EPA-560/13-80-037; PB82-229105,

Chemical Selection Methods:  An Annotated Bibliography (November 1980)
  EPA-560/TIIS-80-001;  PB81-241481

Chemical Information Resources Handbook (January  1981)
  EPA-560/TIIS-81-002;  PB82-225657
                                      iv

-------
Toxic Substances Control Act Grants to States (July 1981)
    EPA-560/TIIS-81-003; PB81-232969

TSCA Chemicals in Commerce Inventory:   Regional and State  Perspectives
    (August 1981) EPA-560/TIIS-81-005
           For copies, contact:

           National T-echnical Information Service
           4285 Port Royal Road
           Springfield, VA  22161

-------
vi

-------
                                 INTRODUCTION

This  bibliography  is intended  to  provide individuals  interested in  chemical
selection (priority  setting,  ranking,  indexing,  sorting) with a  collection  of
relevant  materials.    Examining  such   materials   may  prevent   unnecessary
duplication  of  effort  and  aid  in  developing  new  methods  to  help  select
chemicals of concern.  The selection methods listed here were  developed  by the
SPA,  other  federal  agencies,  state  agencies,  and  contractors  for   these
agencies, as well as industrial, academic, and  public interest  organizations.

Included  with  each  entry  is  a brief abstract.   Some  of  the abstracts  were
prepared  specifically  for   this  effort;  others  accompanied   the   original
documents.  This second edition adds  15  new citations  to the  original  report;
these new citations are asterisked in the Author  Index.

Chemical  selection  has  been  and continues to be  of  interest  to many  groups
concerned  with  the  assessment  and  regulation  of  toxic  substances.   For
example, the Toxic  Substances  Control  Act (Public  Law  94469)  requires  the EPA
to  compile  an  inventory   of  chemical  substances  manufactured,  imported,  or
processed in the United  States  for  commercial purposes.    By June  1982,  over
70,000 chemicals were included  in  this Inventory.  The  EPA  is  also responsible
for determining which of  these  chemicals may require testing, which  should  be
subject  to  information-gathering  rales   and  which  might  require  regulatory
controls.   Making  these  types   of  determinations  requires  "selecting"   small
groups of substances from  larger groups.

Because of the large number of  potentially toxic substances  and  the  variety  of
actions possible under numerous  federal  and state  authorities,  the  development
and use of  systematic selection methods  has  become necessary.  Of course,  use
of  systematic   selection  methods  is  appropriate  only  at   certain  stages  in
decision  processes  affecting  chemicals.    For  example, early  in any  process,
candidates  must  be  screened   to identify  substances   clearly  requiring  some
assessment;  later,  candidates  from  the  first  group  must  be  chosen for  more
in-depth  consideration;   and,   finally,   a  few  candidates  are  selected  for
intensive analysis.  Thus, selection methods can be applied  to  candidate pools
that  are large or small and  to information ranging from limited  to  extensive.
However,  selection   schemes must  be  developed  and  employed  "operationally,"
that  is,  with  regard  to  the  requirements  of  the  particular   stage  in  ths
decision-making process  that  is being considered.

Limitations

There are also some  inherent  limitations  in  the  use of  any  chemical  selection
method.  Most  selection methods involve   consideration  of  some  or  all  of  the
following    factors:     carcinogenicity,    mutagenicity,     teratogenicity,
neurotoxicity,   and other   chronic  human health  effects;   acute   human  health


  Excludedfrom  this inventory  are  mixtures,  pesticides,  tobacco,  food,  food
additives, drugs, cosmetics,  firearms, and nuclear  materials.

-------
effects;  environmental  toxicity;   types  and  quantities  of   emissions   and
environmental exposure;  bioaccumulation;  costs of  regulation;  and  regulatory
authority.   Any  method   that  involves   combining   various  categories   of
information  to  arrive at  some single  index of  concern  risks  obscuring  the
problems created by the original data.

Existing  systematic  selection methods generally  use  some  form  of  scoring  or
ranking which involves  assigning  a dimension!ess or  arbitrary number to  each
of the  factors  being  considered.   Conversions of  this type,  while  summarizing
large amounts of  information, also tend  to  obscure the original  information,
and fail  to  incorporate a  measure  of  uncertainty.  Suppose,  for example,  that
the  number   "3"   is  assigned  to any  substance  to  which   between  10,000  and
100,000  persons  are  exposed.   Not  only  will   this  number  fail  to  convey
accurately the  number  exposed,  it  will  also fail to  provide  an indication  of
the level of uncertainty in the estimate or an indication  of  the basis for ':he
estimate (e.g.,  monitoring, modeling).

Another limitation  to the  use of  many scoring  methods  is  that  they  employ
simple  additive  or  multiplicative  algorithms  to  combine  factors.    Such
algorithms often  ignore or  distort  che  true relationship  between  f act :>r^.   IT
the case  of  an  additive  system, suppose  that  for Chemical X, carcinogenicity
is assigned  a "5",  no information  is available for six  other factors  so  each
is assigned  a  "0", and  persistence  is   assigned  a  "3".   The total  score  for
Chemical X is "8".  For Chemical Y, carcinogenicity is assigned  a  "1",  each of
six other factors  is  also  assigned a "1", and  persistence is assigned  a  "3".
Thus, the total  score for Chemical  Y is  "10".  It is  unclear, however, whether
Chemical  Y   is  really deserving  of  more  intensive  review  than  Chemical  X,
regardless of the policy framework  governing the evaluator.

Some  methods utilize  "flagging"  or  "triggering,"  as  a  means  of  selection.
"Flagging"  refers to  setting  threshold levels or  employing  "discriminators"
for  various  factors  in order to  select  candidates.   Any  chemical  may  be
"flagged"  if  it  exceeds   the  discriminators  for  certain   combinations  of
factors.  The use of "flags"  or "triggers"  in  a selection scheme avoids  the
complex data manipulations  required for  the  systems discussed above.   However,
selection methods  using  "flagging"   are only  effective  if   the  factors  are
chosen carefully.

System Comparisons

It is beyond the  scope  of  this  work to  perform a  comprehensive review  and
critique  of  past  priority  scoring  systems  developed  for  use  in  selecting
chemicals  for   study  and/or  possible  regulatory action   in  other  contexts.
However, Table  1  summarizes  the number  and  types  of variables used  in several
different systems.   In general, these  few selected  systems  were  designed  to
accommodate  a  wide  variety  of information.   Three  of  the  systems  call  for
scoring more than two dozen different variables.

-------
                                   Table 1
                    NUMBER AND TYPES  OF  FACTORS  CONSIDERED
       (Numbers represent  the  number of variables within each category.)
System


A
B
C
D
E
F
Produc-
tion or
release

1
7*****
0
2
4
i
L
Expo-
sure


2**
9***
2***
0
12
14
Biological Responses
for humans*


5
5
6
3
6
6
Other


0**
5
1
1
3
4
Total


5**
10
7
4
9
10
Multi-
media
charac-
ter
1******
0
0
0
0
0
Total
factors
scored

9
26
9
9****
25
25
*In cases where the scoring factor could apply either to humans
group of animals (e.g., "terrestrial  animals"),  it  was  counted
      or  to  a  larger
      as  applying  to
those that could  not
                                                               as  an  exposure
humans.  The only  factors  considered under "other"  were
include human data, such as plants and aquatic animals.
**Bioaccumulation,  1  of  the  2,  was  treated by  the  system
variable.
***Persistence  and  bioacruraulation  were  counted here  as  exposure  variables,
although they were treated on  par with response variables  in  the  system.
****Included   three   miscellaneous    variables    (ignitability,    reactivity,
corrosivity)  that   relate  to  neither  production,  exposure,   nor  biological
ef feet s .
*****Included  production,  consumption,  imports,  exports,  and   release   to
various specific  environmental media.
******A factor  was  included  that  gave extra  points  if  other  agencies  had
standards  for the chemical being scored.

-------
Table 2 compares some  general  properties of  the  systems  surveyed.  The  first
column  (ranking  vs.  screening  as the  overall approach) distinguishes  systems
that are intended to rank a number  of  candidates  for action from  systems  that
use  a  series  of  screens  (pass/fail   tests)  arranged   in  a decision  tree  to
select a group  of  candidate  chemicals  that  meet  a  specified set  of  criteria.
The second column  discusses  the  method for computation of  final  scores.   All
but one  system had  some  provision  for  doing this, usually based on  derived
scores rather than raw data.

The final column of  Table  2  represents an  independent  assessment  of Che  need
for research-level expertise.  Generally, the criterion for this  judgment  was
that if the system required extensive  evaluation  of raw research  results  in  a
particular field, it would usually  be  necessary to  have a Ph.D. or  equivalent
in that  field involved in some  way in  at  least checking  the  analysis.

Summary

Systematic selection methods will certainly continue to be significant:  in the
selection  of   chemicals  for  assessment  and   regulation.    Such  methods  can
provide  consistency in  decision-making.   They  also  can  promote   efficient
resource  allocation.   Finally,  use   of  these   methods   can  aid   in   the
defensibility and effectiveness of regulatory  actions.

The selection schemes presented in  this publication  range from  quite  simple to
very sophisticated and, with minor  tailoring, can be adapted to virtually any
chemical  selection  need.   This  publication  is   intended  to   provide  quick
reviews of many approaches to  chemical  selection, with  the hope that an  early
overview will  save  time and  money  for  those wishing  to  select  or  create  a
system of their own.

-------
                                   Table 2
       GENERAL PROPERTIES OF THE PROCEDURE FOR ARRIVING  AT FINAL  SCORES
System
Ranking or
screening
as overall
approach

Ranking
           Ranking
           (within
           components)
           Ranking,
           but
           screening
           at end
           Screening/
           Selection
           Ranking fol-
           lowed by a
           sort into
           three "zones'
           of overall
           hazard
Selected properties
Uses only derived scores.
Related factors grouped into
"components"; components com-
bined in various ways (e.g.,
addition, multiplication).
Component scores normalized,
weighted, and aggregated to
form final score.

Uses raw data for production/
release , but derived scores
for all pther factors.
Authors opposed combining
scores because of different
scales used .

Uses only derived scores.
Scores combined by addition.
                Uses quantitative data cut-
                off for production/release,
                but derived scores for
                biological response.
                Yes/no decision tree  used
                to select  substances;  no
                scores combined.

                Uses quantitative data ex-
                clusively.  Each  factor's
                data added or multiplied;
                some factors weighted.
Overall ex-
pertise required
                                                           Ph.D.  or  equivalent
                                                           in one specialty
                                                           (biology)
                                                Ph.D.  or
                                                equivalent  in
                                                many specialties
Ph.D. or equivalent
in many specialties
(aquatic biology,
environmental
transport,
toxicology,
carcinogenesis,
genetics)

Ph.D. or equivalent
in one specialty
(biology)
                                Ph.D. or equivalent
                                in many specialties
           Ranking
           (within
           components)
                Uses only derived scores.
                Addition of factor scores
                into four "component"  scores;
                these scores normalized.   No
                further aggregation of com-
                ponent scores,  so no final
                score for each  substance.
                                Ph.D. or equivalent
                                in many specialties

-------
Chemical Selection Methods

-------
Allport, J.;  Casey, S.;  Cook,  J.;  at al.
Stanford Research Institute, Menlo Park,  California
A Study of Industrial Data on Candidate Chemicals  for Testing—Final  Report
Prepared for U.S. Environmental Protection Agency
Office of Toxic Substances
EPA-560/5-77/006, Research Request No. 1
NTIS //PB-274 264/1GA, 1977


This report summarizes  the  work  done by Stanford  Research Institute  for  EPA's
Office  of  Toxic  Substances  and   includes  three  major   parts.    (1)  Data
previously collected on a NSF  study  were  supplemented  to  provide  the following
information,  where available,  on  667 industrial  chemicals:   (a) United  States
production; (b)  estimates  of  quantities  released  annually to  the  environment;
(c) major  uses;  and (d)  references  on mutagenicity tests.  Mutagenicity dac-a
on 25 chemicals  were evaluated after developing a list of assays  and criteria
for classifying  the results  as  either positive  or negative/inadequate.   (2)
Tables which  contain economic information on  1,791  chemicals  belonging  to  26
structural classes  considered to  represent   potential  industrial  carcinogens
and  mutagens   were  prepared.   For   those  chemicals  with  annual  production
greater  than   one  million  pounds,  market  forecasts  which  present  a  brief
summary of production, consumption  patterns, major uses,  possible  substitutes,
and growth trends  were  prepare'd.   (3) Carcinogenicity  data  for all  chemicals
belonging  to  three classes (epoxides, alkyl  halides,  and vinyl halides) were
used  to  correlate  structural  features with  carcinogenic activity.   Criteria
were  developed  for estimating the  potential carcinogenicity  of   chemicals  in
each  class.   These  criteria  were  applied  to those  chemicals in each  class
known  to  be   produced  commercially  or  for  which  there  was   evidence   of
significant human exposure.  The results of the study were summarized in  three
separat e reports .

-------
Ross, R.H.
Oak Ridge National Laboratory
Welch, J.
U.S. Environmental Protection Agency, Office of  Toxic Substances
Proceedings of the EPA Workshop on the Environmental Scoring of Chemicals
Washington, D.C., August 13-15, 1979
EPA-560/11-80-010, May 1980


The  environmental  scoring of  chemicals is  viewed  by  the EPA  as  a  tool  to
assist in the ranking or ordering of  the  universe  of chemicals that are under
the jurisdiction of the Toxic  Substances  Control  Act.  The purpose  of  scoring
is  to  identify  most  of  the  chemicals   that  have  a  high  probability  for
requiring  review  for regulation  or  testing.   This  report  describes  a  3-day
workshop  held  in  Washington,  B.C.,  August  13-15,   1979,  to   develop  an
environmental   scoring    system.    Initial   discussions   centered   on   the
determination of a safety  factor  (calculated  as the  concentration at which  an
effect is  observed divided by environmental concentration) that  would  allow a
numerical score to be assigned  to a chemical  to reflect its  potential  hazard.
Further discussion, however, indicated  that the environmental  concentration  of
a  chemical   is   usually   not   available  and  that  the   estimation   of   an
environmental concentration  is  not  readily accomplished;  therefore, a  scoring
system was developed that  does  not  require  environmental concentrations.   This
system relates  environmental exposure  to  toxicity  by using a multiplier  (3x,
2x, or  Ix)  which is assigned  on  the  basis  of  the  concentration  at which  an
effect is  observed.   The   applicability of  the  scoring system is  demonstrated
by scoring selected chemicals.

-------
Flinn, J.E.;  Thomas, T.J.;  Bishop,  M.D.
Battelle Columbus Laboratories
Columbus,  Ohio
Identification  Systems  for   Selecting   Chemicals   or  Chemical  Classes   as
Candidates for Evaluation
Prepared for U.S. Environmental Protection Agency
Office of  Toxic Substances
EPA-560/1-74/001
NTIS //PB-233 196/OGA, November 1974


This  report  is  a  review  of  various  systems  for  selecting,  assessing,  and
ranking chemicals for their health and environmental effects.   Systems  for  two
general  areas  are  considered:    health  planning,   including   environmental,
occupational, and  general  health;  and environmental management.   It is  noted
that  the  term  "system"  is  broadly defined  to  include  not  only  formalized
organizational  structures,  models,  and  methodologies, but  also  less   fornal
tools, methods,  and working  groups  which  have  been created  to rank chemical
substances or effects.  Existing systems are classified in several ways.  Some
are  considered  information   repositories  for   chemical  data,   particularly
toxicity  data.   Other systems  are  classified  with  respect  to whether  their
function  is  to  identify  chemical  substances  before  or after general exposure
of  the  public  and  the  environment  occurs.   Each  of  these  types  is  further
categorized into those which  identify new  or  unrecognized chemical  "stressors"
and  those which  evaluate  recognized  "stressors."   Particular  attention   is
given to methods used by various federal  agencies.   It is coir-laded  that most
existing  systems  are  deficient  in  that  they  focus  on  acute  rather  than
long-term  effects;  have  a  limited  domain of  concern;  and  are aot designed  to
identify hazards from degradation  products, synergistic effects, or  effects  on
the  nonliving  environment.   An  appendix  summarizes  results  of  a  seminar   on
"Early Warning  Systems  for Toxic  Substances"  held  January  30  -  February   1,
1974.

-------
Venezian, E.G.
Arthur D. Little, Inc.
Cambridge, Massachusetts
Pre-screening  for  Environmental  Hazards  -   A  System  for  Selecting   and
Prioritizing Chemicals
Prepared for U.S. Environmental Protection Agency
Office of Toxic Substances
Washington, D.C.
EPA/560/1-77/002, April 1977


Alternatives  were  considered  for  prescreening  chemicals   to  assess   their
potential  environmental  hazard.   A  system  design  concept  which  takes  into
account both the  toxicity of the chemical and the eventual levels which  it  can
be expected to reach in the environment was selected for further  analysis.   li
is noted that, although neither  toxicity  nor  eventual  levels can be  predicted
with  great  accuracy,  the  accuracy  attainable  by  simple  methods  appears
adequate for  selecting  and  ranking chemicals  for additional  investigation.   A
specific design which relies on data which are  usually available  was  developed
to the  point  of  testing the feasibility of collecting  the necessary data  and
performing the required computations  on five chemicals.
                                      10

-------
TSCA  Interagency Testing  Committee:    Initial  Report  to  the  Administrator
Environmental Protection Agency
42 FR 197:55026-80,  October 1977
Section  4(e)  of  the  Toxic   Substances  Control  Act   (TSCA)   requires   the
establishment  of   a  committee   to   identify  and   recommend   to   the   EPA
Administrator chemical  substances  which should  be  tested  to determine  their
hazard to human  health or  the  environment.   This report documents  procedures
used  by  the TSCA  Interagency  Testing Committee  for  selecting  those  chemical
substances  recommended for testing.   Reasons  for  each  recommendation  are
outlined.   Available  data  and  potential  data  for  carcinogenic,  mutagenic,
teratogenic, and chronic toxic  effects were all considered,  as was  the ability
of  the  substances   to   bioaccumulate  or  cause  deleterious   environmental
effects.   A  scoring  system  which took into account both  available  information
and  the  lack  of  it  for   these factors was  used  in the  screening  process.
Categories  or   substances  recommended  for  further  testing   include   alkyl
paraffins,   chloromethane,  cresols,   hexachloro-1,3-butadiene,   nitrobenzene,
toluene,  and xylenes.  The  formation of  the Interagency Testing  Committee,  its
responsibilities,  and  the  approach  used  in  forming   the recommendations  are
discussed.   Sources of data used in  the  preparation of the  initial  substances
list  are given.   A  linear weighting  scheme  used  to rank  the  substances  is
discussed.
                                      11

-------
Enviro Control Inc.
Rockville, Maryland
Scoring Chemicals for Health and Ecological  Effects  Testing
Proceedings of TSCA-ITC Chemical Scoring System Workshop
February 25-28, 1979


Since  its  inception   in   early  1977,  the  Toxic  Substances  Control   Act
Interagency Testing  Committee  has  submitted  to the  EPA  Administrator  four
reports  recommending  a total  of  33 chemicals  and  chemical  categories  to  be
tested for  their  potential  to cause unreasonable risk to human health or  the
environment.    In  order   to  identify  these   chemicals    from   among   the
approximately  44,000  chemicals  in  commercial  production in  the United  States
today,  it  was  essential  for  the   Committee  to  have  procedures  for  ranking
chemicals as to the need for testing to determine their hazard  to human  health
and the environment.

The Committee  used existing  source lists  of  hazardous  chemicals   to  form  a
Master   File,   as  described  in   the  Committee's   Initial  Report  to   the
Administrator.  This  list  of 1,700 chemicals  was  subjected to scoring,  first
on the basis of  production,  release, operation,  and exposure,  and subsequently
on the basis  of biological  activity.    The  scoring  procedure was developed  in
conjunction with  Clement Associates, Inc.,  the Committee's   support  contractor
during its first  2 years of  operation,  and is described in this report.

In  an effort  to  make any  possible   improvement  in  this   numerical  scoring
system,  the Committee  decided to subject  the  scoring system  to the  scrutiny
and criticism  of  individuals  from  academia, industry,  and government who  were
expert in  the  various  aspects of release, exposure, and effects  that  comprise
the system.  Enviro Control, Inc. was  awarded  a  contract to  organize  a  scoring
workshop and to prepare a report on the proceedings.  The workshop was  held in
San  Antonio,   Texas,  February  25-28,   1979,  with approximately  80  technical
experts  participating.
                                      12

-------
Fuller, B.;  Hushon,  J.;  Kornreich,  M.;  et  al .
Mitre Corporation, McLean, Virginia
Preliminary Scoring of Selected Organic Air Pollutants,  5  Volumes
Prepared for U.S.  Environmental Protection Agency
Office of Air Quality Planning and  Standards
Research Triangle  Park,  North Carolina
EPA-450/3-77/008a,b,c,d,e
NTIS #PB-264 442/5ST, October 1976


This  report  presents a  scheme  for evaluating  the  relative  hazard  to humans
resulting   from   air  emissions   during   production   of  synthetic  organic
chemicals.  Data  on  production, fraction  lost  during  production, volatility,
and toxicity have been  compiled for 637  organic  chemicals.   A scoring system
using  these  data  elements  was  developed and  is  described  in  this repot" .
Scores  assigned  to  the   637  chemicals are  presented.    Four  appendices  were
published with this  report.  The appendices are dossiers  containing  chemistry,
production,  and toxicity data for the 637  synthetic  organic  chemicals.
                                      13

-------
Fiksel, J.; Segal,  M.
Arthur D. Little, Inc.
An Approach  to  Prioritization of Environmental Pollutants:   T'ns Action  Alert
System — Final Draft
Prepared for U.S. Environmental Protection Agency
Office of Water Regulations and Standards
Monitoring ind Data Support Division
Washington, D .C .
Contract No. 68-01-3857, June 1980
The  Action  Alert  System  was  developed  to  assist  the  Office  of   Water
Regulations and  Standards in  evaluating available  data,  setting  priorities,
and  determining  appropriate  actions.   The  system  can  aid in  sorting  large
groups of chemicals into "manageable clusters"  for  further  examination.

The  two  required  data elements  are  (1)  concentrations,  including  drinking
water, human diet, and  ambient  water;  and  (2)  effects data, including  chronic
mammalian,  acute  human or mammalian,  and  aquatic   toxicity.   Six auxiliary
modules  have  been developed  to  serve  as  surrogates  in  cases where  data  are
lacking or to enable use of more extensive data where they  exist.   This  report
includes a detailed description of  the conceptual  development of  the system,  a
presentation of  the  six  auxiliary  modules,  and a user's  guide.   The  user's
guide  provides detailed  instructions  for the  application  of" the Action  Alert
System to the ranking of specific chemicals.
                                      14

-------
Dorsey, J.A.; Johnson, L.D.;  Statnick,  R.M.;  et al .
Environmental  Assessment   Sampling   and   Analysis:    Phased   Approach   and
Techniques for Level 1
Prepared for U.S Environmental Protection Agency
Industrial Environmental Research Laboratory
Research Triangle Park, North Carolina
EPA-600/2-77/115
NTIS //PB-268 563/4ST, June 1977
The  report  discusses  a  three-level  approach  to  sampling and  analysis  for
environmental source assessment.  A research  program was initiated  to  develop
a  sampling  and  analytical  approach  for   conducting   environmental   source
assessments of the feed, product, and waste streams  associated  with  industrial
and energy processes.  An environmental source assessment identifies  potential
air,  water,   and  terrestrial   problems  for  both  regulated  and  unregulated
pollutants.  The three-level sampling and analysis approach resulted  from  this
program.  Level  1  is  a  complete survey  of  all  streams,  using  simplified,
generalized  sampling  and  analytical  methods  which  permit  priority  ranking;
i.e.,  hazardous  streams  are   distinguished   from  those  less  hazardous   or
relatively innocuous in  nature.   Level  2  is detailed sampling  and analysis  of
the  streams  ranked   in   the  highest  priority  by  the  Level  1  survey.  Other
streams may then be  addressed according to  potential  hazard.  Level  3 involves
continuous monitoring of "key"  indicator materials  to  evaluate the  effect  on
emissions of  process variability.
                                      15

-------
Archer, S.R.;  McCurley, W.R.;  Rawlings,  G.D.
Monsanto Research Corporation, Dayton,  Ohio
Source  Assessment:    Pesticide   Manufacturing   Air  Emissions—Overview   and
Prioritization
Prepared for U.S. Environmental Protection Agency
Industrial Environmental Research Lab
Research Triangle Park, North Carolina
EPA-600/2-78/004d
NTIS //PB-279 171/3ST, March 1978


This  report is  an overview of  the  pesticide  manufacturing industry  and  ranks
80 major pesticides  based  on  their  potential  environmental burden  from an  air
pollution  standpoint.   Production  of  synthetic organic  pesticides  was  about
640,000 metric  tons  in  1974.   Thirty-seven major synthetic  organic  pesticides,
those with  annual  production  of 4,540 or more  tons,  accounted for 742 of  the
market.   Elemental  chlorine   is  common  to   most   pesticides,  but  other  raw
materials  include hydrogen  cyanide,  carbon  disulfide,  phosgene,   phosphorus
pentasulfide,    hexachlorocyclopentadiene,  various  amines,   and   concentrate!
acids  and  caustics.   Air  pollution aspects  of  the  pesticide  manufacturing
industry  are  essentially  without  quantitative  data.   For  some  plants,  the
pollution caused  by  loss of  a-ctive  ingredients  is less significant  than that
caused  by   unreacted   by-products.   Evaporation   from  holding  ponds   and
evaporation lagoons  may  also  be an  emission  source,  although few quantitative
data  are  available.   Emissions  emanate  from  various  pieces  of  equipment  and
enter  the  atmosphere  as   both  active  ingredients  and  as  raw  materials,
intermediates,   and   by-products.    Air  emission   control  devices   include
baghouses, cyclone  separators,  electrostatic  precipitators,  incinerators,  and
gas  scrubbers.   Synthetic  organic  pesticide  production  in  1985 will  be  about
806,000 metric  tons.
                                      16

-------
Brown, S.L.; Holt, B.R.;  McCaleb, K.E.
Stanford Research Institute, Menlo Park,  California
Systems for Rapid  Ranking  of Environmental Pollutants:  Selection  of  Subjects
for Scientific and Technical Assessment Reports
Prepared for U.S. Environmental Protection Agency
Office of Research and Development
Office of Health and Ecological Effects
Washington, D.C.
EPA-600/5-78/012
NTIS //PB-284 338/1ST, June 1978


This document  reports the  results of the  development  and  testing of  a  system
for  rapidly ranking  environmental  pollutants.    One   potential  use  for  the
system  is  in   choosing   the  most   important  candidates  for  Scientific  and
Technical  Assessment  Reports  (STAR).    Of  several  possible  approaches  to
ranking  environmental  agents,  a  system  depending  on  expert  opinion  but
assisted by an objective subsystem  was  selected for development.   The  system
defines  procedures   for   collecting,   processing,   and  evaluating   data  on
production  and use;  environmental   transport,  transformation, and  rate;  and
human  health   welfare  and  ecological  effects.   A  test  of  the  objective
subsystem confirmed  the  utility  of   the  system.   Of ten candidate  agents,  the
three highest  ranked were cyanides,  carbon disulfide, and beryllium.
                                      17

-------
Stacey, G.S.;  Flinn, J.E.
Battelle Columbus Labs, Ohio
Development of an  Economics-Based  Methodology for Projecting  Future  Pollution
Problems
Prepared for U.S. Environmental Protection Agency
Office of Research and Development
Office of Health and Ecological Effects
Washington, D.C.
EPA-600/5-78/011
NTIS //PB-284 337/3ST, June 1978


The  report  describes  a   project  designed   to   develop   a   methodology  for
identifying potential  future  toxic  substance  pollution problems.  An  approach
was  desired that  would  be  systematic,  comprehensive,  and  futuristic.   The
methodology developed is  based on exposure and  initiates  the  identification of
problems by focusing  on  the potential  for their occurrence in  the  production,
exchange,  and  consumption  of  goods  and   services.   Products  are  ranked
according  to  the  potential  they  have   for  being  associated  with  future
pollution  problems.   For  the high-ranked  products,  additional  information  on
the chemical constituents of the product are  identified.  The  final  step is to
analyze  the   chemical  constituents    to  determine  which   chemicals   occur
frequently and in  large quantities.  At the  same time the  potential  that each
of  the  chemicals  has  for  resulting   in  toxic substance  problems  would  be
assessed.   In  ranking the  products,  parameters concerning historical  growth,
future  growth,   dispersion,  technical  change,  and   value  of  shipments  were
developed and applied.  A  specific group of products was  examined  to determine
their  chemical  content.   The  results  of   this  effort showed  that  identifying
chemical constituents  of  products  required considerable resources.  The final
step of  analyzing  chemicals to determine  frequency  and quantity was  developed
conceptually,  but, due to resource limitations, could not  be applied.
                                      18

-------
Cleland, J.G.; Kingsbury, G.L.; Sims, R.C.; White, J.B.
Research Triangle Institute
Research Triangle Park, North Carolina
Multimedia Environmental Goals for Environmental Assessment.  4 Volumes
Prepared for U.S. Environmental Protection Agency
Industrial Environmental Research Laboratory
Research Triangle Park, North Carolina
EPA-600/7-77/I36a, EPA600/7-77/136b, EPA600/7-79/176A, EPA600/7-79/176B
NT IS //PB-276 919/8ST, //PB-276 920/6ST, //PB80-115108, //PB80-115116, August 1979


The  report  gives   results   of   a   study   of   the  derivation  of  Multimedia
Environmental Goals  (MEG's).   MEG's are levels of  significant  contaminants  or
degradents  (in   ambient  air,  water,  or  land,  or  in  emissions   or  effluents
conveyed to  the  ambient media)  that  are judged  to be either  appropriate  for
preventing  certain   negative   effects   in  the   surrounding   populations   or
ecosystems  or   representative   of   the  control   limits   achievable   through
technology.   In  the  context of  deriving  MEG's,  Volume I:   offers perspective
on the broad range of contaminants  whose control  is vital  to both industry  a.i
-------
Pielmeier, G.R.
Tracer Jitco, Inc.
Rockville, Maryland
Identification of High Risk Occupational Groups and  Industrial  Processes  Using
RTECS/NOHS Data—Final Report
Prepared for U.S. Department of Health,  Education and Welfare
Public Health Service
Center for Disease Control
National Institute for Occupational Safety  and Health
Division of Surveillance, Hazard Evaluations and Field Studies
Cincinnati, Ohio
Contract No. 210-78-0076, November 1979
Two NIOSH data files, the Registry of Toxic Effects  of  Chemical Substances and
the National Occupational Hazard Survey, contain data that  pertain  to the risk
posed  to  workers  by  toxic  chemicals  in  the  workplace.    In  order  c'lit
occupationally and  industrially  defined groups  of workers  at  high  risk mig'ic
be  identified  and   given   their   proper  priorities  in  NIOSH  activities,
algorithms  were  developed  which  combine  data  from   the   two  data  files
identified above,  produce indexes of the potential risk to  workers  in specific
occupations  and   industries,  and  rank  the  chemicals  to  which  workers  are
exposed in terms of their toxicological hazard and their  potential  risk to all
workers.   The steps  involved  in  development of  these algorithms  are described
in  this  report.   Five  indexes  are developed:   (1) Hazard   Risk   Index,  (2)
Adjusted Hazard  Risk Index,  (3) Occupational  Risk  Index,  (4)  Industry  Risk
Index, and  (5)  Occupation  within  Industry Risk  Index.   Representative  pages
from the indexes are presented.
                                      20

-------
Stanford Research Institute
Menlo Park,  California
Criteria and Procedures for Chemical Selection
Prepared  for Chemical  Selection  Working  Group,  National  Cancer  Institute,
Bethesda, Maryland
Contract:  N01-CP-95607,  October 1977
This  report  describes   the   criteria  and  procedures  used  by  the  Chemical
Selection Working Group  to  select  chemicals for NCI's  carcinogenesis  bioassay
program.  Chemicals are nominated for consideration by  a  variety  of  government
agencies and  others.   A  two-step  process  is  then  used  for  selection.   The
first step involves gathering  limited data  to determine if  the chemical  meets
qualifying criteria.   In order  to  qualify,  a  chemical must:  not  currently be
undergoing  testing  by the NCI  bioassay  program  or  other  programs;  not  have
been  adequately  tested  previously;  and   have  high  annual   consumption  or
evidence of  exposure from  environmental occurrence.   If  the  chemical  meets
these criteria, more extensive  selection criteria are  then  applied.  This step
involves  analyses  of:    exposure  information,  including  consumption,   use
patterns,  human  exposure,  and  environmental  occurrence;  and  evidence  for
possible carcinogenic activity,  including human data, animal data,  short-term
tests, metabolism, and structural/activity  relationships.
                                      21

-------
Stanford Research Institute
Menlo Park, California
An  Automated  Procedure   for  Assessing  Possible  Carcinogenic   Activity   of
Chemicals Prior to Testing
Prepared for National Cancer Institute,  Bethesda,  Maryland
Contracts:  N01-CP-33285  NIH-NCI-71-2045,  1975


This report discusses  the systematic  application  of information  regarding  the
carcinogenicity of chemicals  to  the  prediction  of carcinogenicity activity  of
untested  chemicals.   The  method  focuses on  structural   relationships  between
the  chemicals  being  examined  and known  carcinogens.   The  development of  an
activity  tree  is  presented for implementing a procedure  for  ranking  chemicals
prior to  testing  in laboratory animals.   The activity tree method  classifies a
large number  of  chemicals  by  asking  a  series  of  increasingly more  specific
questions  about  their   chemical  structure.   Expert  consultants  then  offer
estimates  regarding  the probability  of  any chemical in  an  end  point of  the
tree being carcinogenic  and the  relative potency of the  carcinogenic  activity
if  the  chemical  proves  to  be  carcinogenic.    A  level   of  confidence   is
attributed to  each estimate.  Estimates  are  also  made   specific  to   the  four
routes  of  administration (i.e.,  oral, inhalational, dermal,  and  prenatal)  of
the  compound  to  the  test  animal,  corresponding  to the  routes  used ii   :he
exposure  estimates.   For  the "development  of  this  activity  tree, six experts
offered  their  considerations on  five classes  of  substances  known to concala
carcinogens:    (1)  naturally occurring  substances,  (2)   aliphatic   nitrogen-
containing  compounds,  (3)  polycyclic   aromatic   hydrocarbons,  (4)   aromatic
amines,  and  (5)  inorganic compounds.  Their  reports  are  summarized  and  the
current  activity  values  for  the  tree  are  given.   An  appendix   describes
computer  implementation of the activity  methodology.
                                      22

-------
Becker, D.S.
IIT Research Institute
Design of a Chemical Hazard Ranking System — Final Report
Prepared for U.S.  Consumer Product Safety Commission
Bethesda, Maryland
Contract No. CPSC-C-77-0068, December 1978
The Consumer Product  Safety  Commission would like  to rank  the  health  hazards
of  the  chemicals   in  consumer  products  to enable   them   to  allocate  their
resources  in  an  appropriate  manner.   The  three  main  factors  of  consumer
chemical  hazard  due  to  consumer  products  are:  (1)  toxicity,  (2)  dose  per
person,  and  (3)  population  exposed.   Unfortunately,  these  factors  are  not
available  as  standardized  statisitics,   nor   can   they   be   calculated   by
standardized formulas.   Population exposed  could  be  computed  from  marketing
data for  each  product,  but  this  would be  prohibitively  expensive.  Dose  per
person  should  be  computable  from  a  product's known  characteristics  (aerosol,
skin contact,  frequency  of  use,  etc.), though  accepted standardized  equations
do not  exist.  Chemical  toxic  strength factors for carcinogens,  mutagens,  and
teratogens  are known for  only  a  few percent  of   consumer  chemicals.   No
standard  method for extrapolating latent toxic  strength  factors from  data  on
the tested  chemicals  to  the untested  ones  exists.   Typically,  these  problems
have been overcome by having panels of experts  rate the three  factors  based  on
unstated  rules.   However,  the scope  of the problem  for   consumer  chemicals
favors a  computerized approach based  on systematic procedures.   First,  such a
system  could efficiently re-rank  all  chemicals  as  new data became  available.
Second,   it   is   less   expensive   to  use   automated  procedures  for   this
application.    Third,  it   is  more   precise  to make  all the decision  criteria
explicit.   This  report  presents   uniform  techniques  for   estimating  the
toxicity, dose per  person, and population exposed factors,  and for calculating
a hazard  score from them.   The procedures  are  sufficiently general that they
can be  applied,  using  available   data,  to   all consumer chemicals  within  an
automated system.
                                      23

-------
Nelson, N. ;  van Duuren, B;  Goldschmidt,  B.
Final  Report  of  NSF Workshop  Panel  to Select  Organic  Compounds Hazardous  to
the Environment
National Science Foundation, 1975
A  selection  process  was developed  to  identify  manufactured  organic  chemicals
that may be of special  interest because of  their  effects  on environment and/or
health.  This process  is  conducted by an  advisory  panel comprised  of  members
from  government,  industry,  academia,   and   public  interest   groups.    The
selection process was  divided into two phases.   Phase I  identified  and ranked
275 compounds  in descending order  of  release  rate.   The release rate  (R)  is
defined as R = (P+I)FD + PFPL

           where:  P      =  the  annual U.S.  production,
                   I      =  the  annual imported  quantity,
                   FJ-J    =  the  fraction of compound  which goes to
                            non-intermediate dispersive  uses,
           and     FpL   =  the  fraction of production lost  during
                            manufacturing, conversion, and product formulation
                            and  that which escapes from the  plant site.

Each advisory panel member  identified  10  chemicals,  regardless of the quantity
produced, which  they felt merited  inclusion on  the list.  Sixty-two  additional
chemicals were  thus  identified and added  to  the 275 chemicals  on   the  list.
Each chemical  was independently  rated on  a  scale  of  0  (no  interest)  to  5
(highest  interest)  by  each member.   The  list  was  re-ranked  using   the  mean
value of the panels'  interest and  the  top 80 chemicals were  selected  for Phase
II  evaluation.    Phase  II  evaluation consisted  of   obtaining  data  on  three
factors:  (1) production  volume,  losses  to the environment,  impurities,  etc.;
(2) toxicity, covering  mammalian,  nonmammalian, and  environmental effects;  and
(3) chemical properties, including persistence.

The  data  for  each  chemical were reviewed  by  each panel  member  and  each
chemical  was  discussed  by  the   entire  panel.   A  statement  of  the  panel
discussion was prepared for each  chemical  and  the  compounds  were again  rated
independently by  each  panel member for health and environmental  effects.   The
selection  process used  the combined  interests  of   its  diverse  membership  to
rank chemicals for further research.

-------
Brown, S.L.; Chan, F.Y.;  Jones, J.L., et al.
Stanford Research Institute, Menlo Park, California
Research Program on Hazard Priority Ranking of Manufactured Chemicals
5 volumes
Prepared for National Science Foundation
Washington, B.C.
NTIS  #PB-263  161/2ST;   //PB-263   162/OST;  //PB-263  163/8ST;  #PB-263  164/6ST;
//PB-263 165/3ST, April 1975


This  research  effort by  SRI was  divided  into  two  phases.   In  Phase  I,  the
universe  of  manufactured  organic  chemicals  was  reduced  to  250  chemicals
according  to the highest  calculated values  for the  release  rates  for these
chemicals.   Data  on relevant  physical   and  chemical  properties  for  these
chemicals  were  presented  to  the Advisory Panel during its  meeting  in Menlo
Park, California, on August  26-27,  1974.   From  this list  of 250  chemicals  and
from  a  list:  of  41 chemicals designated by the Advisory Panel as  "Most Wanted
Chemicals,"  the members  of  the Advisory Panel selected 80  chemicals  for Phase
II  study  as  the  chemicals  with  the greatest  potential  for  environmental
effects.   In Phase  II   of  this  two-phase  effort,  80 chemicals  having  the
greatest  potential  for   environmental  effects were  studied.   Information  was
collected  on  the  extent  of  .the environmental  exposure   to  these  chemicals
(during manufacture and  use) and on  the possible environmental effects of this
exposure  (chemical,   physical,  and  biological  properties;  persistence;  and
toxicity).   A questionnaire  survey of  industry  was conducted  to  determine  the
amount and types of  losses  in  the manufacturing  plants.   Information from this
survey  is tabulated.  The following  five summary reports  are  submitted  for
each  of  the  80 chemicals:  (1) a  flow  diagram showing industrial data  on  the
amounts used for  various  applications; (2) a  one-page assessment of  the data
gathered on  toxicology and environmental  persistence;   (3) a  computer  printout
of a  tabular summary  of  the  important  data gathered;  and (4)  copies  of the  25
most  pertinent  abstracts  found  in  the area  of  toxicology.   The  report  is
divided into five volumes.

The  first  volume  contains chemicals 1-20:  Tetrakis(hydroxymethyl)phosphonium
chloride;    Benzo(a)pyrene     (BAP);    Tetraethyl    lead;    Vinyl    chloride;
Hexachlorobenzene; o-Cresol;  Ethylbenzene; Nonylphenol,  ethoxylated  (9  moles
of   ethylene  oxide);   Hexachlorobutadiene;   Vinylidene  chloride;   Toluene;
Ethylene  dibromide;   Tris(2,3-dibromopropyl)   phosphate;  Ethylene  dichloride;
Trichloroethylene; 1,1,1,-Trichloroethane; Carbon  tetrachloride;  Chlorinated
paraffins (35-64% chlorine); Perchloroethylene; and Dichlorodifluoromethane.

The second volume contains chemicals 21—40:   Benzene  (chemical uses);  Silicone
fluids;   Nitrobenzene;   Toluenediisocyanate   (TDI);   Xylenes  (mixed   total);
Aniline;  Dimethyl  terephthalate;  Trichlorofluoromethane;  p-Dichlorobenzene;
Tetrabromoethane;  Methylenebis(2-chloroaniline);   Polyhalogenated   biphenyls
(Aroclor  1254); Tricresyl  phosphate; Fluorescent brightening  agents  (no.  23);
Polyvinyl   chloride;  Methylene   chloride;   Dichloropropene,   Dichloropropane
mixture; and Ethyl chloride.
                                      25

-------
The  third  volume  contains  chemicals  41-60:    (Ethylenedinitrilo)tetraacetic
acid,  tetrasodium  salt;  Benzidine;  Zinc  di(butylhexyl)   phosphorodithioate;
N-(Dimethylpentyl)-N-phenyl-p-phenylenediamine;  Vinyltoluene;  Methyl  bromide;
Mercaptobenzothiazole;  Chloroform;  Acetontrile;  Di(ethylhexyl) phthalate;  Vat
Blue Dye  no.  6  (Dichloroindanthrone);  Di(ethylhexyl)  adipate; Dimethylamine;
Dichlorobenzidine;  Hexamethylenetetramine; Acyclic  xanthic  acid salts  (Sodium
isopropylxanthate-Dow  Z-ll);   Sulfolane-p-nonylphenyl   manganese   phosphite;
Nitrilotriacetic  acid   (trisodium  salt);  and   Diarylarylenediamines  (mixed)
(Wingstay 100).

The  fourth  volume  contains  chemicals  61-79:    Polyacrylonitrile  ' (fibers);
Naphthalene; N,N-Dimethyldodecylamine oxide;  Chloroprene; Formaldehyde  (37%  by
weight);  Bis(hydrogenated   tallow   alkyl)dimethylammonium  chloride;   Methyl
chloride;  Dioxane;  Ethylene  oxide;  Allyl   chloride;   Ethylenimine;  Phenol;
Tri (chloroethyl)phosphate;   Polyethylene  glycols   (MW   400);   Methoxye thariDl;
Dodecylbenzenesulfonic  acid   (sodium salt);  Bis(chloroethyl)   ether;  Dodecyl
mercaptan; Polyurethane and diisocyanate resins;  and Ethylene.

The fifth volume contains  a  listing of the  references  for the data  collected
and general notes describing  the data.
                                      25

-------
Stephenson, M.E.
An  Approach  to   the  Identification  of  Organic  Compounds  Hazardous  to  the
Environmental and Human Health
National Science Foundation
Ecotoxicol. Environ.  Safety, 1:39-48, 1977


Results  are  presented  from  a  workshop   ranking   hazardous   chemicals   and
identifying key  problem  areas for future  research.   Eighty organic  compounds
were ranked according to  their  environmental impact and  their hazard  to  human
health.  A scatter plot  of  the  80 compounds indicating  their  relative ranking
in  terms   of  both criteria  showed  that  10  organohalides  required  immediate
study.   Seven major impact or problem areas  around  which future  research  could
be  organized  include:    (1)   natural   sources  of  organic  compounds;   (.°.)
contribution  to  biochemical and  geochemical  pools;   (3)  effects of  transport
and translocation of  toxic  elements  and micronutrients;  (4) toxic  degradation
products  and  formation   of   secondary  pollutants;  (5)  remote   effects;   (6)
persistence in the absence  of other effects;  and (7)  bioaccumulation.   It  is
also recommended  that the  environmental  impact and human health hazard of  the
carbon-chlorine  bond  be  studied.   Production figures  and  estimated  annual
release rates for the  80 compounds are  given.
                                      27

-------
Bechmann, R.B; Carhart, H.W.;  Cunningham,  W.A.;  et al.
System  for  Evaluation  of   the  Hazards  of   Bulk  Water   Transportation  of
Industrial Chemicals
National Academy of Sciences
Contract No.  DOT-CG-41.680-A,  1974
An  evaluation scheme,  developed  for  the  U.S.  Coast  Guard,  to  assess  the
potential  hazard  of  industrial  chemicals  transported  in  bulk  by  water  is
presented.   Each  chemical  is  rated  on a  scale  of  0  to  4  (which  reflects
increasing hazard) on nine hazard parameters:   (1) fire,  (2)  contact  of liquid
with  skin  and  eyes,  (3)  occasional  short-term inhalation  of  vapors,  (4)
occasional  short-term  inhalation of  gases,  (5)  repeated  inhalation  of  gases
and vapors,  (6)  human toxicity  due  to water  pollution,  (7) aquatic  toxicity
due  to  water  pollution,   (8)   reaction  with  water,  and  (9)   self-reaction
(usually polymerization or decomposition).   The human toxicity rating  is based
on  the  oral  LE>50  in a  suitable  mammalian   species.   The  aquatic   toxicity
rating is based on the 95-hour TLm in adult or  juvenile fish from upper levels
of  the   food  chain   and  includes  special  notations   on  bioaccumulation,
biochemical oxygen demand,  and  insolubility.   An appendix  evaluates  potential
concentration  ranges  of  discharged  material   into  three  selected  receiving
system:  (1) rivers,  (2)  estua.ries,  and (3)  coastal waters.

-------
Brown, S.L.; Cohen, J.M.;  Macrae, N.I.;  Small, M.J.
SRI International
Setting Priorities for R & D on Army Chemicals--Final Report
U.S. Army Medical Research and Development Command
Fort Detrick, Frederick,  MD.
Contract No. DAMD17-75-C-5071; CRESS No. 56, May 1978


The Environmental  Protection Research  Division of   the  Array Medical  Research
and   Development   Command   is  charged   with  developing  the  database   for
environmental and  occupational  health  criteria  dealing  with  the  manufacture,
use,  and  disposal of  chemicals  in Army  activities.  This  report  presents  a
method  that  can  assist  the  Army  in  allocating   resources  among  candidate
research and development studies on the environmental  and occupational  effects
of Army chemicals.

The basis  of the method  is a mathematical model of the process  leading  from
the initial  pollution  of  air, water,  land, or  the  workplace  to  the  eventual
environmental or occupational health  effects  of  the  chemicals  in  question.
Both  environmental and  occupational  hazard   assessment  models  consider  the
chemical discharge  source;  environmental  medium;  concentratton as  determined
by  transport, transformation, .and  transfer; population at  risk; exposure;  and
effects.  Effects  in  humans include carcinogenicity,  severe chronic toxicity,
recoverable  chronic  toxicity, and  acute  toxicity.   Effects  in  fish  include
acute and  chronic  toxicity.   Effects are  assigned  a relative  dollar  value  to
allow comparison among different chemicals.   For incomplete  or  missing  data  on
a chemical,  the value is estimated, assigned  a  default value,  or assumed  to  be
unimportant.  The  model  estimates  a  total  hazard value, weighted  among  human
and ecological effects, with  a  corresponding  uncertainty value  due  to  lack  of
knowledge.    The   allocation  method   then   compares  the  reduction  in  hazard
uncertainty  expected  to  be  achieved  after  a  research study with  t'na  cost  of
the study,  and ranks candidate studies according to  the ratio.
                                      29

-------
Margler, L.W.;  Rogozen, M.B.;  Ziskind,  R.A.;  et al.
Science Applications, Incorporated, Los Angeles, California
Rapid Screening and  Identification  of Airborne  Carcinogens  of  Greatest  Concern
in California
J. Air Pollut.  Control Assoc., 29(11):1153-1157, 1979
The method used by  the  California  Air Resources Board to quickly  identify  and
rank potentially  serious airborne  carcinogens  is described.   Eight  lists  of
carcinogens were  compared  after  eliminating  chemicals not used  in California,
pesticides,  chemicals  unstable  in   air,  or  doubtful  carcinogens.   Further
eliminations  were made  if production  or use  was  under 100,000  pounds  per
year.   The  investigators   included   a  few  additional  compounds.   Candidate
substances  were   ranked  by additive  and multiplicative  algorithms  and  by  a
panel of  experts.   For  the additive algorithm,  the  rating  of a  sabstance  was
the sura of  the scores of each  rating  factor  multiplied by a  weighting factor.
For the multiplicative  algorithm,  the substance rating equaled  the  product  of
the rating  factors.   The six  criteria  rated  were:   (1)  present use,  (2)  use
trends,  (3)  emission  potential,  (4)  stability  in  ambient air,  (5)  dispersion
potential, and (6)  evidence of carcinogenicity.  Those compounds  appearing  in
the  top   11  of  at  least   two  lists  were  selected  for  further  study.    The
substances   selected   were  a-rsenic,   asbestos,   benzene,   cadmium,   carbon
tetrachloride,    chloroform,    ethylene   dibromide,   ethylene   dichlorids,
nitrosaraines,  perchloroethylene,   and  polychlorinated  aromatic  hydrocarbons.
The authors found this screening approach efficient.
                                      30

-------
Powers, R.;  Bednarz,  R.;  Hurlburt, G.;  Koebbe,  J.A.;  Kavanagh,  T.
Michigan's Critical Materials Register and Chemical Hazard Assessment Program
Michigan Critical Materials Register 1930
Department of Natural Resources Environmental Protection Bureau
Environmental Services Division
Publication Number 4833-5324, 1980


Michigan's  Critical  Materials  Register  (CMR)   system,  developed  to  protect
water quality, is described.  A list of  suspect  chemicals  is  selected annually
by an advisory committee from  established  priority lists  and  from  chemicals  of
concern in Michigan.   The CMR  list  includes:  inorganics,  organics,  pesticides,
herbicides,  fungicides,  pharmaceuticals, food additives, and  natural  materials
which  may  be  used  or discharged  in Michigan.   Each chemical  is scored  for
eight  factors:   (1)  acute  toxicity;  (2)  adverse   effects   such  as  chronic
toxicity,    fetotoxicity,   embryotoxicity,   phytotoxicity,   and   effects   of
metabolites   and  degradation products;  (3)  persistence;  (4)  bioaccumulation;
(5) carcinogenicity;  (6) teratogenicity;  (7)  hereditary mutagenicity;  and  (3)
esthetics.    Factors  are weighted  such  that phytotoxicity,   persistence,  and
esthetics  have maximum  scores  of  3,  4,  and  1,  respectively,  and  remaining
factors have a maximum score of 7.   When the data are insufficient  to  score a
factor, an  asterisk  is  assigned  and  the  factor  is  scored  as  data  become
available.    Hazard assessment * criteria  and  rationale are  presented for  each
category.   Chemicals with  a  cumulative  score of  7 or greater are  included  in
the  register.   The  hazard  ranking  system  has  been  modified to  incorporate
potential  hazards from air  pollutants.  Further  review of  the  CMR  list  focuses
on chemicals  with  the most number  of  asterisks  ind  chemicals with  cumulative
scores of less than  7.   Businesses  are  required to  file  an   annual  report  on
the use and  discharge  of  CMR  chemicals.   The authors  note three  problems  in
the assessment program:  (1)  chemicals lacking sufficient data  are  not included
on  the  list,  (2)  lack  of  adequate  chemical  composition  data  on  commer.. Lil
products causes  difficulty  in  identifying  products  which may  contain  toxic
substances,   and  (3)  many  chemicals  which may  not   be  used  in  Michigan  are
included on  the  CMR list  because Michigan  laws do  not  require  reporting  of
inventory  data prior  to  chemical manufacture and use.
                                      31

-------
United Nations Environmental Programme
Initial Report on the Priority Pollutants Project
Nairobi, October 1978
Results  are  presented  from  the  initial  phase  of  the  priority  pollutants
project.  The  objectives  include:   compiling a  list  of  hazardous  chemicals
which  have  been  given  priority  consideration  by   governments   and   other
organizations,  proposing   a  systematic   process    for   selecting   priority
pollutants  with  emphasis  given  to  the  needs  of  developing  countries,  and
identifying  several   lists   of  chemicals  or  categories  for  further  United
Nations  Enviroa-nental  Programme   (UNEP)   attention.    Section  2   includes  a
compilation  of  approximately ' 230  selected substances  of concern  to  various
national  and international  institutions.  Section   3  discusses  the  general
problems  associated  with  identifying  and ranking  environmental hazards.   It
includes  a  discussion  of  the distinguishing  features  between developed  and
developing  areas  which  give  rise   to   different   sets  of   priorities  for
environmental pollution.   This section  concludes  by  outlining the  selection
criteria  to  be  used  in a  ranking system  for both  developed and  developing
areas.     Criteria     include    exposure,    carcinogenicity,    mutagenicity,
teratogenicity,   acute  toxicity,   other  health  effects,   the   influence   of
nutrition and infectious disease on toxicity,  persistence  and  bioaccumulation,
and  environmental  agents.    The  method  uses   expert  opinion,  information  on
structure-activity relationships,  and  production  and  environmental  exposure
data  in a multistage  screening  process  where a  relatively  large  number  of
substances are considered initially.   In  subsequent stt?ps, a  smaller subset is
selected  for collection of  more   data  and more  thorough review.   Se'ction  5
suggests  several  categories  of hazardous  pollutants  which may be of  interest
to  the  UNEP.   For  each  area,  an  interim  list  of   priority  pollutants  is
proposed.
                                      32

-------
Brown, S.L.; Gofer, R.L.; Eger,  T.;  Liu, D.H.W.;  Mabey,  W.R.;  Tus'e,  D.
SHI International
A Ranking Algorithm for EEC Water Pollutants—Final Report
Contract No. ENV/223/74-EN: CRESS Report No. 136, September 1930


This  report  describes   an  algorithm  used  to   rank  1,500  possible  European
Economic  Community (EEC)  water pollutants  for   prioritization  of  regulatory
action  on discharge  limitation?.   Pollutants such  as   persistent  synthetics,
mercury, or cadmium compounds already under EEC  regulation,  compounds  produced
or used in the EEC in amounts less  than 100 metric  tons  per  year,  or compounds
which could  not  be idantified  were eliminated   from  the list.  The  remaining
426   pollutants   were   then   ranked  by  a   "screening   algorithm"   based   on
production, environmental  half-life,  and acute  toxicity to aquatic  organisms
and  mammals.   The top  130  compounds   were  then  ranked  using  a  "complete
algorithm"  which  incorporated  information on:   sources (amount  produced  or
consumed  in  ESC   and  estimated  fraction used as an  intermediate);  discharges
(industrial, agricultural, domestic, commercial,  water  treatment,  or un'x'iovn);
persistence;  bioconcentration  partition  coefficient;   acute   aquatic   effect
level (LC5g);  chronic aquatic  effect  level  (TC  low);  acute  mammalian  effect
level (LD^Q);  chronic  mammalian effect  level  (TD low);  and chronic  mammalian
effect  reported  (carcinogenicity,   mutagenicity,  or  teratogenicity).    Each
factor was  labeled  as  to whether it was based on reported data or  analogy  to
other compounds  and each factor except  discharge was scored  for  reliability.
The  authors  recommend   that  priority   regulatory  attention  be  given  to  top
ranked  pollutants with   good  reliability  indexes and  that  additional data  be
collected on  pollutants  with  low reliability  indexes before taking  regulatory
action.
                                      33

-------
Gori, G.B.
Ranking of Environmental Contaminants for Bioassay Priority
International Agency for Research on Cancer,  ISS  No.  16:99-111.   1977


A ranking index was  developed  to classify chemicals by  their  potential  hazard
and  to  indicate  which  chemicals  need  to  be  examined   first   for   their
carcinogenic properties.  The  index is  based  on a quantitative  evaluation  of
their  prevalence  and  distribution  in   the  environment,  intake in  man,  and
similarity  to  known  carcinogens.   Two  major  indicators  used  to  determine
carcinogenic  potential  are  (1)  exposure  and   intake   conditions,   and  (2)
suspicion  of  activity.   Exposure  and  intake conditions were  formulated  by
analyzing  the  number of  persons  exposed  and  the number  of  estimated  annual
intakes per person  in  each  subpopulation.  The  second indicator,  suspicion  of
activity,  was   formulated   from   the   probability   that  the   chemical   was
carcinogenic for each  intake  route and  the  estimated  relative  potency  of  the
chemical  for  each  intake  route.   Possible  contamination  from  manufacturing
processes can affect exposure to a  compound  and  tracing  techniques  ars used  to
identify  sources  of  contaminants.   Forward  tracing  involves  analyzing  raw
materials  of  a  chemical up  to  the  intake  or use  by man  and  identifying
contaminants.   Backward tracing starts by  analyzing an environmental  situation
and  proceeding back to the cause  of contamination.   Forward  tracing  examines
chemicals  with   significant   production  while  backward  tracing   examines
compounds with high  user  exposure.  Another met'iod  used to aid  in  confirming
suspicion  of  a   chemical   as   a   carcinogen  is  the   chemical   structure.
Carcinogenic  probability   and  potency   estimates  for   known   compounds  are
determined by an expert panel and  ranked  into  a  structure tree.   New  compounds
are  compared  to   the   structure  tree  and,    for  each  intake   route,   the
carcinogenic activity is  estimated  based  on  the  carcinogenic  probability times
the  estimated carcinogenic  potency.   With the information incorporated  into  a
single ranking index, any new  compound  structure may be  compared and  assigned
an  index  of  suspicion.  Further  development   of  the  system   could  lead  to
identifying the carcinogenic hazard of chemicals  before they  are produced.
                                      34

-------
Van NeCten, C.
Simon Fraser University, Burnaby, British Columbia, Canada
Critical  Review of  Strategies  Aimed  at  Identifying  Chemicals  Hazardous  to
Human Health and the Environment
Prepared for Department of National Health and Welfare, Ottawa, Canada,
March 8, 1978


Criteria are  presented for evaluating  the  effectiveness of  existing  chemical
screening strategies.   Environmental  criteria include  production and  release,
distribution  in the  atmosphere,  degradation,  bioaccumulation,  and  toxicity,  as
well  as  synergistic and  antagonistic  effects within  the environment.   Human
health criteria  include exposure of  the  total population or  specific groups,
distribution  of the  substance  within  the  body,  accumulation and  excretion,
general  toxicity,  rautagenicity,  teratogenicity,  carcinogenicity, and  possible
synergistic and antagonistic  effects  in humans.   Decision-making criteria  are
used  to  evaluate  whether the  recommendations .nade by  the  different  screening
procedures allow easy  access  to  information so that specific  decisions  can  be
made.  Six screening methods  are then assessed by  these  criteria.  The  author
suggests that an  optimum  combination  of  these methods could  be derived  which
would reject a maximum number of chemicals at a minimum cost.

-------
Astill, B.D.; Lockhart,  H.B.,  Jr.;  Moses,  J.B.;  et al.
Eastman Kodak Company,  Rochester, New York
Sequential Testing for  Chemical Risk Assessment
Presented at the Second International Congress on Toxicology
Brussels, Belgium, July 6-11,  1980


A  method of quantitatively  evaluating environmental  and  health  hazards  ar?
presented.  Four categories of effects were established:  (1)  the  magnitude  .>r
environmental exposure  (ME), (2) the magnitude of human exposure  (MH),  (3)  the
effects  on  human  health  of  an  exposure  (H),  and  (4)   the  effects  on  the
environment of an exposure (E).  The  criteria for MH are production  per  year,
number of people  exposed,  duration  of  exposure,  and number of groups  exposed.
The criteria for ME are amount discharged per year,  number  of  discharge  sites,
number of discharges per year, and half-life  in  the  environment.   Criteria  for
H  are  the  oral  ^^Q  in  rats,   reversibility  of  immediate   affects,  and
reversibility  of  prolonged  effects.   Criteria  for E  are  the  5-hour  median
inhibitory concentration on  waste  water  treatment  microorganisms, the  median
lethal concentration on fish, and  the  octanol-water  partition  coefficient as a
measure  of  bioconcentration potential.   Each criterion  is given a  maximum,
median and minimum  range and  these are ranked 1,  2,  or 3, respectively.   The
cumulative health or environmental  score  for a  compound  is evaluated so that
scores of 9 or less, 10  to 13,  14  to 17,  or 18  or more suggest  testing  levels
I,  II,  III,  or  IV,  respectively.   Level  I testing  includes  physical  and
chemical  properties,  acute health  effects,  in  vitro  mutagenesis,  and  acute
environmental  tests.   Level  II  includes  2-week   feeding  or inhalation  tests,
skin  painting  tests,  additional mutagenesis  tests,  activated sludge  effects,
photodegradation, biodegradation, effects on  plant growth and  germination,  and
partition  coefficients.    Level  III  includes 90-day  feeding  or  inhalation,
fertility   tests,   teratology,   half-life   and   metabolites   in   rodents,
nitrification  inhibition,   algal  toxicity,  14-   to   21-day   biodegradation,
simulated fate study, larval fish  studies,  and bioconcentration  factor.   Le^el
IV   includes   2-year   feeding   studies,   3-generation  reproduction   studies,
teratology,  pharmacokinetics,  biodegradation products, soil   interaction,  and
long-term aquatic studies.   If  testing is  conducted,  the  results may  then  be
entered into the  scoring and the total score  reassessed.  The  authors conclude
that  this screening protocol  is  both  cost  and  time  effective,  as  over  500
chemicals have been tested  in 2 years.
                                      36

-------
Nees, P.O. .
Hooker Chemical Company, Niagara Falls, New York
Assessment  of  Oncogenic Potential  - A Scoring  Matrix to  Determine  Oncogenic
Potential Proposed for Application in Risk Assessment
In:  Toxic Substances Control, Vol.  Ill - Implementing the Regulatory Program
Miller, M.L., ed. , Government Institutes,  Incorporated
Washington, B.C.  pp. 168-82, 1979


A  scoring  matrix for  Oncogenic risk  assessment  that assigns  weighted  number
values to  subjective  and  objective  results of studies  is described.   Positive
lifetime animal  studies  receive a primary score  of  A,  which  is  multiplied  by
metabolism, route of administration, and  quality  of  study scores.   This  result
is  then  weighted with  a  dose level factor and  a time until  tumor appearance
factor.   Negative  results  start  with a minus  primary  score  and  the  time
weighting  is correlated  to  survival.    Epidemiology  studies   have a  primary
score of 10  multiplied  by  specificity  of  tumor  type,  suitability  of  controls,
indirect association,  relative  risk,  mixed exposure, and specificity  factors,
and  added  to   a  detectability  of  increased cancer incidence  factor.   The
resulting  adjusted primary  score (AP)  is weighted by adding  AP  times  a  dose
response factor  to AP  times a  consistency of association factors  to AP times
an  exposure  level   factor.  , There  is  also   a   weighting   for   repetitive
epidemiological  studies to  compensate for low  confidence  level  in a  single
study.   Negative  studies  start  with  a  primary  score   of  6   and  have  fewer
factors.   For short-term genetic, microbial,  and  fluid  assays,  individual  test
scores are  small, positive  or  negative,  and  the  only weighting factor  is  for
supportive data.  The sum of all  the scores is ranked on  an index  of  oncogenic
potential  that  correlates   the  total   score  to  an  appropriate   ra^uLilory
response.  The author concludes  that  the  scoring matrix  can be used by  people
with  little  experience.  An  expert  review panel  would  be  necess-t.-'/  ^nly  if
questions arose.  The assessment could change  with the results  of  new studies.
                                      37

-------
Kimerle, R.A.;  Gledhill, W.E.;  Levinskas,  G.J.
Monsanto Co., Environmental Assessment Department,  St.  Louis,  Missouri
Environmental Safety Assessment of New Materials
In: Cairns, J.; Dickson, K.L.;  Maki, A.W.; Eds.
American Society for Testing Materials STP No.  657:132-146,  1978
A procedure is presented for evaluating the hazard  to  aquatic  organisms  of  new
materials prior  to  their  commercialization.   The procedure  uses  environmental
fate and aquatic organism toxicity data in  the  saqaential  phases  of screening,
predicting, confirming, and  monitoring  to  reach a decision  either  to  continue
the  toxicity  testing,  terminate the project  because  of unacceptable  risk,  or
cease  testing  because  of  an acceptable  risk.   The  screening phase  involves
short-term acute  tests,  while   the predictive  phase  involves  short- and  long-
term  laboratory  studies.    In  the  confirmation  phase,  environmental   field
studies are designed  to answer  critical questions of  environmental  safaty.   In
the  monitoring  phase,  field  studies  under   actual   use   conditions   after
commercialization  are  conducted  to  confirm   the  ultimate   safety  of   the
material.  The  types  of tests  in  each  phase and the  criteria that  determine
which  specific  tests  are needed  are presented.  The  criteria for evaluating
whether or not a hazard exists  also are  discussed.
                                      38

-------
Booz Allen Applied Research, Inc.
A Study  of  Hazardous  Waste Materials.  Hazardous Effects  and  Disposal  Methods
Vol. 1, 1973
A system is described  for  rating hazardous materials.  One  rating  equation is
based  on known  data  and   a  second  rating  equation  is  used  for  assessing
compounds with unknown data.  The first equation is HR = TER x HER

           where:  HR is the hazard rating,
                   TER is the total effects rating,
           and     HER is the hazard extent rating.

TER is the sum of values assigned  for  air,  water,  and soil contaminant hazards
on  human populations;   effects  from  flame,   explosion,   or  reaction  of  the
material; and  toxic  effects on ecological populations.  Toxic  effects include
acute  and  chronic ^£50  and TLV  for  air  and  water,  and  effects  of  contact
exposure for  soil.   Each factor can be weighted  to reflect  importance.   The
score  for  each  of   the  factors may be  U (unknown),  1   (minimal  hazard),  2
(slight  to  moderate hazard),  or  3  (severe  hazard).    HER  is  the  sum  of
production and  consumer  distribution scores.    Production  scores  are  based  on
the amount produced per year and may be U  (unknown),  1  (low),  1.2.5  (moderate),
or  1.5  (large).   Consumer distribution   scores  may  be  0  (limited),   0.25
(moderate),  or 0.5 (wide).   Criteria for scoring TER and HER are presented.

The second equation is MPHR = (3U + TER) x HER

           where:  MPHR is the maximum potential hazard rating,
           and     3U = a rating of 3 to each  unknown to produce a conservative
                        hazard rati.ig.

Ratings are  given for many hazardous materials used as warfare agents.
                                      39

-------
Jones, C.J.
The Ranking of Hazardous Materials by Means of Hazard Indices
J. Hazard. Mater., 2:363-389,  1978


Several indexing models for use  in  landfill  management,  transport of dangerous
goods,  environmental  quality  assessment,  and  other  areas   are   reviewed.
Suggestions are  then presented  for developing a  hazard index  for  evaluating
waste  management  options.   The  procedure involves  listing  the materials  or
wastes of  interest  and the management  or disposal  options to  be  considered.
The properties of the materials or  wastes  which are  relevant to the  evaluation
process   ar«   obtained   from   the  existing   literature  or   from   laboratory
measurements.   A  combination  model appropriate  to  the  management  situation
must  then  be  constructed.   The  author  recommends  the additive  utility  model,
but notes  that other  models  may be  more appropriate  in  certain cases.   The
normalized indexes  for  each  material and  the  relative  utility  range  for  each
property are  derived  from  the data and  the  overall  index  is  calculated  using
the combination model.  The overall indexes  must  reflect  the actual  properties
of the material in  combination with the  relative  importance attributed to  ea;h
property by the weighting factor.   The author  observes  that I.f 'c'i =  combination
model  is  mathematically  sound,  the  use  of an  indexing  and  combination  model
allows versatility  in quantifying  value  judgements about  aspects  of a  given
material's environmental behavior.
                                      40

-------
Cramer, G.M.;  Ford, R.A.;  Hall, R.L.
Estimation of Toxic Hazard — A Decision Tree Approach
Food Cosmet. Toxicol., 16(3) :255-276,  1978


A procedure using  33  criteria  for establishing toxic hazard is  presented  as  a
decision   tree  organized   into   branches    dealing   with   major   chemical
classifications.   The  procedure  is  intended  for  use  with  all  ingested,
structurally-defined organic and metallo-organic substances.   The  criteria are
based  on  features of  chemical  structure,   occurrence  in  body  tissues  and
fluids, and natural occurrence in food.   The  logic  of  the tree rests  heavily
on  known  data  on  metabolism  and  toxicity.    The  classification  according  to
presumptive toxicity can be combined with knowledge  of  human  intake  to  provide
a protection  index for each  substance.   The  index can  be  used  to  establish
priorities  and  define tentatively  the  extent of  appropriate  testing.   It  is
noted  that  the procedure has  been applied  to a  large  number  of  pesticides,
drugs,  food additives, and  industrial  and  environmental  chemicals  of  known
biological  properties.   Because  the  procedure  has  not  yet  resulted  in  any
underestimation  of   toxicity,   it   is   seen  as   a   practical   means   for
discriminating effectively among different levels of probable  hazard.
                                      41

-------
Cairns, J.;  Dickson,  K.L.,  Maki,  A.W.
Estimating the Hazard of Chemical Substances to Aquatic  Life
Hydrobiologia, 64(2) :157-166,  1979
A  conceptual  framework   for  conducting  a  hazard  assessment  is  presented.
Various toxicity  tests  and procedures for  evaluating  hazards to  aquatic  life
are  compared,  and  the   decision  criteria  used   in   these   procedures   ar^
discussed.  The use of safety factors or "uncertainty factors"  is  discussed as
a  central concept  in  a  sequential  testing  approach  in  which  estimates  of
expected  chemical  concentrations  in the environment and  their effect.; can be
made with an increasing degree of accuracy.   The state of  the  art  of  assessin^
hazards from chemicals to  aquatic life is  reviewed;  safety,  hazard, and  risk
concepts  involved  in   such  assessments   also  are  discussed.    Particular
attention is given to  hazard assessment procedures  developed by  the  American
Society for Testing Materials,  the  American Institute of  Biological  Sciences,
and Monsanto.
                                      42

-------
Ingrahara, T.R.;  Zechel,  E.M.
Selection of Candidate  Substances for Control  in an  Environmental  Regulatory
Agency
J. Air Pollut.  Control Assoc., 29(l):38-42, 1979


A  quantitative  decision-making  process   based  on  the  serial  application  of
three  models  and  incorporating  measured data  and  subjective  judgment  is
presented for  selecting  environmental  regulation  candidate substances.   The
first model defines  the  severity  of  a pollution  problem.  Values  assigned for
damage to an individual organism, concentration  of the  pollutant,  percentage
retention of the  pollutant,  duration  of  exposure, hazard to  the  organism, and
the total population of organisms are  multiplied  to  attain the  tocal  damage
index.  The second model defines  the  social,  economic,  and  political  impact of
regulation.   Values  assigned on  a  scale   of  -5  to +5  for  nine  factors:   (1)
public health  protection;  (2) employment  and income  stability;  (3)  lifestyle
changes;   (4)  creation   of   income   disparity;  (5)   alterations   in   market
structures;   (6)   changes  in  regional,   provincial,  or  national  balance  of
payments; (7) public acceptance;  (8) effects on  territorial sovereignty;  and
(9) contribution  to  inflation,  are  totaled  to  determine  Lnpact.  The  third
model   assesses   the  current  technological  capability  for   implementing  a
potential  regulation.    Values  assigned  to  reliability,  efficiency,   general
applicability,  dollar  and  manpower  resources,  and   development  in  a  given
period of  time  are  multiplied  to attain the  magnitude  of ''chft  cechaological
capability.   Illustrative cases are  given for each model.
                                      43

-------
Keith, L.H.; Telliard, W.A.
Priority Pollutants.  I.   A Perspective View
Environ. Sci. Technol., 13(4):A16-423,  1979


The historical origins of  the EPA Priority  Pollutants and the  development  of
the  Priority (Water)  Pollutants  Protocol  are   related.   Events  are  traced
beginning with the court  decision in 1978  that resulted  in  adoption  of the EPA
Consent Decree  to control  the  levels  of  pollutants  in  industrial  wastewater
discharges.  Components of  the decree  are  reviewed, including  formation of the
Toxic Pollutant List  to  aid in  the  classification of harmful  substances.   It
is  noted  that,   since  the  decree  lacked  allowance  for  time  La  solving
analytical  problems  in testing  for  these substances  in wastewaters,  the  EPA
adopted four  criteria to prioritize and  select   representative  compounds  from
each group.  (1)  All  compounds  specifically named on  the Toxic  Pollutant  List
were  automatically included.   (The availability  of  chemical   standards  for
verification and  quantification  was considered  mandatory.   (2)  Compounds  not
found on  the list should be  tested if they  accounted for 5^  or more  of  the
total known  listing  for  the class  of  compounds.   (3) All  chemical  praduction
data should  be reviewed, where  available.   (4) Other  sources  were  examined  to
determine  if  the  compound  was  a  recognized   water  pollutant.   Next,   the
screening,  Aerification,  and  monitoring   strategy developed  by  the EPA  for
testing  water  for  pollutants  is  presented.   Unresolved  problems  in   the
screening  stage  are  discussed,  along  with  automated  software  programs  being
adopted to  speed  up  the  analysis process.   Future plans and  first  drafts  for
implementation of  the monitoring  phase are considered,  and  a  list of  the  129
compounds on the Toxic Pollutant List is included.
                                      44

-------
Langner, R.R.;  Norwood, S.K.; Socha, G.E.; Hoyle, H.R.
Two Methods for Establishing Industrial Hygiene Priorities
Am. Ind. Hyg. Assoc. J., 40(12):1039-1045, 1979


Two approaches,  health team rating  and  calculation rating,  are  described for
determining  the  health  hazard   of   industrial   chemicals  and   establishing
industrial hygiene  priorities.   In health team hazard  rating,  a  team composed
of  an  industrial  hygienist,  physician,  toxicologist,   safety  engineai',  and
process engineer rates  the  degree  of  exposure and consequences of overexposura
to a  chemical  for  each job classification  in a facility.  Degree  of exposure
ratings are  no  contact,  minor  contact,  occasional  daily  contact,  and  gross
contact' likely.  The assessments of  the  consequences  of overexposure are based
on  acute  toxicity,  chronic  toxicity,   and  available  data  on  eye,  ski-i,
inhalation, and  ingestion exposures.   The  health  team then  ranks  each che'TiI: i1
for each  job classification on a  scale  of 1  to  5.   Class  1  chemicals  h*^  i
significant  probability  of  a  serious   overexposure  and   require  immedia-;-
investigation.   Class  2 chemicals  have a possibility  of overexposure, but with
less  severe   consequences   than   Class   1,   and   should  be   monitored  after
investigating  Class  1.   Class  3 chemicals  have  satisfactorily  controlled
exposure and regular monitoring  is recommended.   Class  4 chemicals  have only a
remote  possibility  of  overexposure with  serious consequences  and  require only
confirmatory monitoring.  Class  5  chemicals pose no  health hazard  and require
no monitoring.

Calculated  ratings  can be  used  to  assess  the  health   hazard  of  numerous
chemicals  under varying  conditions   and  use.   A formula  is  presented  for
calculating hazard due  to inhalation  of  gases and sapors.   Parameters included
in  the  formula  are:    degree  of  exposure,  consequences  of  overexposure,
atmospheric boiling point, and the maximum airborne workplace concentraL i _•> i  i. .•>
which a  chemical should be  controlled  to  prevent significant adverse  healc'i
affects to  an  employee.  Degree of  exposure  is  rated  on a  scale  of 1  to  ~~>,
including the  four categories  described  previously plus a  penalty  factor or ^
that  is added  when the process  is operated  at  greater than  5 atmospheres  of
pressure.    Consequences of  overaxposure  are  rated  on  a   scale  of  1  to  4
according  to  the  seriousness  of   employee  health  effects.   The  calculated
hazard  rating  corresponds   to  five  subsequent action  steps  similar  to  those
previously  described.   The  authors  note  that  both  approaches  are  based  on
hazard  rather   than  toxicity,  thereby allowing  appropriate   establishment  of
hygiene priorities.  They  further  note  that  the calculated  health  hazard may
be a  more useful method in  determining regulatory action and  is  adaptable for
computer use for broad-based studies.
                                      45

-------
Branson, D.R.
Prioritization of Chemicals  According  to the Degree  of  Hazard in  the  Aquatic
Environment
Environ. Health Perspect.,  34:133-138, 1980


An   aquatic   hazard   assessment   technique  is   proposed  for   prioritizing
chemicals.  Aquatic hazard  is  defined  as the margin between  the  no observable
adverse effect  concentration (NOAEC)  and  the  ambient exposure  concentration.
In  fish,  the safety  margin is the  NOAEC in chronic  or  sensitive  life-stage
tests divided by the  ambient exposure  concentration in water.   In fish  eaters,
the  safety margin  is  the acceptable daily  intake  of  residues  in  fish  divided
by  the  ambient  concentration in fish.   Ambient exposure concentrations  may  be
measured  directly  or  predicted   from  a  nomogram  based   on   the  ambient
concentration   in   fish   and   the   ratio  of   relative   dissipation   and
bioconcentration potential.  A chemical with a safety margin  of  5 or higher is
classified as a priority chemical.
                                      46

-------
Kornreich, M.R.;  Nisbet, I; Fensterheim, R.;  et al. ,
Clement Associates, Inc.
Priority Setting  of Toxic Substances for Guiding Monitoring Programs
Contract No.  OTA-C-78-372,  1980
Thirty-two  previously existing  priority  lists  for  monitoring,  testing,  or
regulating chemicals are reviewed.   The  values of the lists  are  discussed and
a new  priority  system for  monitoring  potentially • toxic  chemicals  in  food  is
proposed.  A master  list  of priority chemicals was  comprised from established
priority lists  as  well  as   chemicals recommended  for  inclusion by an advisory
panel.   Through  panel decision,  select  classes of chemicals  were  removed from
the master  list to  shorten  the  list and  concentrate  on food  contaminants  oc
greatest concern.  The chemical classes  removed  from the list were pesticides,
drugs,  intentional food additives,  chemicals  not  in  commercial  prod i~l i.M, and
chemicals with  limited  occurrence  in the  environment.  The  remaining organic
chemicals on  the list were grouped into  class categories,  £.3.,  chlorinated
benzenes, cresols, xylenes,  etc.,   to further reduce  the list.  All compounds
of each  metallic 'il^nant were considered  as  a single group  'dad all  forms  of
asbestos were  grouped together.    The  problems associated  wi th such  grouping
and evaluating  representative  chemicals  from each  group  are  discussed.   The
chemicals were evaluated on  exposure and  biological  factors.   Exposure factors
included  data   on   bioaccumirlation   potential,   environmental   persistence,
occurrence in water,  production,  use pattern,  occurrence  in- foods,  potential
for post-sale contamination, and  volatility  of organic  chemicals.   Biological
factors    included    acute    toxicity,     carcinogenic!ty,    mutagenicity,
ter itogeaicity,   other  toxic effects,  and  health  effects  attributed  to  food
contamination.    Weights  were assigned  for  each  factor  and  a  total  score v/a-;
conputed.  Objective  criteria  were used  to  score  each factor when  pos'^o"!-.1
(e.g.,   oral  LD5Q   for  acute  toxicity),  but,   when  necessary,   subjactl ' ;
judgment  from   the  expert   panel  supplemented  the  objective  data.    Problems
associated with ranking systems  such  as  this one,   e.g.,  unknown  data,  are
discussed.  A separate ranking  system is proposed for  radionuclides.

-------
Oiler, W.L.; Cairns, T. ;  Bowman, M.C.; Fishbeln, L.
A  Toxicolog ical Risk  Assessment  Procedure:   A  Proposal  for a
Index for Hazardous Chemicals
Arch. Environ. Contain. Toxicol., 9(4) :483-490, 1980


The  Surveillance  Index (SI)  is a  quantitative  method  to  rank  the  potential
risk  of  suspect foo-1  residues  for  prioritization in food  residue  monitoring
programs.   The  SI  is  equal   to   the   sum   of   the  toxicity  factor  (TF) ,
environmental  factor  (SF) ,  and  biosafety  factor  (BSF) (SI =  TF  + ,£F  + BSF).
The  TF  is  equal  to  the  kilotons  of  a  compound  released  annually  into  the
environment (KT) multiplied  by  the  relative  toxicity ratio (RTH) of  :Ye oral
LD5Q  in  rats  of  dieldrin  to  the  ^050  of  the  suspect  compound.   The  5?   is
equal to the effective half-life  (1/2E),  which is  calculated  from the physical
half-life  and  biological  half-life,  multiplied   by   the  sum  of  crop  values
(CV).  Values  are  assigned   to  crops  in  ./nich residues are  likely  t:o occar
based  on   preparation  required   for   the   commodity  prior   to   consumption,
principal   consumers,  and    the   time   interval   between  crop   harvest  and
consumption.   Values  are  high  for nonprocessed crops  such as vegetables  and
fruit, and  low for  fully processed  crops.  The BSF is determined  by  values f re
the  propensity  to  biomagnify (PB),  reactive  sites in man, population  at ri-.v.
(PAR), and  the no observable  effect  level  (NOEL) according  to the equation 3SF
=  PB x  S  x  PAR/NOEL.   To  evaluate  banned  compounds,  naturally  occurring
compounds,  and contaminants, a  modified SI  (MSI)  is calculated  according   to
the  equation MSI  =  (RTR  x EF)  + BSF.   The SI is  determined  for  several known
food  residues  and  found  Lo  ne  valid.   The  authors  note  that  the  effect   of
nontoxic changes,  such as  usage/production,   on the  SI is minimal  corn-Dared   to
the  impact  of changes  in  toxicity data.
                                      48

-------
McErlean, A.J.;  Duke, T.W.
On the Use of Indices in Aquatic Toxicological Hazard Assessment
Regulatory Toxicol. Pharmacol., 1(3-4):344-354,  1931


An environmental  index  to assess  the potential aquatic  toxicity  of chemicals
is described.   For  each  environmental  factor,  the  chemical  is  ranked on  a
subindex  scale  ranging  from  0  to  100.   The  range  for each subindex  scale  is
determined  from  the  highest  and  lowest  values  of  its  known  compounds  and
converted  to  a  0  to 100  scale.  Each subindex  score is  raised  to  a power  and
multipled  together  to  calculate  the  total  index  value.   The  sura  of  the
subindex  exponents equals  1  to ensure  a  range  of  0  to 100  for  the  total
index.   Variable  weighting  of  the  subindex exponents  can be  used  to  reflect
relative  importance  of  individual  factors.   The  calculated total  index value
of the  chemical can  be  compared to  total  index values for known  compounds  in
order  to rank relative  aquatic  hazard.   An  example  is  presented  using  three
factors:   (1)  an invertebrate  ^£50  toxicity  test,  (2) a  vertebrate  LC^Q
toxicity  test,  and (3)  the  octanol-water  partition  coefficient.  The  authors
note  that the  advantages of  this  technique include uti li^a^'I  11  >f  existing
data and rapid identification of hazardous chemicals.

-------
Squire, R.A.
Ranking Animal Carcinogens:  A Proposed Regulatory Approach
Science, 214(4523):877-880, 1931


A  method  is  proposed  for  ranking  identified  animal  carcinogens  into  five
classes  with  varying  regulatory options.  It  is assumed  that  the  carcinogens
hav3  been  identified by  testing  at  multiple   doses  and  in  at  least  two
species.   Carcinogens are scored  for  six  factors:    (1)  number of  different
species  affected, (2)  number of histogenetically different  types of  neoplasms
in  one  or more  species,  (3)  spontaneous  incidence  in  appropriate  control
groups   of   neoplasms   induced   in    treated   groups,   (4)   dose-response
relationships,  (5)  malignancy of  induced  neoplasms,  and  (6)   genotoxicity  as
measured  in a  battery of tests.  Possible total  scores  range  from 13  to 100.
Class  I  carcinogens  would have total  scores  of  86  to 100 and  the  regulatory
options would be  restriction or banning.  Class  V carcinogens  would  have total
scores less than 41 and  there would be  several  regulatory  options  including no
action,  limited  use,  labeling  requirements,  and/or  public  education  programs.
Regulatory options on Class  II, III, and IV chemicals would  fall between those
for Class  I and Class V chemicals.
                                      50

-------
Schmidt-Bleek, F.; HaberJand, W.; Klein. A.W.; Caroli, S.
Steps Towards  Environmental Hazard  Assessment of  New Chemicals  (Including  a
Hazard Ranking Scheme.  Based Upon Directive 79/831/EEC)
Chemosphere, 1KA ) :383-415,  1982


A scheme  is presented for  environmental  hazard ranking  and  bazar-5 assessment
of  chemicaJs  prior to  marketing.   Appropriate  parameters  of air,  water,  and
soil/sediment  a^e  scored and  used  to  calculate  effect  and  exposure  in  each
media.   Effect   parameters   include:   acute  fish   toxicity,   acute  Daphnia
toxicity,   oral   or   inhalation   28-day   mammalian   toxicity,   mutagenicity
screening,  and skin  sensitivity.   Exposure  parameters   include  use  patterns,
bioaccumulat i on    potential,    environmental     compartmenta] i zat •>' on,    and
persistence.   Effect  and exposure point  allocations  are  presented.    Inherent
risk for each medium  is  determined by  multiplying  effect and  exposure and then
is  standardized.   Based  on  the  inherent  risk  scores  as  compared   to  known
chemicals, chemicals are then assigned  to  one  of  three categories:  ("! ) white,
no  immediate  regulatory  interest;  (2}  gray, keep  under  consideration;  and  (3^
black,  immediate  further testing required.   A case  i"" lustration  is  presented
for DDT.
                                      51

-------
                              Author Index





                                                                       Page




 Allport, J.  et al. 1977                                                 7




 Archer, S.R. et al. 1978                                                16




 Astill, B.D. et al. 1980                                                36




*Bechmann, R.B. et al.  1974                                              23




 Becker, D.S. 1978                                                       23




*Booz Allen Applied Research,  Inc.  1973                                  39




*Branson, D.R. 1980                                                      46




 Brown, S.L. et al. 1975                                                 25




*Brown, S.L. et al. May 1978                                             29




 Brown, S.L. et al. June 19*78                                             17




*Brown, S.L. et al. 1980                                                 33




 Cairns, J. 1979                                                         42




 Cleland, J.G. 1977                                                      19




 Cramer, G.M. et al. 1978                                                41




 Dorsey, J.A. et al. 1977                                                15




 Enviro Control, Inc.  1979                                               12




 Fiksel, J. at al. 1980                                                  14




 Flinn, J.E. et al. 1974                                                  9




 Fuller, B. et al. 1976                                                  13




*Gori, G.B. 1977                                                         34




*Ingraham, T.R. et al.  1979                                              43




 Jones, C.J. 1978                                                        40




 Keith, L.H. et al. 1979                                                 44




 Kimerle, R.A. et al.  1978                                               38
                                   52

-------
122 IQI	
EPORT DOCUMENTATION  1  REPORT NO                       ?.               |3  R.c.p.en; , AC
      PAGL                EPA 560/TI I S-83-003   _.                ;          	

1CHt5?ICAL!tSELECTION METHODS•   An Annotated Bibliography,  Second :5 March°atf983
 Edition                                                             \\  -
. Doug Sellers^. Robert E,  Janney
 Reforming Organization Name and Address
 TRACOR-JITCO,  Inc.
 1776 E. Jefferson Street
 Rjockville, MD  29852
1 r,ponioring Organization Name and Adrlress

 J.  S.  Environmental Protection Agency
 401 K Street, S.  W.
 Washington, D.  C.   20460

» Supplement any Notes
 Supersedes first  edition
                                                                      I 8  Performing Organization Ren! Nc
                                                                      I
                                                                      I

                                                                      I 10. Proiect/TasV/Work Unit No

                                                                      j   05
                                                                      '• 11. ContracKC) or GrsntfG' Nc

                                                                       :> 68-01-6651
                                                                     -I
                                                                       (G)

                                                                       13.  Type of Report 4 Period Covered

                                                                          Final
  ABStraCt (Limit ?00 words)
 "arsons interested in  the control  of  toxic chemicals have  exarrined  and are exarinine
 irethods to select choricals of concern frorr the  universe of chemical  substances.
 Chemical selection refers to such  acti\dties as  priority-setting, ranl-'ine, indexing,
 and sorting.   These efforts have included the  development  of various  systematic
 selection methods, such  as scoring systems.  This annotated bibliography is intended
 provide brief  summaries  of a variety  of chemical selection methods.
'. Document Analyst*  a Descriptors
  b  Identifiers/O-pen-Ended Terms
  r COSATI Field/Grouo

I  Ay.liability Sute-irnt


 Release unlimited
                                                        19 Security Class (This Reoort)

                                                         Unclassified
                                                        70. Security Clus (Tim Ptgr!
                                                         Unclassified
Jl. No of Pa|es

    61    ..
        18'
                                                                                      M fOBU 77? f«-77
                                                                                       f NTI'j 3 *-'

-------
                            Author Index (Cont'd)




                                                                          Page




   *Kornreich, M.R.  et al.  1980                                             47




   *Langner, R.R.  et al.  1979                                                45




    Margler, L.W.  et al.  1979                                                30




   *McErlean, A.J.  et al.  1981                                               49




    Nees,  P.O. 1979                                                          37




   *Nelson, N. et  al. 1975                                                   24




   *011er, W.L.  et  al.  1980                                                 48




    Pielmeier, G.R.  1979                                                     20




   *Powers, R. et  al. 1980                                                   31




    Ross,  R.H. 1980            "                                              8




   *Schmidt-Bleek,  F. et  al.  1982                                            51




   *Squire, R.A.  1981                                                       50




    Stacey, G.S.  et  al.  1978                                                 13




    SRI 1975                                                                22




    SRI 1977                                                                21




    Stephenson,  M.E. 1977                                                    27




    TSGA Literagency Testing Committee  1977                                  11




    United Nations  Environmental Programme  1978                              32




    Van Netten,  C.  1978                                                      35




    Venezian, E.C.  1977                                                      10
* Addition to Second Revision.
                                      53  •

-------