,-„-.,,„,, k
01
i, '
; United States
Environmental Protection
/Agency
Office of Pollution
and Toxics
(7406)
Design for the En
EPA 744-S-93-001
June 1994
ironment
A a... ,
» ib 4* ,
—IS- i O ' *f . , ,
'%" r*" r
f""
Workshop on Identfedng a
Framework for the fkure of
Human Health and ||,
Environmental RisM Ranking
, , i • ' j • *. . , II, J7, 1 " ", &
l!.
Proceedings Doc
}r»»ybiHu/necyciaDle
Erifited with Soy/Canda Ink on paper that
n at >
a
nf at >««« 60% post^msumer racydedflber
» .
-------
DM mm i i si i MMHHM
I
.'af'-t . i iiiy
-in ! 'Si! > Sf4iM» « is * ailii! i Si <
: Nnln'U'liiiiLiWiMUnililill wi IliUli ! I'lOil niliiqKniiiiLVWiiililllfVI'H'dHliWIJllK
'
• :
'.
-------
EVALUATING CLEANER TECHNOLOGIES
WORKSHOP ON IDENTIFYING
A RISK RANKING FRAMEWORK
June 30-July 1,1993
Washington, DC
PROCEEDINGS DOCUMENT
Submitted to:
Office of Pollution Prevention and Toxics
U.S. Environmental Protection Agency
Washington, DC 20460
Submitted by:
Eastern Research Group, Inc.
110 Hartwell Avenue
Lexington, MA 02173
July 8,1994
-------
-------
PANEL SESSION II: METHODS FOR EVALUATION
Basic Principles of Risk Assessment .
and Related Issues
Dorothy Patton, EPA Risk Assessment Forum
PANEL SESSION HI: INSTITUTIONAL APPLICATIONS
CONTENTS
ACRONYMS _____ ........ .
................ ................... ....... vi
EXECUTIVE SUMMARY _____
............. ••' ....... ....... . .... ---- . ..vu
INTRODUCTION ... ....................... .
PANEL SESSION I: USER NEEDS ........................
Product and Process Design .......................... 4
Sill Hoffman, Motorola, Inc. ...... " ' * * " ..... '
Product Purchasing ____ . ................
Arthur Weissman, Green Seal « • • • • ........ .......... ....... 6
Risk Ranking as a Regulatoiy Tool ......... ..... ____ ..... s
Richard Kimetie, Monsanto Corporation ...... " "
Mark Rossi, Massachusetts Toxics Use Reduction Institute
Life Cycle Assessment ................ :......... ^
Frank Consoli, Scott Paper Company .......... ....... 16
t ' •"
Process Design and Substitute Assessment . . ____ 1»
Libby Parker, EPA Design for the Environment Program '•"'••"••:'."".• ....... *•*•
•> ...20
The EPS Scoring System ..... 2i
Sven-Olof Ryding, Federation of Swedish Industries and " ! '
Axel Wenblad, AB Volvo Corporation
The GW System ..... 24
Barbara Glenn, George Washington University ' ''
The Michigan Critical Materials Register 26
GaryHulburt, Michigan Department of Natural Resources '"
-------
CONTENTS (cent)
PANEL RESPONSES TO QUESTIONS • JU
ISSUES IN RANKING AND SCORING SYSTEM DEVELOPMENT ...... v 31
Gary Davis, Center for Clean Products and Clean Technologies,
University of Tennessee
33
Purpose of the System • "
leuvuuiuuvii^ ~-r— • • •' '' 33
Measures of Exposure • ; ''
Selection of Health and Environmental Impacts ... <
..33
34
Missing Data .......................... ............... * ........ '"
1/f
Aggregation and Weighting of Impacts ..... ...... ...... ..... ,.....• ..... ->
„ , . .......... ......35
Conclusion ............. ............... ..........
INTRODUCTION TO BREAKOUT SESSIONS .............. ................. 35
- ; • •' ...... ' * ' 37
BREAKOUT SESSIONS
ut Groups
Presentations on Purpose, Criteria, and Framework for a
Breakout Groups I, H, and m ..... . ........ ...... • * • ---- ..... ....... 37
Risk Ranking System .............................. .......... 3V
Discussion on Purpose, Criteria, and Framework
for a Risk Ranking System .............. ........ ... • •: ......... 3y
Breakout Groups IV, V, and VI ---- . ................................. 40
Presentations on Purpose, Criteria, and Framework for a
Risk Ranking System ............... .......... • • • ...... ; ------ 40
Discussion on Purpose, Criteria, and Framework
for a Risk Ranking System ............................... ..... 42
-11-
-------
CONTENTS (cent.)
Breakout Groups Vn, Vm, and IX ;........... 44
Presentations on Purpose, Criteria, and Framework for a ,
Risk Ranking System ......:.. . ^.................. 44
Discussion on Purpose, Criteria, and Framework
for a Risk Ranking System < 46
Next Steps: Breakout Group Recommendations . 48
Development of a Work Plan,
Validation of the System 48
Centralized Coordination, Development of a
Draft Risk Ranking Model for Review ... .... 49
Framework Development: Strategic Action Plans ..... .„ ,....... 50
Review of Existing Systems ....;.„ 52
Case Studies <: 52
Assessment of Valuation Procedures 52
Detailed Assessment Procedures .53
Clarify the Relationship Between
Sustainable Development and Pollution Prevention 53
Consensus on System Components 53
i*.
Separate Framework Development from Data Ranking .54
Data Needs 54
Training 55
Communication and Information Exchange 55
Development of a National Strategy: A Workshop Series .55
Legislative Reform , _ 55
Foster Public-Private Partnerships 57
-m-
-------
CONTENTS (cont)
CONCLUDING REMARKS 57
APPENDIX A: Workshop Agenda * A-l
APPENDIX B: List of Participants B-l
APPENDIX C: Mission Statement * C-l
APPENDIX D: Background Materials '," * ' D
APPENDIX E: Overheads from the Product and _
Process Design Presentation • • • • E-1
APPENDIX F: Overheads from the Risk Ranking as
a Regulatory Tool Presentations:
(1) Overview of Regulatory Issues
(2) State Regulatory Use >• • • F'l
APPENDIX G: Overheads from the EPS Scoring pi
System Presentations > • •'. G-l
APPENDIX H: Overheads from the Michigan Critical
Materials Register Presentation H-l
APPENDIX I: Critical Issues in the Development of
Human Health and Environmental
Risk Ranking and Scoring Systems • • > 1-1
Gary A. Davis and Sheila Jones,
Center for Clean Products and Clean Technologies
APPENDIX!: Graphic Representations of Two
Risk Ranking Frameworks ••; J-l
APPENDIX K: Health Impacts and Life Cycle Assessment ...... K-l
Anders Schmidt, Lisbeth Engel Hansen,
and Allan Astrup Jensen,
dk-TEKNIK
Jens Erik Jelnes,
Danish Technological Institute
-IV-
-------
1
CONTENTS (cont)
APPENDIX L:
Computerized Data as a Tool for
Substitution Analyses
/. Niemda, Danish Environmental Protection Agency
L. Seedotff, COWIcomidt, Denmark
APPENDIX M: PROBAS: The Danish Product Register
Data Base—A National Register of Chemical
Substances and Products
Mori-Ann Ffyvholm, Paul Andersen, Inge D. Beck"
and Nanna P. Brandoiff
Danish National Institute of Occupational Health
M-l
-v-
-------
ACRONYMS
EPA U.S. Environmental Protection Agency
EPS Environmental Priority Strategies
GW George Washington University
ISO International Organization for Standardization
LCjo lethal concentration (median-level concentration, a standard measure of toxicity)
LCA life cycle assessment
MIT Massachusetts Institute of Technology
NAS National Academy of Sciences
NGO nongovernmental organization
OECD Organization for Economic Cooperation and Development
OPPT Office of Pollution Prevention and Toxics \
SETAC Society of Environmental Toxicology and Chemistry
TRI Toxic Release Inventory
TSCA Toxic Substances Control Act
-VI-
-------
EXECUTIVE SUMMARY
Risk ranking has emerged as an important tool for achieving pollution prevention.
Although a number of different systems currently rank the relative importance of environmental
risks and impacts, a comprehensive system broad enough for use by diverse industrial sectors,
comparative risk analysts, product designers, consumers, and others involved in pollution
prevention has yet to be developed. The Workshop on Developing a Risk Ranking Framework,
held June 30 and July 1,1993, convened approximately 90 people working in pollution
prevention, life cycle assessment, design for the environment, risk assessment, chemical ranking,
and toxics substitution to discuss development of a risk ranking framework, including:
• How environmental risks associated with chemicals, products, and processes can
be best assessed. '.
• What methods exist or need to be developed to rank enviroiamental risks.
• What specific actions are needed to develop a comprehensive risk ranking system
or systems. 6 *
Characteristics of a Risk Ranking Framework
Workshop participants agreed that a risk ranking framework should have the following
characteristics:
Flexibility
Transparency
Dynamic quality
Iterative structure
Practicality
Multiple-use orientation
A risk ranking system should be flexible enough to incorporate new (data and concerns
and serve a variety of purposes. A transparent framework is one that would ensure that all steps
involved in using the system are readily identifiable, thus facilitating ease of use and system
evaluation and replication. A framework should also be dynamic so that (1) decisions can be
vu
-------
changed if necessary based on new data, and (2) the system can still be used if toxic chemical use
is reduced, product design is improved, and/or new products are introduced.
An iterative framework is one that provides companies with mechanisms for using
product assessment results as feedback for needed changes in analytical and/or valuation
methods. The system also must be a practical tool that people find useable. A risk ranking
framework must also accommodate multiple uses, including development or redesign of products;
evaluation of processes; prioritization of information; dissemination of information regarding
products, industry segments, and processes; and support for monitoring and regulation.. More
specific uses might include serving as an evaluation tool for smaller companies (with few
resources for other modes of assessment) and product labeling.
Objectives of a Risk Ranking System
Key objectives of a risk ranking system discussed at the workshop included:
• Facilitating evaluations of the potential human health and environmental risks
that product systems, materials, and human activities pose.
• Identifying areas for which more detailed analysis is warranted.
• Assessing the potential for evaluation of alternatives.
• Empowering industry to make environmental protection decisions and increase
environmental protection efforts.
• Aiding environmental decision-making at both global and local levels.
• Enabling prioritization of environmental issues.
• Incorporating adaptability for various use contexts (e.g., a single chemical with
multiple uses, a single use cluster with multiple chemical options).
• Promoting sustainable development.
ym
-------
Development of Health and Environmental Criteria
Workshop participants suggested development of criteria in several health and
environmental risk categories, including:
Human health
Ecological health
Environmental fate
Resource depletion
Energy usage
Sustainability
Energy usage
Ozone depletion
Global warming
Other indirect releases
Criteria for Assessing a Risk Ranking System
A risk ranking framework should provide a systematic method for evaluating
environmental risks, agreed many workshop participants. Criteria suggested for assessing risk
ranking methodologies included:
• Consideration of data quality and data gaps
• Reproducibility of objective results
• Sensitivity of the system (robustness)
• , Uncertainty of the system; error analysis
• Validation of the system
• Definition and measurement of success
Issues in Framework Development
Workshop participants identified several important issues in developing a risk ranking
framework. One issue was how to proceed with risk ranking if only limited data are available.
Does one proceed as if a chemical or process is highly toxic until more data are available proving
otherwise, or as if the chemical or process is environmentally benign until proven otherwise?
IX
-------
Another issue was the relationship between life cycle assessment (LCA) and comparative
risk assessment, and clarification of the particular role that each plays regarding environmental
risks. Should comparative risk assessments of chemicals and processes be expanded to
incorporate new information from LCA, or should LCA be expanded to include comparative risk
assessment?
Participants also raised the issue of how a risk ranking framework and health and
environmental criteria should be evaluated. Many agreed that evaluation of a ranking system
should focus on how successfully it can be used as a tool for making decisions rather than on
how strictly it adheres to a specific ranking methodology. In addition, while all participants
agreed that health and environmental criteria must be evaluated, some suggested that political
and socioeconomic factors also must be considered. Another issue was how a risk ranking
system could be developed for diverse purposes such as screening, detailed analysis, and
decisionmaking.
Possible Risk Ranking Frameworks
Participants offered two possible frameworks for a risk ranking system. The first
framework focused on assessment of risks and recommended clearly separating the objective (i.e.,
data analysis) from the subjective (Le., ranking) aspects of the framework. Because various
stakeholders have different values, this approach suggests that scientific data be used to develop
assessment techniques rather than actual ranked lists. Assessment of risks should first occur
within the same categories of impact (e.g., human health, ecological, resource depletion, or
social/economic) prior to further evaluation.
The second framework focused on environmental valuation, which involves weighing the
relative importance of risks or impacts based on the assessments conducted. While the
assessments provide the decisionmaker with a better understanding of the implications of a
potential action (e.g., development of a new product, choice of a new material or process,
promulgation of a new regulation), valuation explicitly involves setting priorities. Like the
x
-------
assessment approach described above, the valuation approach also clearly distinguishes between
environmental analysis and environmental ranking.
Strategies for Framework Development
Workshop participants developed three general strategies for developing a risk ranking
framework:
The first strategy involved development of a "strawman" framework by either EPA
workgroups or another small group established by the steering committee. Other
workshop participants would then review and critique this strawman framework
The enure group, which might be expanded to include additional members, could
meet again to finalize individual parts of the framework and consolidate it A
SUggCSted to develop and Produce a definitive, formalized
The second strategy for framework development recommended that the steering
committee identify areas needing further work and form small workgroups
composed of people with a particular interest in these areas. The steerine
committee would need to determine how to separate the issues into workable
pieces for each of the groups. The entire group could then come together for
presentations and discussion of the different areas and to perform further work.
.nV°lved brfnging devel°Pers ^d users of risk ranking systems
together and applying an already existing system in a hypothetical situation The
advantage of using an existing system would be that people could readily see
where additions work is needed and how a system should look when finished; this
Sdl^h1^^
could then begin developing the weaker areas identified. Case studies could be
developed using an existing or strawman system for practical application.
Workshop participants agreed that different parts of these three strategies could be
combined to determine more specifically what the next steps should be. Unification of language
and definitions could occur as part of the framework development effort.
XI
-------
Identification of Next Steps
A key purpose of the workshop was to identity "next steps" needed toward development of a
risk ranking system. There .was consensus on four steps:
• Disseminate information on existing tools and ranking systems
• Unify language/definitions for risk ranking
• Expand the steering committee
• Develop a draft risk ranking framework
Some participants suggested that EPA or another organization should establish a central
coordinating center to administer development of risk ranking frameworks. Such a center could
also .serve as a liaison between workgroups (e.g., groups focusing on risk assessment, lifecycle
analysis, or design for the environment). Another important endeavor identified was further
development of methodologies for assessing and ranking risks and impacts.
xn
-------
INTRODUCTION
Mark Greenwood, director of the U.S. Environmental Protection Agency's (EPA's) Office
of Pollution Prevention and Toxics, opened the meeting, Evaluating Cleaner Technologies:
Workshop on Identifying a Risk Ranking Framework, by noting that EPA's work in the area of
pollution prevention has matured over time into a more fundamental examination of how
pollution prevention is carried out and how interested parties can collaborate to improve
pollution prevention efforts. An important issue involves how environment considerations can
be factored into the design of manufacturing technologies, processes, and products. The design
for the environment ethic is not only an EPA program, but a broader effort that many people are
attempting to practice in a meaningful way.
To implement design for the environment through pollution prevention, fundamental
values must be defined. A person or a company must decide which issues are most important
from an environmental point of view, thus establishing "environmental value points." To this end,
EPA has focused on, among other things, how to prioritize and compare risks and how to better
analyze these issues.
EPA began developing tools for comparative risk analysis to set priorities under the Toxic
Substances Control Act (TSCA), and sought and received feedback on the usefulness of these
tools. During the process of sharing this information, the Agency found that many others also
were trying to develop systems and analyses for prioritization and comparative risk ranking
While the purposes of these systems might differ, the basic goal is the same: to sort through a
sea of information and attempt to develop relative values regarding what issue or effect is more
important than another.
This workshop on developing a risk ranking framework brought together many people
from industry, government, academia, and environmental organizations working in similar areas
Goals of the workshop included the following:
• Identify key components of existing relative risk ranking systems.
• Discuss uses of such systems.
-1-
-------
• Specify actions needed to develop risk ranking systems further and identify
existing work that can contribute to developing a comprehensive risk ranking
system or systems.
The purpose of the workshop was not to produce a master system that everyone could use, since
development of a final, comprehensive system will take time. Rather, a key goal of the workshop
was to encourage a coordination of international efforts in developing a useful system or systems
for risk ranking across industries.
Workshop facilitator Abby Arnold, environmental dispute mediator at Resolve, reviewed
the agenda for the workshop, which included panel discussions on user needs, methods for
evaluation, and institutional applications; a presentation on issues in the development of ranking
and scoring systems; facilitated discussion sessions; and breakout group sessions. The purpose
of the breakout sessions was to involve workshop participants with diverse interests in the process
of developing a risk ranking system by addressing several key questions, including:
• When conducting an environmental evaluation of technologies and products, what
criteria should be included in a ranking system?
• What work needs to be done to develop such criteria?
• What are some of the next steps that individuals, organizations, and companies
take to develop a ranking system both nationally and internationally/
can
• Is one system appropriate, or is a series of different ranking systems needed?
After the breakout sessions, individual groups reported back to the entire workshop
forum with their deliberations regarding these and other key questions that arose during the
workshop. The entire group then discussed the breakout group reports and identified possible
next steps. This proceedings document summarizes the workshop activities, focusing on
discussions generated from the breakout groups' deliberations and on appropriate next steps
identified by workshop participants to develop a framework for environmental risk ranking and
impact assessment.
-2-
-------
The document's appendices include additional relevant information, including the
workshop agenda, a list of workshop participants, a mission statement, background materials,
graphic representations of two possible risk ranking frameworks, overheads! used by some of'the
speakers in their presentations, and published papers describing various risk ranking systems.
PANEL SESSION I: USER NEEDS
Bob Ferrone of Digital Equipment Corporation began the panel discussion by providing
background on the origins of the workshop. Mr. Ferrone and Alex Wenblad of AB Volvo
Corporation sowed the seed for this workshop at a presentation Mr. Wenblad gave at the
Massachusetts Institute of Technology (MIT) on mercury in lighting. Mr. Ferrone, noting that
the presentation included information on the Swedish approach to a risk scoring system, said that
bringing together an international group on the subject would be productive, and Mr. Wenblad
agreed. Mr. Ferrone then talked with Claudia O'Brien and Mark Greenwood of EPA about
involving the Agency in convening such a group to explore the issues of environment and
technology and the role of risk ranking systems.
Fran Irwin of the World Wildlife Fund, the panel moderator, said that over the past 15
years a number of workshop participants have been involved in developing ways to evaluate
cleaner technologies, including ranking systems. The new element in the workshop, she said, was
that a demand for better tools for evaluating technologies is emerging, as indicated by the
response to this workshop. People using these evaluation tools are asking how to design and
produce products in an environmentally sound way and how to label products so that customers,
whether institutions or individuals, can make better choices. In traditional manufacturing, by
contrast, manufacturers focus on which chemicals and products to produce and use outside of an
environmental context. The panel members addressed three questions, including:
• How are questions about evaluating cleaner technologies framed from different
perspectives (e.g., a product or process designer, an ecolabeling group or
consumer buying a product, a regulator or group promoting regulations)?
technokT .^te'°f'the-art methods
-3-
information tools for evaluating cleaner
-------
• How can technologies be evaluated better?
Each panelist's presentation concerning user needs is summarized below.
Product and Process Design
Bill Hoffman of Motorola, Inc., said that while he could not speak for all industry, or
even for Motorola, he would give his own views on developing a framework for a risk scoring
system for material and product use and development. Overheads from his presentations are
provided in Appendix E. First, Mr. Hoffman described briefly how products generally are
designed. Currently, many products are designed by multifunctional work teams that include
people from all the different aspects of a product's life cycle: mechanical design, electrical design,
finance, production, manufacturing, perhaps maintenance, as well as others. Hus approach,
however, works only in large companies with sufficient resources for such a team approach. In
small companies, a multifunctional individual might have to cover all of these different areas
himself or herself. In such cases, knowledge of environmental issues may be minimal, since small
companies typically lack an environmental professional on the team. The point person in the
smaller company therefore needs an "intelligent advisor," which, for example, could be a
software system that scores a product's environmental risk and helps identify areas that need
improvement.
The three primary industry users of a risk rating system include the design engineer, the
environmental and safety professional, and management, each with somewhat different needs.
Most of these people do not have time to add a new dimension, such as an in-depth evaluation
of environmental impacts, to their product development cycle or to their duties. Instead of
spending time learning a whole new system in these different areas, the designer wants to know
what the environmentally preferred materials are. Rather than be told in a general manner that
he or she should recycle and use less hazardous chemicals, the designer wants to know which
chemical to use and how to design for demanufacture, recycling, and disassembly. Hie designer
also wants to know how altering these processes for environmental reasons would affect the
quality of a product and what the tradeoffs would be. Will a recyclable material pass the drop
-4-
-------
also wants to know how altering these processes for environmental reasons would affect the
quality of a product and what the tradeoffs would be. Will a recyclable material pass the drop
test? Will it function properly? How much will it cost? How much wiU it add to the design
cycle time? A common assumption is that when environmental concerns are included in the
manufacturing process, costs increase, but this assumption is not accurate. Costs can be reduced
because environmental concerns are included (e.g., waste minimization). For example, the
current cost of recycled paper (i.e., deinked rather than bleached paper) could prohibit people
from using it. If enough of it is used, however, the market can be altered so that the recycled
paper will be produced in larger mills, thereby reducing the cost.
The environmental and safety professional* another primary user of risk ranking systems,
traditionally has been involved in product design, focusing on a product's potential effects on
waste disposal and safety conditions. An environmental and safety professional's questions might
include: What are the EPA guidelines? What are the rules, regulations, and restrictions on
using this material? How should it be handled? Does it require special equipment or facilities?
How should it be disposed of? If it is toxic, can it be incinerated? Landfilled? Put down the
drain? What are the cleanup procedures and costs if the material is spilled? A relatively new
question from these professionals is: What are the life cycles and environmental impacts?
Management, the third primary user of risk ranking systems, is concerned with cost.
Management's questions focus on issues of marketing and profit: Can we produce an
environmentally sound product at the same costs? Will it sell as an environmental product? If it
costs more to make this a low-impact product, will it still sell? If it does sell, what are possible
future liabilities?
Currently, the environmental and safety professional is more concerned with the
environmental impact of a product than the other two primary users of risk ranking systems. As
environmental issues become more important, however, designers, too, will want to improve their
products and will need a measure of environmental impact. A risk scoring system can provide a
place to start, a reference point, and can help designers set goals for improving their products.
-5-
-------
A ranking scheme should be user friendly, yet comprehensive. It must be easy to
understand. A designer does not want to have to perform a complicated analysis each time a
small change is made in a product; the general rules and principles should be easy to remember
and apply in conducting analyses to identify areas that need improvement. Yet the system also
must be complex enough to represent the full impact; a system that is too simple will not capture
the interactions between manufacturing processes, design, and environmental impacts throughout
a product's entire life cycle. In addition, the system must be able to impart general principles
that can be applied to most situations.
To incorporate these life cycle issues in a risk ranking system, the most ambitious projects
would require data about every aspect of a product's life cycle, which is problematic because
often the manufacture of a product is not a closed system. A product's life cycle could affect
different types of industries and other products worldwide. Thus, in practice, the scope of a life
cycle impact assessment often needs to be limited. The limits of what to examine typically have
been set at the factory door (i.e., restricted to the manufacturer's direct contact with the
product). But because the heaviest impacts often occur beyond the factory door, issues such as
recycling and the toxicity of the chemicals being used also are now considered. By limiting the
scope of the analysis, a designer can understand the impacts directly associated with a product's
design. ;
Product Purchasing
Arthur Weissman of Green Seal discussed the choices people make in purchasing
products. Some standards for environmentally preferable products have been developed and
made public and more continue to be developed. An important question is, are the right
standards being set in terms of overall impact on the environment (i.e., standards for products
that have the most impact)? A ranking system would help focus on areas with the most risk, but
we must remember that environmental risk and impact are only one factor in purchasing a
product. Companies and governments must consider many other factors in their decision-making
process, such as price, performance, and other factors (e.g., aesthetics). A ranking system is only
a tool and is not in itself a decision model.
-6-
-------
A ranking system needs to be understandable by its various users. Pilot tests should be
conducted to explore whether a system accommodates the end user. System users can include
consumers (i.e., the general public), who have been exposed to confusing environmental claims
for products. Most consumers seek to reduce the environmental impact of their purchases.
According to recent market research and other studies, environmentally conscious consumers
comprise up to 80 percent of the American public. While this percentage does not reflect
possible higher costs for environmentally preferable products, the percentage drops only slightly
when higher cost is considered. Another user is the purchaser in the marketplace, such as a
materials purchaser for an organization or a government body with a procurement plan (such as
the White House, which currently is developing a "green" procurement plan). A third category of
users is the manufacturers who want to make an environmentally preferable product. These
companies also are procurers; they look for suppliers that offer environmentally preferable
products, technologies, or materials. A final group of users includes national and international
organizations that perform labeling of environmentally preferable products in an attempt to
reduce the environmental impact of products through market forces.
Each user group wants certain information. The general public wants to know which
products in a given category (e.g., bags, automobiles, cleaners) have the; least environmental
impact. The more educated consumers might want to know what specific impacts are being
reduced. Purchasers tend to want similar information as consumers. Manufacturers have a
range of information needs. For a product that is not integral to their line, manufacturers, like
the public and purchasers, want to know generally which product has less impact than another.
But for a product that is their own or is integral to their product line, manufacturers want to
know what specific environmental impacts and potential liabilities are atssociated with that
product or material. Labeling organizations want to know all of this information: which products
in the marketplace generally have the greatest environmental impacts, which products also exhibit
a range of impacts suitable for environmental labeling within a particular product line, and what
the specific impacts are.
Little information is available about the life cycles of chemicals and products, the uses
and misuses of products, and product disposal. While frameworks for this information, such as
risk assessment and life cycle assessment (LCA) have been used for approximately 15 years, they
-7-
-------
are very controversial, and LCA is quite preliminary. Even less integrated information is
available about the combined impacts of products or technologies. While some information is
available about specific chemical ingredients, emissions from various manufacturing processes,
and impacts from resource extraction or disposal activities, such data for a given material or
product are not integrated or ranked, and their environmental impact is not well studied.
Environmental information is not readily available in a form that allows for comparisons.
In'the short term, basic data needed for a risk ranking system must be collected, such as
information about energy and material flows and emissions. These data must be provided in a
readily usable form for all end users with varied needs. Consumers want a means to identify and
avoid high-impact products, with the aid of consumer education. Manufacturers want specific
quantitative information about the relative impacts of their materials and technologies. Labeling
organizations want quantitative information about the relative impacts of materials, technologies,
and products.
Risk Ranking as a Regulatory Tool
Richard Kimerle of Monsanto Corporation said that risk ranking, in addition to its other
functions, can be used to guide regulatory priority setting and decision-making. A key element in
the proposed Environmental Risk Reduction Act of 1993, risk ranking can play an important role
in many regulatory guidance and compliance programs. It can provide a means for listing and
delisting chemicals; help in prioritizing which chemicals need additional testing and regulatory
consideration; and contribute to risk assessments, product life cycle assessments, and product
labeling efforts. [
Criteria in a risk ranking system should be risk based and should include exposure as well
as effects data, said Mr. Kimerle. Currently, most risk criteria focus on chemicals, toxicity data,
threshold levels, fate of pollutants, and exposure, all of which are important. But additional
criteria need to be included, such as data on environmental risks and effects other than those
associated with chemicals (e.g., other types of stressors) as well as energy, natural resource,
global, and habitat considerations.
-8-
-------
Risk ranking can help justify decisions to reduce the use of toxic substances and
exposures to environmental hazards. Use of a risk ranking system, however, might raise certain
issues. For example, if a comprehensive risk assessment is not performed, should the use of a
risk ranking system be considered as only a screening tool? Should quantitative or qualitative
(i.e., high, medium, low) rankings be used? How can system users be instiucted to apply the
system properly to avoid misuse? Must a minimum data base first be developed before a risk
ranking system can be used?
To be used for regulatory purposes, a single technical framework needs to be developed
that is scientifically sound and is consistent across EPA offices and state, federal, and
international boundaries. A window of opportunity currently exists to establish such a risk
ranking system that can provide a new basis for regulatory decisions, priorities, guidance, and
compliance. Overheads from Mr. Kmerle's presentation are provided In Appendix F.
Mark Rossi of the Toxic Use Reduction Institute then discussed how states might use a
relative risk ranking system, using the Massachusetts Toxic Use Reduction program as an
example. Appendix F also includes overheads from Mr. Rossi's presentation. States can use a
risk ranking system for several purposes, he said, including identifying "user segments" (i.e.,
production units targeted for toxics use reduction), prioritizing resource allocations, and selecting
chemicals as candidates for restriction. For these purposes, a useful risk ranking system should
allow for broad participation and account for both toxic chemical use and emissions.
Factors to consider in identifying and prioritizing user segments include:
• The amounts and toxicity of toxic substances used by user segments.
• The amount of toxic substances disposed of, discharged, or released to water
land, air, or workplaces by users. '
• The social, health, and economic benefits and costs of targeting particular user
segments.
In setting resource priorities, a risk ranking system could assist in allocating funds and
personnel in research, education, and technical assistance efforts. Such resource priorities
-------
currently are based on legal requirements, amounts of chemical releases and transfers, or
regional needs. Priority often has been given to work involving alternatives to ozone-depleting
chemicals. While all these considerations are important, resource allocation should also be based
on a better understanding of the comparative risks of using toxic chemicals.
Chemical restriction refers to any policy that encourages or directs a company to
eliminate production, use, or distribution of a chemical (or class of chemicals) or of a product
that contains the chemical. To use a risk ranking system for chemical restriction purposes, states
would need to know how a screening tool, such as a risk ranking system, could be tailored to
regional needs.
The Massachusetts Toxics Use Reduction program mandates that the state identify users
of toxic substances^ The program includes an administrative council and an advisory board. The
council is responsible for identifying priority user segments based on recommendations from the
Massachusetts Department of Environmental Protection (DEP) and the Office of Technical
Assistance (both state agencies), as well as from the Massachusetts Toxics Use Reduction
Institute, a private organization which includes both industry and public interest advisory boards.
The council includes representatives from state agencies such as the DEP, the Department of
Public Health, and the Office of Economic Affairs. The state advisory board provides a forum
for discussing the direction of the Toxic Use Reduction program and develops recommendations
for the program. Fifteen representatives from business, government, environmental groups,
labor, health policy groups, publicly owned treatment works (POTWs), and the general public
comprise the state advisory board.
PANEL SESSION II: METHODS FOR EVALUATION
Frank Field of the Massachusetts Institute of Technology, moderator for Panel Session II,
summarized the three types of methodologies currently available for developing risk ranking
systems—risk assessment, life cycle assessment (LCA), and process design/substitute assessment.
Dr. Field said that these methods can be characterized according to three broad professional
-10-
-------
areas: risk assessment is based on scientific study, LCA originated within the engineering •
discipline, and process design/substitute assessment arose from design and management concepts.
Risk assessment, based on biological science and statistics, quantifies dose and associated
response characteristics of biological populations exposed to a variety of environmental stimuli.
Scientists have developed general principles of risk assessment that are used to predict the
behavior of biological systems. An extensive risk assessment often includes a quantitative
description of a complex system (e.g., the environment). Using this quantitative description, a
risk assessor attempts to define the extent of hazard represented by a change in the system so
that appropriate action can be taken to protect the system if necessary,; Scientists have raised
serious questions, however, about the appropriateness of applying atomistic scientific methods,
such as risk assessments addressing a single hazard, to large complex systems in which synergy'
between individual effects may be occurring. Risk assessments generally do not provide guidance
regarding appropriate actions when several hazards are present simultaneously.
LCA is designed to classify and quantify the flow of resources associated with an
industrial activity. LCA extends the engineer's scope by considering source flows throughout a
product's lifetime, both upstream (i.e., resources consumed) and downstream (i.e., wastes
generated) when evaluating a process. Using LCA, processes that reap local benefit but are
globally harmful (or vice versa) can be identified, and alternatives can be devised. The
magnitude of data generated by even* simple LCA, however, can be daunting. Guidelines are
needed to reduce these data to a manageable set of indicators of performance. Also, LCA
generally does not provide decision-makers with enough information to choose alternative
processes reflecting a variety of changes in environmental effects. Developers of an important
component of LCA, called impact assessment, recommend using other impact assessment tools
(e.g, human and ecological risk assessment) in conjunction with LCA to help assess and develop
environmental improvement opportunities.
Process design/substitute assessment, unlike risk assessment and LCA, does address
tradeoffs among competing alternatives. This method reflects how designers and managers
develop modes of action in the face of competing objectives and conflicting and uncertain
information. Process design/substitute assessment focuses on the development of "use clusters"
-11-
-------
within an industrial process, elements of which can be used interchangeably to accomplish a
specific task. Process design/substitute assessment uses some LCA principles, identifies particular
indicators of performance and cost for each member of a use cluster, and ranks the members
according to their potential for environmental risk. Such a ranking, however, reflects the
strategic objectives of the participants; such objectives may conflict with one another among
different stakeholders. In addition, this method does not outline which strategy is most
appropriate.
All three of these analytical methods are powerful tools, but none provides a
comprehensive method for environmental ranking. Each method relies on information .outside of
its field of expertise to address the issue of valuation and ranking of risks, lliis issue is further
complicated when interested stakeholders have different concerns. In the end, analytical
methodologies alone, such as those described here, cannot resolve the problem of environmental
risk and impact ranking. Procedural methods-how the information developed in analytical
methods is applied-are at least as important an element in the development of risk valuation
methods as analytical methods. To a large extent, the compatibility of analytical methods with
the institutions that will be making procedural decisions will determine the success of risk
ranking methods. ,
Panel members then described the risk assessment, LCA, and process design/substitute
assessment methodologies in more detail, as presented below, addressing the following questions:
• What is the methodology (i.e., what problem does it address, what is it based on,
what is not addressed)?
• How is the method used (i.e., what are the data and analytical requirements, what
expertise is needed)?
• What are the strengths and weaknesses of using the methodology to evaluate
environmental performance?
-12-
-------
Basic Principles of Risk Assessment and Related Issues
Dorothy Patton ofEPA's Risk Assessment Forum said that risk can be addressed in many
ways, and that the paradigm in which risk is described must be clearly identified. She presented
EPA's approach for addressing risk and discussed issues that should be considered in conducting
a risk assessment.
In 1983, the National Academy of Sciences (NAS) provided a definition of risk
assessment which EPA has used to develop risk assessment guidelines and for other purposes.
The NAS defined risk assessment in terms of characterizing adverse effects of human exposure to
environmental hazards. While this assessment of risk is scientific, NAS also recognized
nonscientific issues associated with risk, including analysis of perceived risk, comparisons of risk,
the economic and social implications of regulatory actions, control options, and technological
feasibility. The NAS defined these nonscientific areas as risk management, which is part of the
overall process in considering risk.
Over the past 2 years, EPA has been working on a comparable paradigm for ecological
effects. While differing somewhat from the human risk assessment paradigm, the ecological
approach uses the same basic process for characterizing chemical or nonchemical stresses or
exposures. Other methodologies, however, might define these issues differently.
The NAS system focuses on information derived from scientific studies in the laboratory
and the field. In the risk assessment process, information is collected from many scientific
disciplines-biology, chemistry, geology, statistics, modeling, and science policy-to produce a
risk characterization that is used with relevant nonscientific information to reach decisions on
environmental hazard and control options.
This risk assessment system incorporates information on exposure and effects
Information on effects is obtained from laboratory work as well as from human studies, usually
occupational exposure studies. These studies define effects such as cancer, birth defects and
neurological effects; examine whether an association exists between effects and exposure to a,
-13-
-------
particular chemical; and identify the level at which those effects occur. Some uncertainty,
however, always is associated with such information. .
Information on environmental exposure involves studying populations (human and other
systems) that might be exposed to chemicals at levels of concern. Since animal and
epidemiologic studies often are based on populations exposed to very high levels, the actual
exposure level people experience in everyday life should be compared carefully with the high
exposure levels of these studies, and the known combination of effects and exposure levels should
be examined. Environmental exposures may be low and the level at which effects occur high. In
this case there is little basis for concern. In other situations, however, exposure might exceed
the level at which exposures are known to cause effects in well-studied populations (e.g, animals,
occupational exposure). While few would dispute that a risk exists in that case, a debate might
arise regarding how to control the risk. ;
When the levels of everyday exposure and effect are close to the known levels that might
result in risk, determining just how close the exposed population is to the risk levels is difficult.
In these cases, the uncertainties typically result in controversy regarding what the risk really is.
Often a debate ensues about what to do given the information available, and more study is called
for.
While this risk assessment methodology examines hazard in terms of controlled animal
and human studies with a known combination of exposed organisms and effects (or a lack of
effects), it also studies exposure levels for the general population, including ambient levels,
exposure through diet, and the duration of exposure. All this information, including
uncertainties and assumptions in the analysis, is combined to characterize the risk.
Policy also is a major factor in risk assessment. While not part of the original NAS
paradigm, policy involves making choices regarding how to handle uncertainties in the risk
assessment process. The policy choice that people are most keenly aware of involves the use of
animal data to provide information about humans. EPA, other regulatory agencies, and many
others in the scientific community use information from animal studies to make assumptions
-14-
-------
regarding exposed humans in comparable circumstances. Others, however, do not believe that
animal studies should be used to make predictions about human health.
Another policy choice in the risk assessment process is which pDpulation to study. If the
average population is characterized, the assessment will include a larger number of people but a
smaller risk. If higher risks are studied, a smaller number of people will be characterized.
Should studies focus on special subpopulations, such as children, the elderly, or the
disadvantaged? Such decisions, however, often are not identified as policy issues in the risk
assessment process.
To summarize, the risk assessment process involves two areas. The first area is scientific
information (e.g., human studies, animal .studies, studies of subcellular systems, genetic studies
metabolic studies, chemistry, and data on ground-water movement) used to describe effects,
potential exposure, and the relationship between them. EPA guidelines and the NAS process
call for gathering all this information and analyzing the total situation. In describing risk, both
qualitative and quantitative information is used, and there is always uncertainty. The second
area is policy: the labeling of assumptions and policy choices involved in the risk assessment
process.
The source of the risk assessment information also is important. An assessment should
be based on all relevant scientific disciplines and all available studies on a particular topic (e g
not just one study of the LC50 of a chemical or one study on tumors in one population) The '
source of information also includes the perspective from which the risk assessment has been
conducted (e.g., a regulatory agency, industry, environmental group, or academia). Another
factor is the scope of the risk assessment in terms of expertise and cost. A 10-year, multibillion-
dollar dioxm study, for example, is not comparable to a 3-day monitoring study at a Superfund
site.
Finally, for any ranking system, risk assessment terminology must be defined because of
the many different meanings and uses of the term risk. The elements involved in risk perception
assessment, management, and ranking must be clearly, described.
-15-
-------
Life Cycle Assessment ;
Frank Consoli of Scott Paper Company said that life cycle assessment (LCA) originated
as an engineering approach to classify and quantify the flows of resources associated with an
industrial activity. Life cycle assessment is a cradle-to-grave system encompassing raw material
acquisition through impact of a product. For bath tissue, for example, LCA would include an
analysis of harvesting and forestry practices, manufacturing processes, distribution and
transportation, use and maintenance, reuse, and waste management (e.g, recycling, composting).
LCA can be viewed as a three-step process. The first step in LCA is the inventory stage,
in which the environmental burdens associated with a system are evaluated by identifying and
quantifying, where possible, energy and material usage and releases throughout the entire life
cycle. Approximately 90 to 95 percent of people conducting LCAs agree on most of the
methodologies available for this type of evaluation.
The second stage of LCA is the impact assessment, which involves using the information
from the first stage to assess the overall environmental impact of the materials and releases. The
third stage of LCA can be thought of as the improvement stage; the goal is to enhance
opportunities for environmental improvement.
LCA began in the early 1970s as an inventory methodology, which then expanded to
include impact assessment and improvement. Many companies and other organizations now try
to use LCA in a variety of ways. Attempts to evaluate products through LCA must begin with a
systems analysis. A simple industrial system analysis includes a definition of the system itself
(e.g., what is occurring in the system), the system environment, and the boundaries of the system.
The goal of the LCA inventory phase is to examine a system's inputs and outputs across
boundaries. Relevant inputs typically include raw materials and energy, and outputs include
emissions, effluents, and the product itself.
When first developed, the LCA impact assessment method involved simply adding up the
various elements; many people believe, however, that this method is not scientifically valid. The
sheer magnitude of data generated by even the simplest of these evaluations can be .daunting,
-16-
-------
and guidelines must be developed to reduce these data to a manageable set of performance
mdrcators. Current approaches under development involve defining types of impacts and impact
categories and identifying areas that are not included in the assessment. In this way, a
framework for impact assessment is set and the scope of the assessment is limited to a workable
number of measurable categories of major impacts. The three main impact categories identified
are:
• Human health
• Ecological health
• Resource depletion
Social impacts (e.g., the creation of towns, roads, and other infrastructure to support a product's
manufacture) also may be an important category in some systems.
A process known as characterization can translate the information from the inventory
phase (e.g., emission types in pounds) into impact areas. Risk assessment or other scientific
approaches can assist in impact categorization and analysis. Details of the characterization
process are beyond the scope of this discussion.
Once impacts are known, a valuation process is critical in deciding what actions to take
Because values might change over time and with a different set of people, the focus should be on
defining a valuation method, rather than values themselves, using decision-making theories and
processes.
Currently, the goals, inventory stage, and first phase of classification involved in impact
assessment are fairly well defined. Other aspects of impact assessment have been defined only
conceptually, and improvement analysis has not yet been documented. Tfce Society of
Emaronmental Toxicology and Chemistry (SETAC) and others are working to standardize life
cycle protocols.
-17-
-------
LCA is more than a means to make broad, public marketing claims, according to Mr.
ConsolL Life cycle assessment has the power to change the way product design is viewed
conceptually.
Process Design and Substitute Assessment
libby Parker, ofEPA's Design for the Environment Program, discussed the background
of the program. A few years ago, BPA's Office of Pollution Prevention and Toxics (OPPT)
streamled its program for easting chemicals to focus resources earner in the research process
on chemicals with higher risk. This effort integrated all the disciplines and areas needed to
identify and evaluate risk management options-factors s«ch as risk assessment, performance,
and cost The result was a means to screen and organize information about a single chenucal
across different use categories. Tnis is important because risk management options and exposure
scenarios both vary depending on use category.
The next step was to focus on alternatives within a use category. Ihis effort was in part a
response to industry "leapfrogging- from one chemical to another. That is, when one chenncal
becomes regulated, industry typically substitutes a different chemical, resulting in an inefficient
way to expend society's resources. Thus, from a regulatory point of view, EPA began.developmg
a way to examine substitute chemicals a, the beginning of an analysis and to determine whether
industry should be encouraged to use a different chemical or, when a less toxic substitute is no,
available, to manage a particular risk better. This method, known as the m ctoer m-fc* is a
way to organize complex information, no, necessarily to provide definitive answers; the valuafon
process (e.g., cost-benefit analysis) must follow this screening method.
BPA has been examining multimedia risk issues in a regulatory context for some time.
These issues are complex, and the tradeoffs often are difficult to sort out (e.g., between human
heal* and ecological risks). One of OPPTs goals is to provide public access to risk informauon
by working with various industry groups using EPA's methodology to elucidate tradeoffs under
specific circumstances. The use cluster method is based on the assumption that a set of
-18-
-------
Richard Leukroth
Biologistfioxicologist
U.S. Environmental Protection Agency
401 M Street, SW
Washington, DC 20460
202-260-7319
Fax: 202-260-1216
Steven Lewis
Toxicology Associate
Exxon Biomedical Sciences, Inc.
MettlersRoad(CN-2350)
East Millstone, NJ 08875-2350
908-873-6063
Fax:908-873-6009
Kathryn Lindquist
Director of Marketing
Taylor-DeJongh, Inc.
2000 M Street, NW - 7th Floor
Washington, DC 20036
202-775-0899
Fax:202-775-1668
John Markham
Senior Program Officer
International Development
Research Centre
250 Albert Street
P.O. Box 8500
Ottawa, Ontario K1G3H9
Canada
613-236-6163
Fax: 613-567-7748
James McCarron
Director
TNT Technology Company
2121 West University Drive
Suite 123
Tempe, AZ 85281
602-966-9891
Fax: 602-968-9469
Douglas McTavish
Director,, Great Lakes Regional Office
InternationalJoint Commission
100 Ouellette Avenue
8th Floor ,
Windsor, Ontario N9A6T3
Canada
519-257-6715
Fax:519-257-6740
Jed Meline
Chemical Engineer
Economic, Exposure and
Technology Division
Design for the Environment Program
U.S. Environmental Protection Agency
401 M Street, SW(TS-779)
Washington, DC 20460
202-260-0695
Fax:202-260-0981
Harry Milman
Senior Science Advisor
U.S. Environmental Protection Agency
401 M Street, SW (TS-796)
Washington, DC 20460
202-260-1292
Fax: 301-871-5586
Bruce Mintz
Chief, Exposure Assessment &
Environmental Fate Section
Human Risk Assessment Branch
U S. Environmental Protection Agency
401 M Street, SW(WH-586)
Washington, DC 20460
202-260-9569
Fax: 202-260-5394
Victor Morgenroth
Principle Administrator
Environmental Health and
Safety Division
Organisation of Economic
Co-Operation and Development
2 Rue Andre-Pascal
75775 Cedex 16
Paris, France
33-1-4524-9775
Fax:33-1-4524-1675
Cheryl Morton
Manager, Government Relations
SOCMA
1330 Connecticut Avenue, NW
Suite 2IOO
Washington, DC 20036
202-822-6758
Fax: 202-659-1699
Claudia O'Brien
Design for the Environment Program
Office of Pollution
Prevention and Toxics
U.S. Environmental Protection Agency
401 M Street, SW (TS-779)
Washington, DC 20460
202-250-0688
Fax:202-260-0981
Deborah Oberst
Director, Product Stewardship
Environmental Health and Safety
General Ellectric
One Plastics Avenue
Pittsfield, MA 01201
413-448-4827
Fax: 413-448-6550
Libby Parker
Chief
Design for the Environment Program
U.S. Environmental Protection Agency
401 M Street, SW (TS-779)
Washington, DC 20460
202-260-0686
Fax: 202-260-0981
Lynne Patenaude
Proji2Ct Engineer
Solid Waste Management Division
Environment Canada
Ottawa, Ontario K1AOH3
Canada
819-997-3718
Fax:819-953-6881
B-4
-------
Stephen Greene
Corporate Environmental Manager
Polaroid Corporation
575 Technology Square - 1A
Cambridge, MA 02139
617-577-4106
Fax: 617-577-2052
Mark Greenwood
Director, Office of Pollution
Prevention and Toxic Substances
U.S. Environmental Protection Agency
401 M Street, SW(TS-779)
Washington, DC 20460
202-260-3810
Fax:202-260-0981
Wayne Hansen
Group Leader
Environmental Science Group
Los Alamos National Laboratory
P.O. Box 1663 (MS-J495)
Los Alamos, NM 87545
505-667-3331
Fax: 505-665-3866
Trisha Hasch
Conference Manager
Eastern Research Group, Inc.
110 Hartwell Avenue
Lexington, MA 02173-3198
617-674-7321
Fax: 617-674-2906
Bill Hoffman
Project Manager
Corporate Manufacturing
Research Center
Motorola, Inc.
1301 East Algonquin Road
Room 1014
Schaumburg, IL 60196
708-576-7739
Fax:708-576-2111
Yoshitaka Hoshikawa
Director
Japan Chemical Industry Association
3-2-6 Kasumigaseki
Chiyoda-ku
Tokyo, Japan
81-3-3580-1381
Fax:81-3-3580-1383
Gary Hurlburt
Toxicologist
Surface Water Quality Division
Michigan Department of
Natural Resources
P.O. Box 30273
Lansing, Ml 48909
517-335-3312
Fax: 517-373-9958
Fran Irwin
World Wildlife Fund
1250 24th Street, NW
Washington, DC 20037
202-778-9646
Fax: 202-293-9345
Michael Jayjock
Rohm and Haas Company
727 Morristown Road
Spring House, PA 19477
215-641-7480
Fax: 215-283-2554
Allan Astrup Jensen
Head of Department
Department of Environmental
Impact Assessment
dk-Teknik
15 Gladsaxe Mollevej
DK-2860
Soborg, Denmark
45-3-969-6511 . ,
Fax: 45-3-969-6002
Patrick Kennedy
Supervisory Chemist .
Office of Pollution ;
Prevention and Toxics
U.S. Environmental Protection Agency
401 M Street, SW (TS-779)
Washington, DC 20460
202-260-3916 '
Fax: 202-260-0981 j
Raymond Kent
Chemist
U.S. Environmental Protection Agency
401 M Street, SW(TS-778)
Washington, DC 20460
202-260-7974 , '•.
Fax:202-260-1216
Gregory Keoleian
Assistant Research Scientist
National Pollution
Prevention Center
University of Michigan
Dana Building
430 East University
Ann Arbor, Ml 48109-1115
313-764-3194
Fax:313-936-2195
Richard Kimerle
Senior Fellow !
Monsanto Corporation ,
800 North Lindbergh Boulevard
St. Louis, MO 63167
314-694-3286
Fax:314-694-1531
Celeste Kuta
Procter and Gamble
Ivorydale Technical Center
Cincinnati, OH 45217 :
513-627-4849 :
Fax:513-627-6291 ,-.
Leslie Lake
Research Toxicologist
S.C. Johnson & Son, Inc.
1525 Howe Street
Racine, Wl 53403
414-631-2955
Fax: 414-631-3752
B-3
-------
Murray Cohn
Division Director
U.S. Consumer Product
Safety Commission
(HS-590)
Washington, DC 20207
301-504-0994
Fax:301-504-0124
Frank Consoii
Manager of Packaging Technology
Scott Paper Company
Scott Plaza 2
Philadelphia, PA 19113
215-522-5467
Fax:215-522-7132
Cristina Cortinas De Nava
Advisor
Institute Nacional de Ecologia
RioElba20,Piso14
D.F., Mexico 06500
525-553-97-04
Fax: 525-553-97-53
Joseph Cotruvo
Director
Chemical Screening and
Risk Assessment Division
Office of Pollution
Prevention and Toxics
U.S. Environmental Protection Agency
401 M Street, SW (TS-788)
Washington, DC 20460
202-260-3442
Fax: 202-260-1216
Bruce Cranford
Program Manager
Office of Industrial Technologies
U.S. Department of Energy
1000 Independence Avenue, SW
Forrestal Building
Washington, DC 20585
202-586-9496
Fax:202-586-7114
Gary Davis
Center for Clean Products and
Clean Technologies
Energy, Environment and
Resources Center
University of Tennessee
327 South Stadium Hall
Knoxville.TN 37996-0710
615-974-4251
Fax:615-974-1838
Patricia Dillon
Research Associate
Center for Environmental Management
Tufts University
474 Boston Avenue
Medford, MA 01860
617-627-3486
Fax:617-627-3084
Richard Drawbaugh
Director
Environment, Safety &
Occupational Health
Human Systems Center/XRE
2510 Kennedy Circle
Brooks AFB.TX 78235-5120
210-536-4466
Fax:210-536-4475
Gordon Dumil
Chairman, United States Section
International Joint Commission
1250 23rd Street, NW-#100
Washington, DC 20440
202-736-9000
Fax:202-736-9015
Roger Ferenbaugh
Program Manager, Research &
Development
Environmental Management Division
Waste Minimization/
Research & Development
Los Alamos National Laboratory
P.O. Box 1663
(EM-DO/MS-K572)
Los Alamos, NM 87545
505-667-0811
Fax: 505-665-8195
Robert Ferrone
Digital Equipment Corporation
111 Powder Mil! Road
(MS02-3/B16)
Maynard, MA 01754
508-493-5146
Fax: JiOB-493-8353
Fran k Field
Director, Materials Systems Laboratory
Massachusetts Institute of Technology
77 Massachusetts Avenue
Room E40-227
Cambridge, MA 02142
617-253-2146
Fax:617-253-7140
Daniiel Fort
Section Chief
Chemical Engineering Branch
Office of Prevention
Pesticides & Toxic Substances
U.S. Environmental Protection Agency
401 M Street, SW(TS-779)
Washington, DC 20460
202-260-1694
Fax:202-260-0981
Roger Garrett
Director, Senior Scientist Staff
U.S. Environmental Protection Agency
401 M Street, SW (TS-796)
Wasihington, DC 20460
202-260-1256
Fax: 202-260-1283
Emma Lou George
Environmental Scientist
Office o-f Research and Development
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
26 West Martin Luther King Drive
(MS-466)
Cincinnati, OH 45268
513-569-7578
Fax: 513-569-7549
Barbara Glenn
7407 Birch Avenue
Ta!tomaPark,MD 20912
301-585-8955
B-2
-------
Evaluating Cleaner
EPA
Workshop on Identifying a Framework for the Future
of Human Health and Environmental Risk Ranking
Washington, DC
June 30-July 1,1993
Participant List
Braden Allenby
Research Director
Technology and Environment
AT&T
131 Morristown Road (B2176)
Basking Ridge, NJ ^07920
908-204-8440
Fax: 908-204-8217
Victoria Allies
President
TNT Technology Company
2121 West University Drive
Suite 123
Tempe, AZ 85281
602-966-9891
Fax: 602-968-9469
EarleAmey
Assistant Chief
Branch of Materials
U.S. Bureau of Mines
810 Seventh Street, NW
Washington, DC 20241
202-501-9580
Fax: 202-501-3747
Abby Arnold
Facilitator
Resolve
1250 24th Street, NW
Washington, DC 20037
202-778-9653
Fax:202-293-9211
Zebra Aydin
Transnational Corporations and
Management Division of the
United Nations Department on
Economic and Social Development
2 United Nations Plaza
Room 2002
New York, NY 10017
212-963-8811
Fax:212-293-9211
JackAzar
Manager, Environmental Design and
Resource Conservation
Xerox Corporation
800 Phillips Road
Building 317
Webster, NY 14580
716-422-9506
Fax: 716-422-8217
Donald Barnes
Staff Director
Science Advisory Board
U.S. Environmental Protection Agency
401 M Street, SW(A-101)
Washington, DC 20460
202-260-4126
Fax: 202-260-9232
Dennis Berry :
Manager, Environmental
Technologies Support Office
Sandia National Laboratories
P.O. Box 5800
Department 6602
Albuquerque, NM 87185
505-844-0234
Fax:505-844-8170
Gregory Biddinger
Ecotoxicology Associate
Section Head Ecotox Consulting
Exxon Biomedical Sciences
Mettlers Road (CN-2350)
East Millstone, NJ 08875-2350
908-873-6030
Fax: 908-873-6009
Nicolaas Bouwes
Economist
Office of Pollution
Prevention and Toxics
U.S. Environmental Protection Agency
401 M Street, SW(TS-779)
Washington, DC 20460 !
202-260-1567 :
Fax: 202-260-0981
Claudette Cofta
Chemical Manufacturers Association
2501 M Street, NW ;
Washington, DC 20037
202-887-1278
Fax: 202-887-1237
> Printed on Recycled Paper
B-l
-------
-------
APPENDIX B
LIST OF PARHdPANTS
-------
Agenda-Page Two
WEDNESDAY JUNE 30 (continued)
11-15AM PANEL SESSION IIMNSTITUTIONAL APPLICATIONS
MODERATOR: Bob Ferrone, Digital Equipment Corporation
, rt i .... Sven-Olof Ryding
Volvo Scoring System .......................... Federation of Swedish Industries
Axel Wenblad
AB Volvo Corporation Headquarters
GW System ...........
Michigan Critical Materials Register
12-.15PM Lunch
1:15PM
3-.OOPM
3:30PM
4:30PM
5:00-6:OOPM
6:00-9:OOPM
THURSDAY
7:OOAM
8:30AM
10:OOAM
10-.15AM
11:OOAM
12:00-1 :OOPM
1:15PM
2:15PM
2:30PM
3:15PM
4:30PM
5:OOPM
Issues in Ranking and Scoring Systems Development
Breakout Sessions
Break
Breakout Sessions (continued)
Adjourn
Reception
Administrative Assistance for Breakout Groups
JULY 1
Continental Breakfast
Facilitated Panel Session—Question 1
Break
Question 1 (continued)
Facilitated Panel Session—Question 2
Lunch
Question 2 (continued)
Break
Facilitated Small Group Discussion-Question 3
Facilitated Discussion
Wrap-up
Adjourn
...... Barbara Glenn
George Washington University
Gary Davis
CteanTecMogles
University of Tennessee
A-2
-------
EPA
* Framework for the Future
Health and Environmental Risk Ranking
Washington, DC
June 30-July 1,1993
Agenda
TUESDAY JUNE 29
6:00-8:OOPM Advance Check-in
WEDNESDAY JUNE 30
7:OOAM Continental Breakfast and Check-in
8.-30AM Welcome Address :
Mark Greenwood
c .... U.S. EPA
Facilitator Introduction
AbbyArnold
8:45AM PANEL SESSION MJSER NEEDS
MODERATOR: Franlrwin, World Wildlife Fund
ProductandProcess Design
_ t Motorola Corporation
Purchasing
Arthur Weissman
_ , ,. Green Seal
Regulations/Legislation....
Richard Kimerle
Monsanto Corporation
_ . ,, Mark Rossi
9:45AM Break Tox,cs Use Reduction Institute
10:15AM PANEL SESSION II-METHODS FOR EVALUATION
AERATOR: Frank Field, Massachusetts Institute of Technology (MIT)
RiskAssessment....
• Dorothy Patton
i» *. , U.S. EPA
Life Cycle Assessment....
FrankConsoli
Scott Paper Company
Process Design/Substitute Assessment
Libby Parker
; U.S. EPA
A-l
(over)
) Printed on Recycled Paper
-------
-------
APPENDIX A
WORKSHOP AGENDA
-------
-------
APPENDICES
-------
• Further develop a risk ranking framework.
• Further develop methods for assessing and ranking risks and impacts.
• Identity data requirements and gather information.
• Disseminate information on existing systems.
• Reach consensus on:
common language/definitions
impact categories/criteria
valuation procedures
• Develop case studies.
• Expand the steering committee (to perform specific tasks, such as developing a
work plan, and to include broader representation, e.g., from additional industries,
small businesses, more NGOs, user groups, and international perspectives).
An important aspect of this workshop, said a workshop participant, was that people
working on different types of systems, such as life cycle analysis, design for the environment, and
risk assessment, have convened. Maintaining a liaison among these people is key to developing
an effective risk ranking system.
-58-
-------
U.S. and European governments have not invested a lot of money into these areas. Instead,
pollution prevention has been consumer-driven, industry has seen the value of this work, and
governments have promoted it. Industry might want to retain its partnership in this area, rather
than shift pollution prevention to the regulatory arena.
An electronics industry member said that this industry is moving forward in the area of
pollution prevention and risk ranking in part because the industry is driven by ISO, Germany, the
Netherlands, and Sweden, While the United States generally is far behind in its environmental
regulatory approaches, a recently proposed amendment would base EPA's budget priorities on
scientific elements of environmental risk rather than on public perception. In the short term,
EPA should determine whether resources are available to organize another workshop. In the
long term, EPA must determine how the Agency will participate in whatever framework or
system is established, and where resources will come from for developing a national
environmental strategy.
Foster Public-Private Partnerships
EPA and industry need a sense of partnership to ensure a risk ranking system's i success,
said a workshop participant. Once the tools have been developed, a partnership could be
encouraged through case study applications or workshops on use of the tools. Finally, the
formation of a group of experts representing the various environmental risks or impact criteria
was recommended to monitor specific data needs and to ensure that the data arid framework
components work together. , .
CONCLUDING REMARKS
Workshop facilitator Abby Arnold summarized the broad recommendations of workshop
participants:
-57-
-------
Examine industrial metabolism studies.
Identify institutional barriers.
Analyze inventory data of the LCA or EPS models, identify the issues and
problems associated with the data, and determine how to work with the models as
they evolve.
Legislative Reform
Legislative reform is necessary to earmark resources for pollution prevention and risk
ranking activities because priorities conflict with some existing environmental laws, said a
workshop participant. In addition, EPA policy reforms are needed to move from an agenda
driven by legislation to a set of priorities focusing on risk ranking or life cycle analysis. Only
after legislative and policy reforms are made can a specific plan of implementation, including
funding and other resources, be developed for a risk ranking system.
In the absence of legislative and policy changes, would industry and other institutions still
use a risk ranking framework, asked a workshop attendee. An Exxon employee said that his
company already uses such a framework and would continue to do so, even though perhaps 80
percent of the company's environmental resources are directed toward regulatory compliance
rather than innovations such as pollution prevention. Legislative reform might shift the balance
more toward innovation. According to another workshop participant, pollution prevention
systems might be valuable even in the current legislative climate because of customer preference.
Legislation typically determines priorities, regardless of how certain issues rank as relative
risks, one participant said. But as risk-based decision-making is emphasized and definitions of
risk and other important concepts are incorporated into a risk ranking system, then legislation
and policy might move in that direction. Also, a substantial, ongoing financial commitment for
this work is necessary, said a participant.
Illustration of early successes can help move risk ranking and pollution prevention
forward without a major financial commitment, someone commented. Legislative mandates have
not sparked the development of risk ranking and life cycle assessment, another participant said;
-56-
-------
Training
EPA could train system users to ensure that they understand the system and how to apply
it. The training should decentralize the pollution prevention process by preparing users to return
to their own organizations and use the system for their own applications. Flow charts for users
could be developed, outlining step-by-step procedures. Individuals with expertise in specific areas
could develop different parts of the flow charts.
Communication and Information Exchange
Users of a risk ranking system should participate in developing the system's framework,
said one group. In addition, the system should incorporate international input and acceptance
from the beginning; otherwise, some countries could follow a completely different track than the
system being developed. A final recommendation by this group was to identify one or more
institutions that could act as a standards organization to ensure consistency and incorporate
future improvements of the system to keep it viable.
Development of a National Strategy: A Workshop Series
Initiating a workshop series focusing on formulation of a national strategy similar to the
Dutch approach, among other issues, was recommended by another group. The Dutch
environmental policy plan, created by government, industry, and nongovernmental organizations
(NGOs), establishes the goal of achieving a sustainable national economy by the next generation.
The plan includes programs for measuring the performance and/or toxicity of materials, policies
for different sectors, covenants, and specific goal-oriented projects.
Other workshops could be held to:
• Reach agreement on risk ranking terminology.
• Develop a broad system similar to the EPS system.
-55-
-------
Separate Framework Development from Data Ranking
As a risk ranking system is being developed, the process of creating a framework should
be kept separate from the process of ranking data, one group suggested. Otherw-se, the
r r i w.. n*+0rvmtin
-------
Detailed Assessment Procedures
Another group recommended focusing on developing a detailed procedure for assessment
and dedszon-making rather than screening, since screening procedures already exist and the
elements of screening are not as critical. Once a better system for detailed assessment is
developed, the elements in this system could be incorporated into screening procedures.
C^*te Relationship Between sustainable Development and PO
Perhaps a multidisciplinary advisory group could address the relationship of sustainable
development to pollution prevention work, one group suggested.
Consensus on System Components
A group recommended reaching a consensus on categories for environmental impacts and
on valuauon procedures, even though no, every category or criteria would be used in every
assessment, Jndustry, regulators, and environmentalists migl,, choose different criteria, J,
everyone would agree that the criteria all belong in the overall system. ;
Other system components to be developed include:
A mechanism to measure progress. :
A method for developing benchmarks.
-53-
-------
framework. Also, the scientific data base should begin to be compiled, focusing initially on
identifying existing data and needs rather than pursuing new data collection.
Review of Existing Systems
A short-term next step suggested by another group was to collect data and disseminate
information on existing tools and assessment methodologies for both life cycle and environmental
assessment, focusing on procedures to consider in future system development (see Appendices I
andK) Risk and impact criteria should be defined from these state^f-tiie-art procedures. This
information, which should be disseminated before another meeting is convened, would serve as a
benchmark in developing a risk ranking system. Also, existing statistical and modeling techmques
should be evaluated to determine whether they might help fill data gaps.
Case Studies
A workshop could be held to study hypothetical case studies of different existing systems
to identify problems and successes and to learn how the systems work to specific circumstances.
One case study could focus on the internal needs of a company, another could prioritize
resources in a government program, and a third could evaluate confer products. Areas could
be identified in which data are needed and where classification schemes and criteria could be
harmonized.
Assessment of Valuation Procedures
A workshop could be held to identify and evaluate valuation approaches. One method
might be to focus on tiers (e.g., low, medium, and high) rather than numerical rankings. All
subsequent workshops should have an international focus, most groups agreed.
-52-
-------
have
Develop strategic action plans based on the issues discussed at this
"«** — ^ies, addition^
Another group recommended a workshop to develop a conceptual framework document
for people to use and critique. A goal of such a workshop would be to reach consensus on a
conceptual framework, as well as on associated terms and definitions. To prepare for such a
workshop, adequate knowledge of existing systems and how they are used would be needed This
workshop, which could take place within a year, in turn would lead to additional workshops on
an overall, global ranking system. For example, a subsequent workshop should focus on
developing a comprehensive chemical ranking and scoring system. Approximately 40 such
systems currently exist.
The framework should include logical building blocks for ranking as well as
comprehensive criteria for selecting relevant areas and adding others, according to one group
Specific scientific data should be compiled separately. The framework could include an online
data base of: ;
• Information on experts in various, relevant fields
• Centralized information on existing methodologies and data bases
• Ways to access the data
The system should be broad and include end user options that a flexible framework could
address in a single, comparative ranking. The framework and scientific data base together would
guide pollution prevention and ranking applications.
This group also identified short-term next steps for achieving the system they described
A first-cut scoping process should be performed to develop the framework. The focus should be
on developing generic tools that can be used for a variety of applications. Perhaps the
framework should be defined for a single application and then broadened into a generic
-51-
-------
This six-step process hopefully would encourage a diversity of groups to use the system.
The ultimate goal is to move toward a national or international standard so that people evaluate
risks in the same way, make similar assessments, and reach similar conclusions on how to
minimize and ameliorate environmental impacts. \
Framework Development: Strategic Action Plans
The overall goal is to create a framework or tool for decision-making that enhances
environmental quality, said one group member; development of a risk ranking or scoring system
is not an endpoint in itself. Another workshop should be held to develop a conceptual
framework further. The next steps recommended by this group were:
B Gather and disseminate information about existing systems and other baseline
information prior to the next workshop.
• Develop a draft model or working papers for a conceptual framework. Circulate
it in advance of the next workshop for critique at the meeting.
B Discuss the following issues at the meeting:
Develop an overall model built on existing frameworks (e.g., LCA, EPS,
the Dutch national environmental policy).
Clarify system applications and users, including how different user groups
would use the model.
Develop feedback mechanisms to enhance a decision-making tool.
Identify data requirements, including whether existing data are appropriate
and available, what new data is needed, and whether different data sets
can be used together in one model.
Identify institutional barriers (e.g., current regulatory policy); integrate
institutional components (e.g., international systems, ISO standards) as
well as economic and social issues.
Establish partnerships and stakeholders: Who should the partners be?
What public-private and international partnerships should be developed?
Should further work be coordinated by one organization or a consortium
of groups and which ones? Who will be responsible for ongoing work,
-50-
-------
Develop a more detailed framework, through a conceptual framework committee
or another means. In developing the framework, address issues such as:
the role of exposure in a ranking system :
social issues
specific evaluation techniques
making the framework flexible so it can be modified for different uses
Validate the framework, once established, through case studies that test its
boundaries. Determine whether the model works for large-scale goals, such as
national policy planning, as well as small-scale applications, such as choosing
specific materials to manufacture a product. Publishing these case studies could
provide valuable incentives for continuing pollution prevention and risk rankine
work. ,
Develop methods to test the system's success (i.e., whether it has accomplished
short- and long-term goals).
Centralized Coordination, Development of a Draft Risk Ranking Model for Review
Another group identified six steps in a risk ranking process:
• Establish a central coordination center administered by EPA or another
organization to develop one or more generic conceptual framework(s) for
evaluating health and environmental consequences of products and processes.
• Gather all relevant information from state, local, and federal governments,
industry, academia, and international organizations. (Sweden's compilation of 70
data bases should be included in the collection.)
• Develop a draft risk ranking model for further evaluation.
• Hold additional workshops to critique the draft model and propose modifications
as appropriate to address outstanding issues.
• Submit a refined model to stakeholder groups for review, including public interest
groups, academia, industry, government, and anyone else interested in the results
beek feedback from these people on the model's usefulness.
• Distribute the risk ranking model to user groups for evaluation Provide
assistance on how the system works and can be adapted to specific situations
This step would be continuous, with ongoing feedback between the system's
developers and users to improve the model.
-49-
-------
goal must be addressed. Methods must be developed to evaluate whether the process is working.
If human health is not at the core of the system, said another attendee, the system probably has
less chance of being used; if socioeconomic issues are not addressed in the system, decision-
makers will be less inclined to use it.
Next Steps; Breakout Group Recommendations
An important goal of the workshop, according to Libby Parker, director of EPA's Design
for the Environment Program, was to produce a plan of action—a concrete series of next steps.
The workshop should be a catalyst for further action. The workshop first ifocused on what type
of risk ranking system would be useful for making decisions and identifying problems (e.g., data
gaps, .data quality). The next task was to plan how to develop this system further, including
identifying possible next steps, such as whether additional meetings should be held and what
topics they should cover. Summarized below are workshop participants' recommended next steps
for further developing a risk ranking system.
Development of a Work Plan, Validation of the System
While many diverse points of view were expressed at the workshop, fundamental
agreement was reached on the principles and the overall framework of a risk ranking system.
The process of developing such a system involves defining the purpose and scope, which leads to
an overall assessment. In the assessment phase, one can begin with the assessment and move on
to a value system, or begin with a value system and move on to the assessment. Then
improvement opportunities are considered, which could lead back to the beginning of the
process. The process is iterative.
One group suggested several specific next steps, including:
• Develop a work plan based on workshop participants' recommendations and
disseminate this plan (either through another workshop or by mail) to workshop
attendees for feedback. The Steering Committee should take this step.
-48-
-------
A risk ranking system should provide a set of tools for different levels within an economy
to help understand the manifestations of risks and to control them at the appropriate level most
efficiently, another participant said. This process, while it is extremely complex, is what must be
done to move toward sustainability. The Bureau of Mines, for example, is examining industrial
metabolism studies (mass flow balances across an industry or industries, such as cobalt in the
superconductor industry) of various metals within the global economy. This examination goes
beyond wriat most LCA analyses presently do. The question remains: How do we integrate
these types of studies and impacts and develop policies to reduce risk as a whole?
Another person commented that the definition of LCA seems to be evolving. While
some see it as a tool to bring sustainable development to the forefront, others see LCA as a local
decision-making process at the corporate level. Used effectively, LCA can lead to policy
decision-making from the bottom up. LCA also can be used to demonstrate that a business-as-
usual approach is not working and to help define sustainable development. The real solutions to
environmental problems probably will take place at the national or international level; LCA
cannot address this level, but is moving in the right direction. Sustainable development will take
time.
If risk ranking is a direct tool for sustainable development, the elements of sustainable
development should be defined, said a workshop participant. Another workshop member
responded that developing a risk ranking system was complex enough, and this workshop focuses
on further development of such a system or systems. Once this occurs, then the issue of
sustainable development should be addressed. Another participant said that risk is different
from sustainable development; depending on how sustainability is defined, some risky processes
and industries might be sustainable. .
!
The effect of inadequate data on LCA also was raised, Because of the paucity of data,
most LCAs would be conducted on default values. Selection of these default values, therefore, is
important.
An attendee emphasized the importance of validating a ranking system or model once it
has been developed. Whatever the goal—sustainable development or lowering the incidence of
disease in a public health context—the question of whether the model actually accomplishes the
-47-
-------
• Information dissemination to consumers
• Monitoring and regulatory support
In addition, the system should allow the user to divide impacts iinto more specific
classifications and perform analysis within impact categories, said the Group IX representatives.
Also, a common set of principles should be established that can be applied to both the chemicals
approach and the materials/products approach, although the actual procedures for each might
differ. The group also felt that the criteria list should be quite extensive; even if not all the
criteria are used, this comprehensive list would serve as a safety net. 'The group recognized,
however, that making choices and decisions becomes increasingly difficult as more criteria are
added to the system.
Group IX also recommended that the framework explicitly recognize the limitations of
life cycle assessment. The likelihood is small that any given product sysfStn contributes in a
major way to a global or even regional problem. Since a particular product system is just one
small part of a larger scheme, a single life cycle assessment will not allow for global decision-
making. From a societal standpoint, the implications of numerous life cycle analyses need to be
synthesized to understand the global impacts.
Finally, people in Group IX involved in chemical ranking thoujght that consideration of
life cycle information in a chemical ranking system would be a useful and significant extension of
current systems.
Discussion on Purpose, Criteria, and Framework for a Risk Ranking System
The impacts of risk ranking on company viability were discussed. Large corporations are
threatening to stop buying from small businesses unless they become part of a quality control
program. In the face of such demands, small companies now could be asked to rank chemicals
and design new products differently with extremely limited technical resources.
According to a workshop participant, one group's conclusion that human health effects
are more important than ecological effects really is a valuation issue to be decided at the end of
the ranking and scoring process, rather than at the start.
-46-
-------
Group VIH identified the purposes of a risk ranking system as:
• Protection of human health, with a secondary level of protecting all living
organisms
• Empowerment of industry to make environmental decisions
• Reduction of the use of toxic chemicals
" Promotion of sustainable development
• Examination of alternatives for long-range planning
Regarding criteria, the group noted that different types of people will rank criteria
differently; what the government ranks first might be ranked last by the public. This
phenomenon presents an interesting problem.
Group VIH also identified three ways that pollution prevention can contribute to
sustainable development. The first is to replace a nonrenewable resource with a renewable
material, if possible. The second is to find a nonrenewable resource that is not as scarce as the
original material, if a suitable renewable resource cannot be found. The third method is to
eliminate the use of the material altogether, if possible. If the scope of life cycle analysis is
broad enough, according to this group, sustainability could be included in it because LCA
evaluates all possible substitutions.
Group IX focused on the philosophy and psychology behind risk ranking. The group felt
that consistency of thought process was important in providing a systematic evaluation method
regardless of whether the goal is chemical ranking or assessment of a product or process. The
ranking system should provide a systematic procedure for thinking about what will be considered
and how. The ranking system also should be part of a decision-making.framework. The system
would need a broad set of generic uses, including:
• Product development
• Evaluation of processes from a risk perspective
-45-
-------
decide whether to undercoat with a polymer or another material, but it will not influence
whether Volvo continues to make cars.
Another respondent said that two parameters pull together the diverse issues regarding
chemicals or materials used versus types of product use. The first is that, .given access to the
same generic data, both types of investigations could be conducted. Hie socond is the time
element—whether one is seeking results immediately or into the next century will in part
determine the focus of an analysis.
A participant expressed concern about narrowing the focus of pollution prevention to risk
reduction opportunities because this limits the scope of action. Another person commented that
the EPS system superimposes a broader global framework of identifying priority issues on the life
cycle assessment concept, thus allowing this modified LCA to measure impacts on priority issues
involved in sustainability. In contrast, LCA used outside the EPS system only focuses on impacts
without identifying priorities.
Breakout Groups VII, VIII, and IX
Presentations on Purpose, Criteria, and Framework for a Risk Ranking System
Group VII focused on criteria development using a "simple" example and a more complex
example. The simple example was a single chemical with multiple uses; the more complex
example was a use cluster and multiple chemicals or materials. For the simple example, the
group concluded that, taking a global view, risk ranking itself will-not be able to accomplish
pollution prevention. Chemical substitution becomes a subset of evaluating design options and
alternatives because the substitution affects the entire product development process. For the
more complex example, the group felt that a framework incorporating a life cycle analysis should
be developed that would evaluate all alternatives, not only chemical substitutes—a holistic
approach to identifying options. The application or use of a particular chemical or material is
the primary driving force in defining the scope of the analysis. The EPS system moves toward a
holistic approach by examining the relationship between technical and socioeconomic issues.
-44-
-------
ranking methodologies. Opportunities for pollution prevention could then be incorporated into
the analysis. Much more detail would be needed when specific activities are considered. Public
policy applications, however, probably would need to be more complex from the start, since they
require a state or multistate focus rather than concentrating on a single company. '
A cooperative agreement between EPA and industries or individual corporations was
recommended to identity problem areas and possible improvements. Someone suggested that
more emphasis be placed on changing the end use of a product rather than focusing only on
changing the material or substance used in manufacturing that product. Another participant
responded that design and manufacture have been the focus rather than use, based on an
assumption that end users are indiscreet in their handling and use of products. Marketing
strategies can be used, however, to promote certain products. For example, energy usage during
the lifetime of a product may be a selling point even though it is external to the actual
manufacture of a product. In this way, usage can be brought into a life cycle analysis. The
question still remains whether changing the use of products can be emphasized, thus placing
more responsibility on the end user to prevent pollution.
The issue of where to place risk highlights two concepts of risk, a workshop participant
said. The first is a very localized concept that involves various specific methodologies. The
second is a broader concept of risk that involves .determining what prevents us from moving
toward sustainability as an economy. For example, a localized approach to risk for automobiles
would evaluate issues such as how automobiles are manufactured, and the use of solvents in
paint. But this approach does not address sustainability issues such as the amount of fossil fuel
that automobiles consume, the living patterns they have fostered, the impacts of paving large
areas for driving, and the resources used. While it is important to address localized risk, unless
risk ranking is put in the context of achieving sustainability, the seriousness of the problem is not
addressed fully. : f
Someone responded that perceiving life cycle assessment as localized is a misconception;
global questions, such as ecological health and resource depletion, are very much involved in
examining improvement opportunities, for functional use. Another individual disagreed, saying
that LCA does not address the larger issues of whether economies can be sustained using the
level of resources that we have been using on national or global scales. LCA might help Volvo
-43-
-------
correctly ranked high in one use category and low in another. Activities in which pollution
prevention and risk reduction opportunities exist should be targeted, rather than focusing only on
where the greatest need (or risk) exists. This is supported by the fact that little legislation exists
for pollution prevention because it is a cross-media concern, according; to this group; therefore
EPA must look toward corporate decision-making to accomplish pollution, prevention rather than
using the traditional regulatory approach to define what is in the best interests of society.
Group VI also felt that the decision-aiding aspects of a ranking system should be
emphasized more than specific methodologies for ranking because corporations need to focus on
decision-making in pollution prevention activities.
Discussion on Purpose, Criteria, and Framework for a Risk Ranking System
A workshop participant commented that Group V seemed to imply that two separate
systems should exist-a chemical-based and a process-based system-yet the group also said that
the chemical-based system would fit into the process-based system. A Group V member clarified
that the group identified two basic purposes—priority-based and process-based—rather than two
separate systems, and recognized that the final system would need to be capable of being used in
both ways.
Another participant commented that the chemical-oriented approach seems to allow an
evaluation of process, because once a particular chemical is identified as a risk, the focus is on
pollution prevention techniques associated with that chemical's use or use cluster. One is then
working in a different framework.
The distinction between two types of ranking applications is a,n important one, said a
workshop participant. One application is priority setting or screening, and the other is decision-
making. The screening application requires much less stringency regarding data quality and the
process in general because the implication is that there will be further analysis. For decision-
making, much greater care must be taken because the ranking can eliminate or select products.
Another participant concurred that for prioritizing purposes, a ranking system could be simpler,
and that the emphasis would be on understanding where the risks are rather than on distinct
-42-
-------
Group V thought that chemical risk ranking systems would be part of life cycle analyses,
that the purpose of a risk ranking system varies depending on the organization and its goals, and
that no single ranking system can achieve everything. While a ranking system should be designed
to meet the needs of a specific organization, it still should be generic, said the group's
representative. Its purpose could be defined by a set of elements, such as exposure, function,
cost, and toxicity, which would be harmonized. The group's concept of harmonizing involved'
using criteria identified for each purpose in any subsequent life cycle analysis, resource
prioritization, or alternative evaluation. Group V also raised a philosophical question regarding
a ranking system: Is a product considered innocent until proven guilty, or guilty until proven
innocent?
Group VI first assumed that the goal was to develop a single, generic system that could
be adjusted to incorporate different purposes. As discussion continued, however, the group
concluded that a ranking system could have two fundamentally different purposes, each of which
would result in the design of different systems in terms of components and functionality. The
first purpose would be to identify high-priority needs for pollution prevention (e.g., current risks)
The second purpose would be to identify pollution prevention opportunities (i.e., areas of
greatest risk reduction potential). The GW and Michigan systems seem to excel in prioritizing
areas for action. The EPS system seems to address achieving pollution prevention. Very
different systems will be developed depending on which of these goals or functions the system is
to accomplish.
People who are most concerned about prioritizing needs or identifying risk levels tend to
focus on chemicals, while process-oriented people are more concerned with pollution prevention
opportunities, said the group representative. The prioritizing approach focuses on concerns of
the general public, and the process approach emphasizes the corporate perspective. Whether
action occurs at the corporate level or in a regulatory setting, public concerns should be
incorporated. With a process-oriented approach, financial considerations and health risks are
emphasized more than with the prioritizing approach.
Individual chemical risks must be evaluated in the context of their specific uses since the
uses define services, and pollution prevention involves the way changes are made in providing
basic services, said the Group VI representative. The group thought that a chemical might be
-41-
-------
prioritization (e.g., human health versus resource depletion). Developing countries probably will
not have the resources to act on all the environmental fronts identified While human health
would take precedence in these countries, developed countries might address all the
environmental categories.
Participants discussed whether techniques used in other disciplines (such as sociology and
economics) should be considered as scientific methods and included in the risk assessment stage
of risk ranking. People agreed that at the risk assessment stage, reaching a broad-based
consensus is the goal; small groups of people who conduct risk management at the regional or
local level can focus on more specific issues.
The issue of whether a ranking scheme can be used across use categories was raised (e.g.,
materials used in automobile manufacturing versus pesticide use in food production). One
participant responded that as a prioritization tool for resource allocation, a ranking scheme could
probably be used across use categories; another said that much more subjectivity would be
involved in ranking across uses, A third respondent stated that a ranking system is tailored to
one use category, but that if a linkage can be found between categories, then the system can
become a single ranking.
Breakout Groups IV, V, and VI
\
Presentations on Purpose, Criteria, and Framework for a RiskRanldng System
Group IV concurred with other groups that the framework for a ranking system needs to
be broad and generic enough to be used for ranking and scoring of specific chemicals as well as
for comprehensive life cycle impact assessments. The framework needs to be useable for
multiple purposes and by different types of users. Socioeconomic issues should not be ignored
completely, this group suggested; links between different areas of concern (e.g., technical and
societal) should be explored. Agreement might be reached best at the largest level of
aggregation, such as the purpose of the system.
-40-
-------
• Human health
• Ecological health
• Aesthetic values
• Resources '
• Energy use
• Indirect potential risks
" Unknown risks
The last category, unknown risks, describes detrimental effects that are not measurable or
quantifiable resulting from a particular process. In comparing two alternative production
processes, one might want to use the process without any releases, even though a particular
detrimental effect cannot be associated with the release.
After the general risk categories are defined, matrices or lists of criteria would be
developed to identify further potential impacts associated with a particular process. The specific
process being evaluated then would need to be examined closely to determine whether additional
criteria should be added. The group strongly suggested that results or rankings should not be
f * • , ' - * * (' ' » . f -" ' _ ' ,'. *• "'. t' '•*••'• - - '
aggregated between individual categories. While aggregation might be possible in some cases, it
would not work in others. In the final analysis, however, one must examine all the rankings from
all the risk categories and criteria, and mesh them with political considerations to make a final,
informed, rational decision on how to proceed.
Discussion on Purpose, Criteria, and Framework for a Risk Ranking System
The relationship of a ranking scheme to societal issues (such as jobs and health) needs to
be clarified, and its relationship to cultural issues should be acknowledged, said a workshop
participant. Risk ranking frameworks might differ between countries, given different valuations
between resources and societal needs. Someone responded that, first, the ranking scheme is
specifically an environmental tool, and thus societal issues should not be part of the framework.
Second, societal values would be expressed through the different stakeholders, who influence
-39-
-------
• The framework needs to be generic enough to enable decision-making at both the
global and local levels and to accommodate the different priorities and issues
identified by various groups.
• The system should contain rules for making rational and consistent decisions
based on available information.
• The system should enable prioritization of environmental issues.
• The system should be a vehicle for resource allocation (e.g., money, jobs, natural
resources), which is very important for government agencies and companies.
Regarding risk and impact criteria, Group H emphasized the importance of respecting
categories specific to various fields (e.g., chemicals should not be ranked along with waste
impacts) at this stage in the development of a risk ranking system. At this initial stage, the group
felt that different ranking and assessment methods should be established for different criteria
categories (e.g., human impact, ecological impact, resource use), and that these methods should
remain separate rather than be aggregated. Also, stakeholders with different values about
technologies, processes, improvements, and/or resources need a model that enhances different
types of decision-making, said the group's representative. Rather than a single ranking system
with a list of the 100 most important items, users need techniques and methods to weigh and
rank different impacts or risks.
These models and techniques should have a clearly defined purpose and scope, which
would direct the other steps and methods in the process, such as further defining the scope of
the analysis and identifying areas for further study. The first part of the method should be based
on scientific protocols, analogous to risk assessment (see Figure 1 in Aj>pendix J); and the second
part, similar to risk management, should allow stakeholders to rank and valuate the information
from the first step (see Figure 2 in Appendix J). The group felt that these two steps could not
be joined. This two-tiered process would lead to identification of improvement opportunities and
control options and would facilitate decision-making.
Group m thought a generic model that included all potential impacts would be most
useful; these impacts could be tailored for a particular purpose. In developing a generic model,
the group felt that specific criteria would be difficult to define; insteacl, more general risk
categories first could be defined, such as:
-38-
-------
BREAKOUT SESSIONS
A representative from each of the nine breakout groups summarized their discussions on
the purpose, framework, criteria, and next steps for a risk ranking system. Three groups at a
time gave presentations on purpose, criteria, and framework, followed by a discussion among all
workshop participants. All breakout groups then presented recommendations for next steps in
developing a risk ranking system, interspersed by questions and discussions from the entire
workshop group. A summary of these presentations and discussions follows.
Breakout Groups I, II, and III
Presentations on Purpose, Criteria, and Framework for a Risk Ranking System
Group I identified several purposes of a ranking system, including:
• Providing a means to demonstrate and measure success.
• Communicating and educating the public, companies, government, and others
about environmental priority issues and prospective decisions.
• Helping to balance short- and long-term environmental goals.
• Accommodating different levels of analysis, from screening to detailed decision-
making involving the use of chemicals or raw materials.
» Assisting in evaluating tradeoffs.
Group I also identified several important generic elements of a risk ranking framework:
« The system should be flexible because future issues and forthcoming data sets are
unknown. A dynamic system is needed that is able to incorporate new
developments so that decisions can be changed when new information is available.
• The system must be a practical tool that people find useable.
» The current framework for comparative risk assessment should be expanded by
incorporating life cycle, cradle-to-grave impacts in addition to toxic chemicals.
-37-
-------
• What are some of the next steps that need to be taken to develop a framework
for ranking? •
Ms. Arnold advised the breakout groups to begin by addressing the purpose of the
ranking system, which perhaps could be clarified by asking, what is the framework being
developed? Should it be an international or national framework? Should it be an industry
sector framework? Should it be for regulatory affairs or research and development? Should
more than one framework be identified, depending on the application needed?
Ms. Arnold suggested that determining the purpose also might involve identifying the
audience and understanding the results being sought. For example, is the purpose of the ranking
system or framework to assist design engineers in considering the environment, or to aid
environmentalists in considering technology? The purpose should be focused on pollution
prevention and on human health or environmental risk ranking, and leus focus should be placed
on environmental restoration or waste management, Ms. Arnold said. The focus should be on
including environmental considerations in the design of a product, a production process, or
valuation, rather than on how much cleanup is required at a site before it is considered to be
clean. Good waste management, however, could help close the pollution prevention cycle, since
the whole lifecycle of chemicals or products is being examined, from cradle to grave.
After defining the purpose of a ranking system and framework, the breakout groups
should identify criteria for at least one framework. In determining what criteria should be used,
the breakout group should consider the role of different interested parties, such as industry,
labor, or environmentalists, and ask what the criteria should be. The groups' would then try to
address the other questions posed regarding problems, obstacles, work needed to develop
criteria, and .next steps, including the issues raised by Mr. Davis, such as types of impacts,
measures of exposure, data gaps, and aggregation of impacts. Different breakout groups
probably will return with different approaches, perspectives, and methodologies, Ms. Arnold said.
She also asked each breakout group to designate one member as a reporter who would
summarize the group's discussions when the entire workshop group reconvened.
-36-
-------
impact valuation every time he or she designs a product, and thus aggregation may be
appropriate for a ranking system used for this purpose. But if a system is used for regulatory or
policy purposes, then the decision to aggregate or not is a political one; aggregation may or may
not be appropriate depending on the type of decision to be made. We must realize that a
perfect, scientific approach to aggregation and weighting cannot be devised.
Conclusion
Mr. Davis concluded by emphasizing that the development of a comprehensive human
health and environmental risk ranking and scoring system is in its infancy. An appropriate
framework for such a system is only beginning to be addressed. Most people agree that simple,
readily usable tools are needed. The issues of aggregation of impacts and of whose value system
(e.g., the scientific community, the general public) is used in the valuation process are very
important. The number of chemical ranking and scoring systems being developed has
proliferated; perhaps a consensus based on such systems can result. A chemical ranking and
scoring system can be a component of a comprehensive human health and environmental risk
ranking system.
INTRODUCTION TO BREAKOUT SESSIONS
Abby Arnold, workshop facilitator, said that the goal of the breakout sessions was to
attempt to answer several questions, including:
• What is the purpose of the ranking system?
» In developing a framework for a ranking system, what criteria should be included?
• What are some of the problems associated with developing a ranking system?
What are the obstacles for using certain criteria?
» What work needs to be done to develop the criteria and to address the problems?
-35-
-------
scoring systems because site-specific risk assessments are not conducted for these systems.
Surrogates for exposure frequently are used, such as the amount of chemicals released into the
environment, the use or production of chemicals, and measures of environmental fate, such as
persistence and bioaccumulation. Similarly, with life cycle impact assessment, it is very difficult
to be site-specific given the complex nature of product systems, and thus surrogates for exposure
are used. People often refer to life cycle impact assessment as an attribute of a product system,
but life cycle impact assessment also is a means of measuring environmental impact.
Missing Data
How missing data is handled is a critical issue for both chemical ranking and scoring
systems and life cycle impact assessments. The issue of data gaps is especially important for life
cycle impact assessment because this type of system is more data-intensive than chemical ranking
and scoring systems, and yet much less data are available for life cycle impact assessment. Also,
the broad measures of environmental impact in life cycle assessment are not typically tested or
regulated (e.g., global warming).
Approaches to missing data vary widely among different systems. Some systems use the
most sensitive indicator to determine impact. Other systems provide alternative indicators of
impacts from which one can choose or default to a particular indicator if data on other indicators
are not available. Some systems use quantitative structure-activity relationships to fill data gaps.
This approach involves determining the likely toxicity of a chemical, based on its chemical
structure, using models and expert judgment. Whatever type of system is developed, the issue of
missing data must be considered.
Aggregation and Weighting of Impacts
A major issue is whether the many widely divergent impacts considered in relative risk
ranking systems should be aggregated across impact categories. Whether or not to aggregate
depends on the purpose of the system. A designer of a product does mot want to undergo
-34-
-------
Purpose of the System
To a large degree, the purpose of the system determines how the system is developed.
Chemical ranking and scoring and life cycle impact assessment have had very different purposes.
Chemical ranking and scoring systems have been developed for regulatory purposes, priority
setting, and occasionally for impact assessment. Examples of chemical ranking and scoring
systems for regulatory purposes include the Superfund Reportable Quantities System, the
Michigan Critical Materials Registry, and the GW system. The EPA Use Cluster Scoring
System, developed for evaluating substitute chemicals, is an example of a priority-setting system,
as is the Agency for Toxic Substances and Disease Registry's system for ranking priority
chemicals at Superfund sites in the United States, and a Dutch system for priority setting. Life
cycle impact assessment systems also have been developed for a variety of purposes, including
product improvement, which is a primary goal, product design (such as the Environmental
Priority Strategies, or EPS, system used by Volvo), setting public policy, and environmental
labeling.
Selection of Health and Environmental Impacts
The different impacts included in a system are extremely important. All systems divide
impacts into two categories—human health and the environment. With chemical ranking and
scoring systems, only the direct toxic effects of chemicals are included, whereas in life cycle
impact assessment, additional types of impacts are included, such as ozone depletion, global
warming, acid rain, eutrophication of lakes and streams, and oxygen depletion of water. Since
chemical ranking and scoring is an important part of life cycle impact assessment, the two
different systems could converge in some areas.
Measures of Exposure
Measures of exposure are critical because risk is determined by exposure as well as
toxicity. Certain generic measures of exposure are included in most chemical ranking and
-33-
-------
concerns should have been dealt with all along. The goal of pollution prevention must be kept in
mind as ranking and scoring systems are developed; these systems are tools to achieve this goal.
Several years ago, the University began working with EPA's Risk Reduction Engineering
Laboratory in Cincinnati on a project involving safe substitutes for toxic chemicals. To conduct
evaluations of potential substitutes, a priority list of chemicals was needed. The staff began
developing a ranking and scoring system, since no consensus existed. This system is still being
developed, as is most everyone else's system. This workshop is important because it is one of the
first meetings attended by people developing a chemical ranking and scoring system as well as
people involved in life cycle impact assessment. One goal of the workshop is to bring these two
groups of people closer together.
EPA has been evaluating chemical ranking and scoring systems to improve the Agency's
methodologies. It has identified over 40 such systems that are in use or have been developed for
a wide variety of purposes. EPA also has been involved with SETAC in developing of a
methodology for life cycle impact assessment. Common issues exist between chemical ranking
and scoring systems and life cycle impact assessment, and these issues must be addressed in
attempting to develop consensus, scientifically and otherwise, for a uniform framework and
methodology. These common issues include:
• What is the purpose of the system?
• What human health and environmental impacts are included in the system?
Should potency and severity of effect be included?
• What types of measures of exposure are included in the system?
• How is missing data handled in the system?
• Should the different impacts be aggregated and weighted?
Each of these issues is discussed briefly below.
-32-
-------
Without specific data, product designers have difficulty commenting on the system; they simply
say, "that's interesting," and "come back when you have the data." Another point made was that
ranking of environmental risks or impacts could vary for different industries (e.g., the automobile
versus the paper industry) and this variety should somehow be incorporated into a ranking
system.
Three criteria for the development of an environmental risk ranking system can be
identified, said one panelist:
• Corporate commitment to environmental protection must exist.
• The system must be a tool that answers the users' questions, not only the scientists'
questions.
• The system needs to be readily usable and understandable.
Panelists also raised the issue of regulatory versus nonregulatory motivations for industry
to implement pollution prevention. Companies might be motivated to adopt a life cycle-based
philosophy if proposed product-related legislation would affect them, as is the case in Sweden.
Instituting life cycle and risk ranking approaches before legislation is passed might cost
companies less than figuring out how to comply with regulations after their passage. Tools such
as risk ranking and life cycle analysis, according to several participants, gradually are changing
the way people think about product development.
ISSUES IN RANKING AND SCORING SYSTEM DEVELOPMENT
Gary Davis of the Center for Clean Products and Clean Technologies at the University of
Tennessee gave a presentation on issues commonly encountered in developing risk ranking
systems (also see his paper on this subject in Appendix I). Development of risk ranking systems,
he said, is a revolutionary approach to protecting the environment. For years, pollution has been
addressed by controlling the discharge of pollutants from the end of the pipe and out of the
stack. With the pollution prevention approach, environmental concerns are beginning to be
addressed in the board room and at the designer's drawing table; perhaps this is where such
-31-
-------
information from suppliers who might have been affected by the pollution prevention measures
instituted. Thus, in the process of ranking these chemicals, data is being stimulated for
reevaluation.
PANEL RESPONSES TO QUESTIONS
In responding to questions from workshop participants, some panelists said that the
greatest impact of ranking systems likely will be in the design of products and processes. Other
panelists emphasized that, in addition to the toxicity and environmental fate of chemicals, other
factors must be considered in a ranking process; decisions are made based on the relative rank of
all factors (e.g., costs as well as benefits).
If a substance or group of substances has a known high risk, some panelists suggested it
should not even be included in a ranking system. Instead it should not be used at all if
alternatives are available; corporate policy can support such decisions. Panelists also
emphasized that an international ranking system should be a goal; currently risks often are
ranked on a national scale.
Panelists stressed the need for a system that can be replicated on a daily basis by industry
staff, since experts cannot be consulted.each time an environmental category must be evaluated
and ranked. Thus, background data needs to be provided to system users. In addition,
institutional learning must be built into a ranking system, allowing for feedback from within an
organization (e.g., does the system actually provide a meaningful way to make decisions and/or
provide answers that make sense to the people using the system?) as well as beyond an
organization (e.g., does a feedback mechanism exist to determine whether the system is helping
to improve the environment?).
A workshop participant responded that a feedback loop does erast; once companies have
been identified as having a large use or release of a hazardous substance or product, these
companies tend to look for alternatives to reduce the use or release. Also, at Volvo, providing
actual data to designers has been an important feedback mechanism for the EPS ranking system.
-30-
-------
data and the use data can be used in the risk estimation process. To this end, the register has
stimulated annual compilations of use and environmental release data, particularly for water.
This information has been used as a comparison to and supplement for EPA's Toxic Release
Inventory (TRI). Every Michigan business that discharges wastewater to the waters of the state
must file an annual report on the use, discharge, and disposal of chemicals. In Michigan,
approximately 200 businesses report TRI information and 4,600 report information to the Critical
Materials Register. The combined information from both data bases provides a much more
comprehensive representation of potential exposure to hazardous chemicals in the Michigan
environment.
The register also has been used to develop a list of hazardous wastes in Michigan, to
identify additional storage requirements for certain chemicals, to develop a ranking and scoring
scheme for contamination sites to supplement the Michigan Superfund program, and to enhance
pollution prevention efforts. Since the credibility of any program rests on acceptance of its
scientific validity, an advisory committee of environmental specialists from academia, industry,
government, and public interest groups reviews each step in the Critical Materials Register
process.
The register has been used for pollution prevention efforts to estimate potential
exposures for particular chemicals. In a joint project between the Automobile Manufacturers
Association and the Michigan Department of Natural Resources, both TRI and Register data
were used to develop a list of chemicals recommended for pollution prevention opportunities in
the automobile industry. The chemicals and their volumes released into the environment were
used in an assessment of potential exposure to identify which chemicals were used in large
quantities and/or were released at levels of concern. A relative ranking of over a hundred
chemicals was developed based on this information as well as hazard information and
environmental fate properties of the chemicals based on Register criteria. Chemicals were rated
as, high, moderate, or low levels of concern.
Based in part on the automobile industry's own list of chemicals, another list was
developed of over 60 chemicals that the industry will evaluate from a pollution prevention
perspective over the next 4 years. The list will then be reviewed, in part based on new
-29-
-------
in a ranking system simply because there are more data points to which a score is attributed. In
contrast, a chemical for which very little data exists often scores low in a ranking system, not
necessarily because it is not hazardous, but because the data base is so limited that a good
estimation of the hazard is unavailable.
To address this issue of data availability, the Critical Materials Register incorporates a
threshold approach. In evaluating acute toxicity, for example, the system uses a dose-oriented
approach, but not the quantitative dose-response relationship typically used in the risk
assessment process to calculate acceptable permit limits or occupational exposures. The
register's dose information provides a threshold for ranking that identifies and characterizes a
hazard as slight, extreme, or somewhere in between. The chemicals in the register meet
thresholds in a variety of categories, rather than a single numerical threshold level, since data
availability might ftever be adequate to develop a numeric ranking scheme.
The register evaluates carcinogenicity in a similar manner. Rather than examine potency
directly, the system reviews the relationship between the strength of the; evidence and the quality
of the data. The better the quality of the data, the higher the weight ol: evidence is ranked (e.g.,
a known carcinogen in humans would be ranked high). The lowest scores reflect evidence that is
only suggestive (e.g., data from animal studies only, rather than human studies, for a chemical
that is a tumor promoter rather than an initiator of cancer).
Since toxicity has been studied in much greater depth than environmental fate, the latter
presents greater data quality issues in a ranking system. The Critical Materials Register uses the
categories of bioaccumulation and bioconcentration to develop values in the area of
environmental fate. Data may be collected either in the field or in;a laboratory. The register
also includes data on persistence, but in a simplistic manner because evaluation of persistence
data was poor at the time this scoring system was last revised in 1987.
The Critical Materials Register, while not a complete risk estimation process, does
provide a system for hazard identification and exposure assessment. Once this type of
information is compiled, it can be linked with use information, such as generation of a chemical,
discharge, disposal, environmental media, and contaminant type, and both the risk assessment
-28-
-------
control perspective that have been identified as contaminants in the Great Lakes Basin,
particularly the Great Lakes System. The register provides a methodology to screen or prioritize
thousands of chemicals that industry produces and uses.
The register was developed in the 1970s because the state of Michigan saw a need, from a
regulatory standpoint, to reduce the number of chemicals of concern to a manageable size. After
the original register list of 73 chemicals was subjectively developed in 1971, department staff
wanted to remove the subjectivity in the decision-making process, provide consistency and a
framework for the selection of chemicals, and highlight properties of concern about chemicals. A
decision-tree approach was initiated in 1977 for critical areas, such as carcinogenicity or acute
and chronic toxicity. By 1987, chemicals on the register were selected from eight major areas:
• Acute toxicity
• Carcinogenicity
• Mutagenicity .
• Reproductive and development effects
• Bioaccumulation
• Other toxicity (including subacute and chronic toxicity, and phytotoxicity)
• Physical/chemical properties
• Environmental fate
The register now is based on extensive review of the scientific literature for physical,
chemical, and lexicological properties of commercial chemicals., Chemicals reviewed include
those with well-recognized adverse effects or those of specific concern in Michigan. The review
process incorporates a hazard assessment methodology to develop a level of concern for each
chemical evaluated (see Appendix H).
Sufficiency of evidence and data quality are common issues in any ranking system.
Certain chemicals, such as lead, cadmium, arsenic, and polychlorinated biphenyls, have been
studied in great detail. A chemical which has been researched extensively tends to score higher
-27-
-------
sunsetting if it scored high in: (1) toxicity and release and/or production (excluding pesticides),
or (2) toxicity and persistence or bioaccumulation (including pesticides).
Two general principles guided the GW system's development. First, chemicals that pose
the most hazard (1) are highly toxic (although not necessarily persistent) upon acute or chronic
exposure, and (2) could be released in large quantities. These chemicals pose risks through
short-term intentional or accidental releases. The second principle was; that hazards also exist
with other types of chemicals from long-term, low-level exposures, such as the continuous release
from a frequently used household product or a continuous emission to the environment.
The GW system has been used to evaluate a set of 45 chemicals, 19 of which are
generally considered hazardous and are subject to various management activities, and 26 of which
were chosen randomly from a set of 800 chemicals relevant to the Great Lakes Basin. The
evaluation, based on various data sets, electronic data bases, and summary documents, found that
14 of the 19 chemicals identified as hazardous have already been targeted for use reduction and
selected as sunset candidates. Also, 3 of the 26 randomly selected chemicals had been selected
previously for sunsetting.
An advantage of the GW system is its simplicity. A quantitative basis was omitted
intentionally from the system (i.e., the chemicals were not ranked numerically) so that the effect
and exposure endpoint that selected a chemical as a candidate could be; readily identified.
Another advantage is the system's flexibility—a wide variety of data cart be used. The greatest
limitation of the system is its data requirements. Approximately 10 of the 45 chemicals evaluated
thus far could not be scored because of a lack of data. Use of the GW system would be
optimized by access to a good computerized data system with complete records on study design
and outcomes and by linking the system to a program designed to fill data gaps.
The Michigan Critical Materials Register
Gary Hulburt of the Michigan Department of Natural Resources said that the Michigan
Critical Materials Register lists chemicals of high environmental concern from a water pollution
-26-
-------
for sunsetting. Sunsetting is the term used to identity particularly persistent, toxic chemicals for
restriction, phaseout, and eventual banning of their manufacture, use, transport, storage,
discharge, and disposal. Under the GW system, manufacturing processes and products are
considered, as well as individual chemicals, and specific time frames are set for performing the
sunsetting objectives. Sunset chemicals are chosen based on an evaluation of their toxicity and
the potential for exposure to the chemical.
Selection of chemical candidates is a two-tiered process. The first tier is the scoring
system, which involves specific criteria for exposure and toxicity. A large number of chemicals
are reviewed, but not in depth; thus this tier is not resource intensive. The second tier is.a
resource-intensive review stage in which a small group of chemicals are identified as sunset
candidates and data validating the selection of these chemicals are evaluated. Socioeconomic
considerations and alternative products and processes also are reviewed. The result of this two-
tiered assessment is a set of sunset strategies that are product- and use-oriented and guided by a
presumption of a ban.
A variety of endpoints can be used to consider a chemical as a candidate for sunsetting.
Both exposure and effect parameters are included in the GW system, and both qualitative and
quantitative criteria are used to evaluate the parameters. The exposure parameters are
bioaccumulation, persistence, and release and/or production volume, which are combined with
several effect parameters, which include:
» Ecological disruption.
• Acute and chronic aquatic toxicity (based on potency).
• Acute and chronic terrestrial and avian nonmammalian toxicity (based on potency
and an assessment of severity).
• Acute and chronic mammalian toxicity, including reproductive and developmental
effects (based on a weight of evidence evaluation and on potency).
The chemical being evaluated then is classified as a high, medium, or low hazard, or as an
unknown hazard if no data is available. A chemical would be selected as a candidate for
-25-
-------
expressed uncertainty about how to implement the new environmental polity, Volvo approached
the Federation of Industries and the Swedish Environmental Research ]Institute to help develop a
system that would meet the designers' needs (i.e., a system that is applicable in daily decision-
making). Hie Volvo designers wanted an index that could be multiplied with a quantity or other
measurement of material to obtain an environmental value that would allow for comparisons
among various processes and materials. This led to the development of the EPS system.
Probably several such systems will be needed rather than a single system to achieve the desired
goals, said Mr. Wenblad.
The EPS system has many potential applications, including strategic planning, product
design (as discussed above), process waste minimization, optimization of wztste recycling, and
environmental auditing. The goal of the EPS system is to identify the environmental impacts of
different materials and sources of energy. Average values or specific values based on data from
producers can be used. For a Volvo automobile, examples of these two basic inputs into the
system are the steel or aluminum used to construct the hood (materials) and gasoline required to
run the car (energy). Calculations also can be made for various types of fuels as energy options
(e.g., methanol, diesel, and ethanol). Other aspects can be factored in, such as the state of the
economy and safety requirements.
An example of using the EPS system for strategic planning is the comparison of
environmental impacts of an electric car using an on-board gas turbine generator with that of a
current Volvo family car. Additional calculations can be made using the EPS system, such as
values for using different energy sources (e.g., gasoline, versus electrical energy from a power
plant using oil, versus electrical energy from a hydropower plant, versus energy from a
combination of these sources). The EPS system was used as a tool in the development of an
actual electric car.
The GW System
Barbara Glenn of George Washington University (GW) discussed the criteria and system
that GW developed to identify chemicals in the Great Lakes Basin to serve as suitable candidates
-24-
-------
The first step—defining the study's purpose—is key to the EPS system. If the focus is on
environmental load information, then an inventory analysis might suffice. If the focus is on
specific environmental impact categories, impact classification and characterization would provide
this information. If the goal is to attempt a holistic environmental impact assessment, then
aggregation and valuation of impacts also are needed.
Aft6r a materials and process inventory is conducted, the environmental load values are
calculated by first defining the unit effect of the activity being evaluated on each of the five
safeguard subjects. The unit effects are then valuated using a method similar to the willingness-
to-pay concept (i.e., individuals or society decide how much money or effort they are willing to
allocate to avoid the environmental problems identified). Because this step is associated with
monetary trends, aggregated information can be weighted. A value is then set that reflects any
changes in unit effects. Finally, the change in safeguard subjects (i.e., change in the
>
environment) resulting from a particular resource depletion, pollution emission, or human
activity related to product manufacturing is estimated.
By quantifying environmental impacts of an activity, one can estimate the value of using a
given chemical or process or compare a product's overall impact with that of other products.
Since the EPS system can include actual costs, managers will find it useful. Cost differences can
be calculated using values such as whether cost savings or additional expenses will result from
various environmental improvements. In addition, public interest information (e.g., the
environmentally motivated degree of paper recycling) can be included in this system.
Information on uncertainty is then included in the process through sensitivity and error
analysis. Identifying which area has the greatest relative sensitivity (compared to another area
included in the analysis) may indicate that the reliability of certain types of information can be
increased. Error analysis can be used to identify the probability that the statistical significance of
one environmental impact is greater than that of another.
In discussing the practical applications of the EPS system, Mr: Wenblad said that Volvo
developed an environmental policy in 1989 that was not entirely new to the company's
production and environmental staff, but was a new concept for design staff. When the designers
-23- .
-------
The EPS system categorizes information on environmental impacts into five safeguard
subjects:
• Human health
• Biological diversity
• Production
• Resources
• Aesthetic values
An important goal of the EPS system is to provide a simple way to weight environmental
impacts on these safeguard subjects based on a common scale represented by a single unit or
number. The system measures human health by excess death, morbidity, and nuisance (with
different values given to each). Biological diversity includes extinction and threatened extinction.
Biological production is measured by an activity's impact on the production of seed, meat, fish,
wood, or freshwater. EPS measures natural resources based on resulting impacts on the other
safeguard areas if the natural resources consumed by an activity were restored. Aesthetics are
valued on a case-by-case basis.
The EPS system uses a stepwise calculation and assessment procedure, said Mr. Ryding,
and follows SETAC terminology: inventory analysis, impact assessment, and improvement
assessment. The process begins by defining the study's purpose. The next step is a material and
process inventory. Using numbers obtained in the inventory and information on impacts, a table
of key environmental load indices is developed as an initial reflection of the impact of
construction materials and manufacturing processes on natural resources, raw materials,
emissions to air, soil, and/or water, and/or waste generation. Environmental load values then are
calculated for use by product designers (environmental load index X quantity = environmental
load value). The higher the environmental load value, the greater the environmental impact.
Sensitivity and error analyses also are then performed to avoid making; false conclusions because
of uncertainty of data. The last step is to make comparisons and evaluations of environmental
impacts. The EPS process is discussed in more detail below.
-22-
-------
The EPS Scoring System
Sven-Olof Ryding of the Federation of Swedish Industries and Axel Wenblad of AB
Volvo Corporation described the Environmental Priority Strategies (EPS) system for product
design, which is a conceptual method for ascribing numeric values to various types of
environmental impacts included in life cycle assessments. Overheads from Mr. Ryding and Mr.
Wenblad's presentations are provided in Appendix G. The EPS system supports an ongoing
Swedish project called the Product Ecology Project, which is being refined by Volvo and 11 other
companies. Mr. Ryding presented the framework for the EPS system and associated assumptions
in conducting evaluations. Mr. Wenblad then discussed the practical applications of the EPS
system.
EPS developers identified specific criteria that environmental impact valuation methods
should meet, based on discussions with product designers and on international studies of
environmentally sound product development. These criteria, which the EPS system strives to
include, are:
• A global approach (to address the international nature of raw material acquisition
and trade), as well as a regional or local approach.
• Ability to replicate procedures (i.e., calculations should be able to be repeated
with each valuation method).
• Allowance for sensitivity and error analysis.
" Support and validation for judgments about environmental effects.
• Flexibility (i.e., new scientific findings and inventory data should be incorporated
easily).
• Transparency and consistency (i.e., all separate calculation steps should be clear).
-21-
-------
information, energy impacts, resource conservation, global issues, or other areas that help
characterize a particular use application and associated tradeoffs. Thus, a use cluster assessment
of substitutes includes several tiers. This approach depends heavily on the development of the
science of comparative risk assessment. The DfE program is conducting related pilot projects
with several industry groups.
PANEL SESSION HI: INSTITUTIONAL APPLICATIONS
As products become more complex, designers must reexamine ithe interplay between cost,
reliability, quality, and marketability for their products in the global marketplace, said Robert
Ferrone of Digital Equipment Corporation, moderator for Panel Session 1H. In today's
competitive climate, designers often must compress the timeframe in which a product is brought
to market. Consequently, a plan for production and distribution must be developed as early as
possible in the product design stage. The product development process is further complicated by
the need to consider global environmental factors.
Given the lack of definitive decision tools in the area of environmental risk and impact
assessment, designers are learning to integrate environmental considerations into product design
by experimentation. As a result, companies and researchers are beginning to develop their own
environmental ranking systems to assess the environmental risks and impacts of their products.
Three examples of ranking systems that have been developed by industry and researchers—the
EPS Scoring System, the GW System, and the Michigan Critical Materials Register—are
described below. Other ranking systems are discussed in several appendices (Appendices D, I, J,
K, L, and M). Panel members considered the following questions in their presentations on the
ranking systems they have developed:
• What barriers have slowed the introduction of ranking systems?
• Is it possible to develop an international ranking system?
-20-
-------
substitutes exists for use in performing a function. Once a particular function is defined, which
can be done in a variety of ways, a set of substitutes also can be defined, and a tradeoff analysis
can be performed. Paint stripping is a good example of the importance of defining function.
The set of substitutes for paint stripping is not the same for different types of uses, such as
household and commercial use. For example, sandblasting and other commercial techniques
probably would not be used to strip paint off household furniture.
Once function is defined, whether for processes or chemicals, different ways to
accomplish the function often are identified that previously were ignored. In the case of dry
cleaning, for example, the process of defining function has led to the consideration and
evaluation of nonsolvent alternatives. Focusing on the use cluster, or array of substitutes, creates
a kind of synergism that differs from more traditional approaches to pollution prevention (e.g.,
waste minimization in a particular technological area). The use cluster approach also is useful in
integrating a wide variety of issues (e.g., scientists' concerns with risk and industry's concern with
accomplishing a task). Use cluster analysis considers the type of people affected (such as users
of a product, manufacturers, environmental groups, and regulators) in examining a function and
the variety of available options. Tradeoffs associated with a particular application are examined
so that everyone can agree on and understand them.
Use cluster analysis might involve a quick compilation of known information or a study
design to perform further work. In the printing industry, for example, products are being tested
to generate actual performance data, and risk assessment and characterization professionals are
providing comparative risk information. The industry then can review this risk information and
the concerns associated with a particular chemical use; If, for example, industry wanted to use a
particular chemical substitute (e.g., terpene instead of a solvent) that has associated ecological
risks (e.g., aquatic toxicity from terpenes), then industry might decide to use the substitute but
manage the accompanying risks.
Some principles of life cycle assessment are incorporated into use cluster analysis—risks
from the manufacture of a chemical, the product under consideration, the particular use
application, and examination of disposal methods. In addition, information on control
mechanisms and alternatives are evaluated. Additional areas also can be included, such as cost
-19-
-------
Dorothy Patton
Executive Director and Chair
Risk Assessment Forum
U.S. Environmental Protection Agency
401 M Street, SW
Washington, DC 20460
202-260-6743 -
Fax:202-260-3955
Ward Penberthy
Chief, Chemical Engineering Branch
Office of Prevention
Pesticides & Toxic Substances
U.S. Environmental Protection Agency
401 M Street, SW(TS-779)
Washington, DC 20460
202-260-1730
Fax: 202-260-0981
Lena Perenius
Principal Technica^Officer
National Chemicals Inspectorate
P.O. Box 1384
Solna, Sweden S-171 27
46-8-730-5700
Fax: 46-8-735-7698
Peter Preuss
Director
Office of Science Planning &
Regulatory Evaluation
U.S. Environmental Protection Agency
401 M Street, SW
Washington, DC 20460
202-260-7669
Fax: 202-260-0106
Beth Quay
Director, Environmental Technical
Affairs
The Coca-Cola Company
One Coca-Cola Plaza
Atlanta, GA 30313
404-676-6629
Fax: 404-515-4280
Peter Radeck!
Coordinator
EPA Center for Clean Industrial
Treatment Technology
Michigan Technological University
1400 Townsend Drive
Houghton, Ml 49931
906-487-3143
Fax: 906-487-3292
Ann Robinson
Advisor
International Programme on
Chemical Safety
World Health Organization
80 Quebec Avenue - #601
Toronto, Ontario M6P 4B7
Canada
416-762-6488
Fax:416-762-6488
Mark Rossi
Research Associate
Toxics Use Reduction Institute
University of Massachusetts @ Lowell
One University Avenue
Lowell, MA 01854
508-934-3297
Fax: 508-453-2332
Sven-Olof Ryding
Assistant Professor/
Senior Scientific Advisor
Environment & Energy Department
Federation of Swedish Industries
Box 5501
S-11485
Stockholm, Sweden
46-8-783-8111
Fax:46-8-662-3595
Jacinthe Seguin
Head, Programs & Policies
Office of Waste Management
Environment Canada
Ottawa, Ontario K1A OH3
Canada
819-953-1112
Fax:819-953-6881
Jamine Sekutowskf
Technical Manager
AT&T Bell Labs
P.O. Box 900
Princeton, NJ 08542
60,9-639-2501
Fax: 609-639-2851
Ken Sexton
Director, Office of Health Research
U.S. Environmental Protection Agency
401 M Street, SW(RD-683)
Waishington, DC 20460
202-260-5900
Fax:202-260-0744
Karen Shapiro
Research Associate
Tellus institute
11 Arlington Street
Boston, MA 02116
6r/-266-5400
Fax: 617-266-8303
Anne Smith
Vico President
Decision Focus, Inc.
1155 Connecticut Avenue, NW - #400
Washington, DC 20036
202-429-6506
Fax: 202-466-3428
Hum Sipitzer
Associate Director
ILSI Risk Science Institute
112616th Street, NW
Washington, DC 20036
202-659-3306
Fax:202-659-8654
Sharon Stahl
Design for the Environment
Office of Pollution
Prevention and Toxics
U.S. Environmental Protection Agency
401 M Street, SW(TS-792)
Washington, DC 20460
202-260-2718
Fax:202-260-1764
B-5
-------
Richard Thomas
Director, Toxicology and Risk
Assessment
National Academy of Sciences
2101 Constitution Avenue, NW
Room HA 354
Washington, DC 20418
202-334-2616
Fax:202-334-2752
Maria LivIaTosato
Doctor in Chemistry/Director of Research
Instituto Superiore di Sanita'
ViaieRegina Elena 299
Roma, Italy 00161
39-6-4990-790
Fax: 39-6-4440-140
Roger Tregunno
Department of the Environment
43 Marsham Street - Room A3-34
Romney House
London, England SW1P3PY
71-276-8326
Fax:71-276-8333
Tina Vaarsberg
Consultant
Environment, Technology and
Science Policy
Sandia National Laboratories
480 L'Enfant Plaza, SW
Suite 11-135
Washington, DC 20024-2197
202-646-4451
Fax:202-646-4402
Bruce Vigon
Research Leader
Battelle
505 King Avenue
Columbus, OH 43201
614-424-4463
Fax:614-424-3321
Vanessa Vu
Deputy Director
Health & Environmental Review
Division
Office of Pollution
Prevention and Toxics
U.S. Environmental Protection Agency
401 M Street, SW
Washington, DC 20460
202-260-1243
Fax: 202-260-1283
Randy Watkins
Senior Member of Technical Staff
Sandia National Laboratories
Box 5800
Albuquerque, NM 87185
505-844-3387
Fax:505-844-0116
Mary Ellen Weber
Director of Economics,
Exposure, and Technology Division
Office of Pollution
Prevention and Toxics
U.S. Environmental Protection Agency
401 M Street, SW (TS-779)
Washington, DC 20460
202-260-0667
Fax: 202-260-0981
Arthur Weissman
Vice President of
Standards and Planning
Green Seal
1250 23rd Street, NW
Washington, DC 20037
202-331-7337
Fax:202-331-7533
AxelWenblad
Director, Environmental Affairs
AB Volvo Corporation Headquarters
S-40508
Gothenburg, Sweden
46-3-159-1966
Fax:46-3-159-1044 ;
James Wilson
Regulatory Issues Director
Monsanto Corporation
800 North Lindbergh Boulevard
St. Louis, MO 63167
314-694-8879
Fax: 314-694-8808
R.S. Woodward
United Kingdom Health &
Safety Executive '•
Chepstow Place - Room 456
HPDC2 Baynards House
London, England W2 4TF
44-7-124-3626
Fax: 44-7-124-3603 :
Joseph Yang
Mobil Oil Corporation
P.O. Box 1029
Princeton, NJ 08543
609-737-5547
Fax: 609-737-5570
John Young
Scientific Director
The Hampshire Research Institute
1600 Cameron Street
Suite 100
Alexandria, VA 22314
703-683-6695
Fax: 703-684-7704
B-6
-------
APPENDIX C
MISSION STATEMENT
-------
-------
EPA
Evaluating Cleaner
Workshop on Identifying a Framework for the Future
of Human Health and Environmental Risk Ranking
Washington, DC
June 30-July 1,1993
Mission Statement
Our goal is to establish a network of people who are interested
nr^0™^**?*"?*0™ °f technol°gfe*, Processes and
*v^±' TlSef * deVel°P a Simple' Seneric methodology for
evaluating technologies, processes and products that can be
translated to multiple uses. In order to develop our
methodology, we will:
• Identify the linkages between and strengths and
weaknesses of existing tools for environmental
evaluation, including life cycle assessment, substitute
assessment, and risk assessment.
• Identify how different frameworks for evaluating the
environmental impacts of technologies and products
fit into a scheme of valuation and decision making.
• Make transparent the environmental value choices
that are involved when one process or product is
chosen over another.
c-i
-------
-------
APPENDIX D
BACKGROUND MATERIALS
-------
-------
EPA
I
Evaluating Cleaner Tg
Workshop on Identifying a Framework for the Future
of Human Health and Environmental Risk Ranking
Background Materials
D-l
> Printed on Recycled Paper
-------
INTRODUCTION
TABLE OF CONTENTS
1
Life Cycle Assessment • • • • ' '
2
Health Risk Assessment '
3
Cleaner Technologies Substitute Assessment
4
Congressional Record - Risk Reduction
Summarized from Supporting Environmental Quality: • • • • •
Developing an Infrastructure for Design
Examples of Existing Ranking Systems:
J_ . 11
EPA Office of Pollution Prevention and Toxics •.
Use Cluster Scoring System
12
German Ecological-Controlling System
13
Michigan Critical Materials Register Program
14
Swedish EPS Enviro-Accounting Method
Criteria to Identify Chemical Candidates for
Sunsetting in the Great Lakes Basin !
16
United Kingdom Priority Scheme for Existing Substances ,
GWU Chart on Comparison of 8 Ranking Schemes by GWU •:
D-2
-------
Life Cycle Assessment*
£CA) IS m °bjeCtlVe Pr°cess to evaluate the environmental burdens
,*n n ' Pr°CeSS' °r activity by iden%in9 and quantifying energy and material
usage and environmental releases; to assess the impact of those energy and material usW^d
releases on the environment; and to evaluated implement ^ortunffief to effTrt
ZIT^T0^6"13- The assessme"t i^'"des ^entire life cycle oHhe producf
process, or act.v.ty, encompassing extracting and processing raw
transportation, and distribution; use/re-use/maintenance; recycling;
A complete life-cycle assessment consists of the following separate, burt interrelated components:
An °bJeCtiVe' data'based Proc«* of quantifying energy and
raw material requirements, air emissions, waterborne effluents, solid waste and
other environmental releases incurred throughout the life cycle of a product,
Life-Cycle Impact Analysis - A technical, quantitative, and/or qualitative process
to characterize and assess the effects of the environmental loadings identified in
the inventory component. The assessment should address human health
ecological health, and resource depletion
{-^-Cyc/e /mprovemenf Analysis - A systematic evaluation of the needs and
opportunit.es to reduce the environmental burden associated with energy and raw
material use and waste emissions throughout the whole life cycle of a product
process, or activity. This analysis may include both quantitative and qualitative
imProvement- such » ch^ges in product design, raw material use
processing, consumer use, and waste management.
°entS
UP
approach that, when combined with other
information needed to maximize environmental
* * r ;.7."- could assist LCA practitioners in the impact and improvement
nt stages by facilitating the assessment of the environmental impacts impr°Vement
the evaluation and selection of improvement opportunities.
Society for Environmental Toxicology and Chemistry.
D-3
-------
Health Risk Assessment .
risk assessment is a characterization of the potential adverse health effects pf human
poues^to^nvi%nmen;Sa. hazards. Health risk assessments consist of *efo^,n^ basic
Cements: hazard identification, dose response assessment, exposure assessment, and nsk
characterization.
a qualitative assessment of whether there is a relationship between
a hm ted w and an effect. In assessing the likelihood that a substance causes
toxic TSaliSng evidence is considered, including human evidence, evdence from
animal studies, and any supporting evidence.
In the dose-response assessment, one attempts to identify a quantitative relationship between
maanttude of a response and the dose of the stressor inducing such a response. Since data
TeKavaHa^e on the Sects of chemicals at the level of environmental exposure, usually one
^^ft^Ihe measured data and infer the response at lower (environmental) levels.
SHS^^S^^ extrapolate the dose-response relationship, and drffering results
will be obtained depending on the model chosen.
The exposure assessment looks at the levels of exposure people are likely to ^Perience in the
real world The exposure assessment must consider where the substance is found, the routes
of exposure (e g Sr, water, food), the number of people who are exposed, and the'magnitude
dumtion ^d^ ng of the exposure. One also must decide whether to conduct t^ exposure
SSme^forTheierage population, orfor the population that is most exposed or that ,s most
at risk.
The risk characterization integrates and summarizes the information foundI in^the hazard
Identification, dose response assessment, and exposure assessment to develop publicDearth
^eeflmato. The risk characterization defines the significance of the risk but also presents
?he Sumptions, uncertainties, and scientific judgements that were used to develop the nsk
estimate.
Ecological Risk Assessment utilizes a different framework based on comparable principles but
different terminology and a different approach.
D-4
-------
Cleaner Technologies Substitute Assessment
EPA's Design for the Environment Program has developed a methodology for examining
substitute chemicals, processes, and technologies. This methodology currently is being applied
in the DfE program's cooperative industry projects. Each project has identified, with the help of
industry partners, problem environmental areas for which companies would like to seek
substitutes. . . •
The first step in the methodology is to identify use clusters for the product or industry at issue
A "use cluster" is a set of chemicals, processes, and technologies that can substitute for one
another in order to perform a specific function (e.g., metal parts cleaning). Next all of the
industry use clusters are ranked using an EPA system that incorporates factors such as human
and ecological risks, exposure, regulatory interest, and pollution prevention opportunities The
scoring system helps to prioritize and direct research into more environmentally beneficial
alternatives. Where the company knows that they want to target research into one process
these first two steps are not needed.
Finally, a substitute assessment is conducted for the high priority use clusters. Through a
process of collecting information on currently existing alternatives and through a search for other
promising options, the DfE program lists all alternatives in a "use cluster tree" for chemicals
processes, and technologies that can substitute for one another in performing a particular
function. Drop-in substitutes tend to be rare. Therefore, it is essential to compare systematically
the trade-offs associated with different alternatives.
Cleaner Technology Substitute Assessments (CTSAs) are intended to provide a flexible format
for systematic comparison of the trade-off issues associated with alternatives. Traditional trade-
offs such as cost and performance are brought together with environmental trade-offs, including
comparative nsk, releases, energy impact, and resource conservation for each alternative A
completed CTSA should provide all the information that a designer needs to decide amona
alternatives. a
D-5
-------
Congressional Utcord
V?£££? PROCEEDINGS AND DEBATES OF THE 103 * CONGRESS, FIRST SESSION
.—-——-—--=S=2S=======
WASHINGTON, January 22,1993
Vol 139
Senate
RISK REDUCTION
*• Mr, MOYNIHAN. Mr. President,
environmental decisions are difficult in
say circumstance, and especially so
•when the economy is weak and:there is
competition for resources both, in the
private sector and Government. People
want a clean environment, but they don't
want to pay more than is necessary.
We cant do everything at once. The
question of priorities arises almost
unbidden. We would rather not have to
think this way, but there you are. The
choice is not between, having priorities or
not having them. Rather it is between
setting them consciously or setting them
by default. Thus, I rise todery to
introduce a bill that seeks to establish the
framework needed to ensure that
information needed by an informed and
involved public to set workable priorities.
This legislation, the Environmental Risk
Reduction Act of 1993, is an update of
the Environmental Risk Reduction Act of
Z991.
I am.convinced that risk ranking and
cost-benefit analyses are valuable tools
L
for making environmental decisions. They
offer the means to set priorities and to
measure success. They are not our only
tools, but nor can they be ignored. It
took over half a century for the dedicated
public servants in. the Bureau of Labor
Statistics and the Council of Economic
Advisors to learn- how to fashion
economic indicators, ft wfll no doubt take
a long time to develop reliable method-
ology to assess costs and benefits of
environmental regulations, and to reEably
measure the ..risks associated with
environmental exposures. But this is no
reason not to begin. If we dorft start
now we wfll riever learn. And it must not
be forgotten that there are sonle things
that science cannot tell us. Values, for
instance: Do we care more about this
species or that one? Or fairness. Some
questions are a matter tor the legal
system, not the scientists.
But we do have problems. Dining the
Presidential campaign and transition,
much discussion focused on the economy,
getting and keeping it going. Dr. Paul
S550
D-6
-------
Portney of Resources for the Future
testified at a hearing last September on
my bill S. 2132, the Environmental Risk
Reduction Act that compliance with
environmental laws was about $130
bflKon per year-something like 2.2
percent of GNP. Our nest closest
competitors in terms of environmental
protection, perhaps not surprisingly, are
Germany and Japan, each spending
about 1.6 to 1.8 percent of GNP. 'While
this may not be too much money to
spend on environmental protection it is
too much to spend unwisely. Obviously,
we are seeing a new trend. Federal
environmental laws are being questioned.
State and local governments are signaling
that they can't afford to comply with all
environmental laws. Their resources are
finite and must make do for competing
needs; for example, mflintflfrrfng roads
and buildings, providing social services.
and schooling.
Dr. Edward Hayes of the Ohio State
University testified last September
before the Committee on Environment
and Public Works about the experience in
his State. The city of Columbus, OH,
recently set out to analyze with as much
precision as possible the impact of
Federal environment laws during recent
years. City leaders wanted to know what
effect those changes would have on the
city's budget. The findings were reported
in "Environmental Legislation: The
Increasing Costs of Regulatory
Compliance to the city of Columbus." It
turns out that new environmental
initiatives win cost the city of Columbus
an additional $1.6 bflEon over the next
decade-an extra $856 per year of
increased local fees or taxes for evtry
household in the city by the year 2000
A followup study, "Ohio Metropolitan
Area Cost Report for Environmental
CompfeTice," slipwed a similar impact in
eight other Ohio cities.
As iar as we can tell this pattern is
being repeated in other places. Just
other places. Just this week a bipartisan
group of 114 naayiora from towns and
aaes across the United States sent each
Member of Congress a letter and report
warning of an impending fiscal crisis in
trying to pay forthe increasing costs of
environmental inandates. The President
and "Vice President were sent a copy of
the letter and report when they took
office yesterday. The editorial in the
January 8 issue of Science alerts us to the
"growing questioning of the factual basis
for Federal command and control actions,"
all because of conaans over regulatory
costs. *
The message is clear. State and local
obligating them to spend their resources
on Federal requirements. They will want
proof that there is. a problem and
confidence that ihe legislated solutions
will solve it. California's threat to give
enforcement of its drinking water
program back to EPA last spring speaks
volumes. The% imosr environmentally
advanced State in the Union close to
rcbelhon-a semiring prospect. The
Saence editorial suggests that we are
seeing the "beguuiing of a revolt."
Clearly risk ranking and cost-benefit
analyses aren't the sole factors for
dedaontnaking. Social concems~who
should bear risk for whose benefit-
pubEc preference, basic fairness must be
considered, too. Truth be told, I suspect
that environmental d
-------
field, but we are being overtaken by
events. The questions that are key to
making decisions about the environment
are slowly but surely becoming more
knowable. . .
There are those who discount relative
risk ranking entirely because the assump-
tions needed to assess risk are myriad.
Facts often seem scarce. What ***™
exposed to? What results does tins pro-
duce? What portion of the population is
affected? Or might be affected? There are
no precise answers to these questions.
And so risk assessments ate controversial,
at times very controversial. But we cannot
forget that knowledge need not be preos e
to be usefuL
Relative risk ranking and cost benefit
analyses are tools. Crude tools today,
yes, but perhaps they are sufficient in
some cuesunrt activity "A" as more
risky than activity "B." If costs or political
realities dictate That we should control
"BM before "A" then great. But let us have
the courage and foresight to make a
conscious decision. In public. With people
watching. History shows Hurt crude
tools -give way to more refined ones.
Experience teaches.
Tin introducing the Envnonmental
Risk Reduction Act, to help us learn how
best to practice the trades of
environmental risk assessment and
cost/benefit analyses. The bul wfll put
into law the major findings of the 1990
-Reducing Riak" report by EPA's Science
Advisory Board [SAB]. I agree with
former EPA Administrator waiiam
Rally's belief that science can lend roach
needed coherence; order, and integrity
to costly and controversial decisions.
America's environmental laws are a
large and diverse lot. We have only two
decades of experience on this subject, and
we are still learning, feeling our way.
The relative risk ranking and cost/benefit
analyses called for in this bfll provide
some common ground for looking at our
erwironmental laws. The bill also
provides the public and Congress with
access to the findings. The "Reduction
Risk" report states that "relative risk data
and risk assessment techniques should
inform (the public) judgment as much as
possible." Not dictate it, but inform it.
All this wfll take time, decades
perhaps. But tec us take heart. Questions
that seem difficult now can with a certain
amount of effort yield to the scientific
method.
I urge my colleagues to support this
bill and ask unanimous consent that the
text of this bfll and the Science editorial
be printed in the Record at this time.
There being no objection, the material
ordered to be printed in the Record, as
follows: (jMJCt available)
D-8
-------
Summarized from "Supporting Environmental Quality:
Developing an Infrastructure for Design"1
By Braden Allenby
^ I2S Q^^V'-onmental Management (TQEM) and related technical programs to
* 'S unaooePtable for governments and private firms to continue to relv
ST!**^ know to be inade<"Jate <**" " ""—I*. *K & *
approaches such as -We oyole assessment- (LOA)' or "DesigJ i
methods be9in to i"oorporate »• -«
"* Criti0al * ^ (U- ^""mentally appropriate)
DFE and similar methods cannot be implemented by private fii-ms without the creation
hSiV
D-9
-------
There are several fundamental reasons for this situation, which are inherent in the
structure of our economy:
1.
2
r post vsJue Judgments
that have yet to be resolved-or, in many cases, even 'dentrfied Thus for
eTampteaS analysis of the desirability of substituting ind.um ^ bis "*h£*p *
lead solder in printed wiring board assemblies lead directly to difficun nsK
dilribution questions.6 Environmental impacts of increased use of bismuth and
indium would fall on localities where mining and processmg occurred (some
outsideThe UnLd States), whereas benefits would accrue pnmanly o localities
nSfanSs Incinerators where the electronic items containing aHoys ^ouW
be deposited. This asymmetrical geographic distr.but.on of nsks and benefits
evidentthroughoutthe environmental area (e.g., in waste d.sposal patterns in and
among nations), clearly raises unresolved value quest.ons.
3 Anv economy is defined in large part through the legal infrastructure that supporte
f DFE ^green" manufacturing potentially constitute a far more fundamental
change in the economic structure than most people recognize. Accordingly, rt
should come as no surprise that implementation of such PractlC6S 'f f**
number of fundamental legal questions and, converse^ many ex.st.ng legal
Sures have at least the potential to affect private DFE act,vit.es.
Obviously, existing environmental laws that presuppose linear manufacturing activities
components, subassemblies, and products.
D-10
-------
wrona- hh Social 9oals embodied in these legal structur«s-and many othe/s-^re
SUMMARY
ras.
a DFE inf astruct and tO
a ui-t intrastructure, and begin working aggressively to provide such an
Partnering among stakeholders both in the Unrted States and in othVnatior^
European Community and Japan, should be integral to any such effort'
requin
uro
the
*•
D-ll
-------
1 Total Quality Environmental Management, Spring 1993, p. 303-308.
Prevention Review 3(3):1 67-80.
Ausubel
3 Robert V. Ayers, 'Industrial Metabolism," in Technology 'and
and H.E Sladovick%ds. Washington, DC: National Academy Press, 1989,
4 SETAC (Society of Environmental Toxicology ^anc I Chemtetry), A Technical Framework for
L/fe-Cyc/e Assessments, Washington, DC: The SETAC Foundation, 1991.
the activities of the AEA DFE Task Force, chaired by the author.
6.
, Design for Environment (died In note 2).
D-12
-------
EXAMPLES OF EXISTING RANKING SYSTEMS
EPA Office of Pollution Prevention and Tadca;
Use Cluster Scoring System
Determining which
d
to identify potential risks is a tem tak
concerns] is a promising approach to prtoL^a chfmS? °r ""£"* chemlcals ** potential
been developed in the pit An eSe of 2 ™ f "T S?veral rankl"9 ^terns have
Ranking System (HRS) which is used KPA^^ °Peratlonal ranking system is the Hazard
for Superfund site cleanups Howete? vlvTet L^ Sltef °n the Natlonal Prioritles "« (NPL)
concerns of chemicals thatsubsTSo^
duster Scoring System (UCSS) is a^tem th^ or US6' Th° ^e
(OPPT) has elated chemic^s
offices, government agencies ! oMrrtere^ ^D^ieT-^f ^ T °n the request of other EpA
selected materials, such as those on the K?n ? f ^ °f a targ6ted effort to ev^"ate
createdamanageabtesetlf chemio^ "PP™* has
has not comprehensively addressed potent I tScilstfel ?^f ?r^ reaCt'V6 process tnat
w.ll use in the future to systematically ideX and Srll UCSS ls one too! that oppT
of chemicals in commerce. aT'Ca"y iaen^ ™d screen concerns relied to a greater number
seto chemicals and uses, rather than
The scoring algorithm uses readify a\S method * determine risk.
used in similar applications. It is no^ded^o^^^ ^ **. ?r°UpS Of chemicals
does it generate precise risk numbers s"bstrtute for a detailed risk analysis, nor
D-13
-------
German Ecological-Controlling System'
-. The
ogical-conUoUIng method Jsused
^^
improvement.
sr-.sasBftSB3a.-ssr
criteria:
known.
compliance with laws and regulations
public demands
'normal1 ecological effects
-air pollution
-water pollution
-soil pollution
-toxicity
ecological risk potential
internal environmental cost
product life cycle
-obtaining raw materials
-initial production
-use
-waste disposal
-recycling
waste rate/added value
(A rating of "A" is shown in red, "B" in yellow, and
Ecological
D-14
-------
Michigan Critical Materials Register Program
Overview
unique list of
Advisory Committeo
• from academia,
Director of the DNR to provide advice on the
•|~ — - —•—• • ••»•*-•• wwi i%wi wi i LI its LcurnmiiTBO nfr\\iif+Ao 41*** **.»*^.^i*i^mt*t- *owi n iiOol
Scoring .,-,".'•
^
1 987, contains 28S ootnpound rdm of o, T l'S'Ster' based on c"'teria rsvised
*a "level of concern" of 5 In two or more criteria
an additive "level of concern" greater than 15
support document may be
P.O. Box 30273, Lansing Michigan 489(2. Resources, Surface Water Quality Division,
D-15
-------
Swedish EPS Enviro-Accounting Method*
the quality of life.
the following possible safeguard subjects:
• human health
• biological diversity
• production
• resources
• aesthetic values
Production is measured based on the amount ol ^""£^ MP ' resources according to
compare it with other products.
Swedish Environmental Research Institute. Draft Report, 1992.
D-16
-------
Criteria to Identify Chemical Candidates Ifor
Sunsettlng in the Great Lakes Basin*
Potential exposure is assessed in three categories:
aCCUmUlate ln the «
Ind
- biota for
the persistence of the chemical in the environment
the amount of the chemical that is produced and/or released to the environment.
im"acte
of growth
Chr°nic Chemical Sin9ular
terrestrial and avian, non-mammalian species
or
lnoludsd
HIGH in
in
D-17
-------
United Kingdom Priority Scheme for Existing Substances'
requiring minimum direct expert input.
scheme focussing on human heatth e«ec* u«lizes four .ftp. of data to
for a chemical substance:
Priori*
toxicity
physico-chemical properties
tonnage
use pattern
"*"
*
°"
%tt£Z££3L pa^eters are used:
• toxicity (aquatic organisms and mammals)
• potential for bioaccumulation
• degradation
• use pattern
• quantity produced of imported
,u, each parameter is multiplied to give a final total score. In
PO«aflyPThTmost influential parameter; however, no s,,
parameter can dominate the scheme.
presented in a priority list.
Association, 071-276-8321.
D-18
-------
Comparison of
Great Lakes
Woler Quality
Agreement
(Annex '1
List of
Substances
1989.
(GLWOA)
Wei-notional
Joint
Commission
of the United
Slates and
Canada
Binoliono!
Objectives
Committee.
Effluent
Monitoring
Priority Po&jlanfs
List(1989)
Update Draft
(EMPPL)
Ontario
Ministry of the
Environment.
• Ranking schemes by George Washington University
Compiled from:
1986 Working
List of
Chemicals in
the Great Lakes
Basin. Ontario
MSA Inventory
(1989).
U.S. EPA Toxics
Release Inventory
Oalabose(l989).
294 substances
(listed.
Consists of
hazardous
chemicals delected
in efftjents
or surface
waters, of
30lenlid concern
but not yet
detected, or idenlified
in certain
industrial sectors.
To identify
substances for
development of
Specific Objectives.
List #1 Substances
present and toxic in
the Great Lakes.
Usl #2 Candidates
for oddtional
study.
List jfZ Candidates
for additional
monitoring in the
Great Lakes system.
'Development
of chemical
specific monitoring
regulations under
MSA.
I Classification
Substances
classified
according to
occurunce in the
Great Lakes
Basin and
whelhir data is
sufficiimt to
define them as
toxic.
1. Chemicals
receive a
score f«r
each
'parameter.
! •-
2. Subslonces
are not
prioritized
within the
fat
[Remarks
Estimated. Smiled or
qualified data as
wet as SAR and
nondefinitive tests
be used to
substantiate potential
toxicity.
Professional judgement
is used to
evaluate the
adequacy/utility of
individual studies.
Chemicals ore placed
on the fsf if
exposure criteria are
met ,ond any one of
the parameter scores
equals or exceeds
a specified value.
Qualifier tags are
used to mark scores
based on
questionable date
or SAR.
Candidate-
Substances
List for Bans
or
Phase-Outs.
1992
Ontario
Ministry of the
Environment.
List of 2 1 primary
and 46 secondary
substances that are
highly hazardous
and present in or
discharged to
Ontario's surface
waters.
Michigan
Critical
Materials
Register
(MCMS)
Michigan
Department of'
Natural
Resources
SSiXtif ?<£
Chemicals of high
environmental
concern from a
water petition
perspective which
may be used.
discharged and/or
disposed of in
Michigan.
More than 400
chemicals are listed.
Identifies
candidate
substances for
multimedia
release
reduction,
banning and/or
phase-out.
Establishes a
fat of Critical
Materials for
which businesses
must report use
and/or discharge
information.
Chemicals were
Bsted if they
were given the
highest possible
score for one or
more toxidfy
parameters and.
were defined as
persistent and
bioac-
cumulalive.
Chemicals
receive a score
for each
parameter.
Chemicals ore
placed on the
list if
"sufficient
evidence" exists
for any
parameters or if
a specified
total score or
combination of
scores is
attained.
Substances ore
not prioritized
within the list.
Criteria to
Chemicals were
screened using the
MOE Scoring System.
Study design &
reporting must
approximate certain
guidelines to be
used in hazard
assessment.
SAR not used.
Chemical
-------
Chenicoteof
Environmental
Relevance
(BOA I & I)
|Soc!ely of
German Chemists
(GOCh) Advisory
Commillee on
Existing
Chemicals of
Environmental
Relevance
WWS Scoring
System
(WHS)
Directorate Genera!
for Environmental
Prolection-
Nalherlonds
Toxic Substances
Control Act
Scoring System
(TSCA)
Comprehensive
Environmsnlol
Response.
Compensation.
and UobiSly
Acl of 1980:
Supeffund
Amendments
and
Reovlhorizalion
Acl of 1985.
(CERCIA)
Reporloble
Ouonlily
Ranking
Process
(Substances of
[environmental
Irdevonce selecled
from other priority
Ists of chemicals
which occur in the
environment and ore
induslrialy
important.
List t Substances
rating high for
environmental
exposures.
persistence. &
jiotegicol effect.
Ustfc
nondegrodobte
substances rating
high for effect
potential or high
environmental
exposure.
378 chemicals
compiled from
international and
national isls
of chemicals
known or suspected
to be hazardous
to the environment
or to humans, or
of high exposure
potential.
Environmental
Protection
Agency.
Office of Toxic
Substances.
Existing Chemicals
(Program
Environmental
Protection
Agency.
Emergency
Response
Division.
Office of Sold
Waste and
Emergency,
Response.
Chemicals ore
not Esled.
Scoring system is
used on a
continual basis.
List of about 724
CERCIA hazardous
substances and
I waste slreams
identified by the
statutes and
adjusted reporlable
quantities.
rpose
lot List
Indentifies
existing chemicals
of health or
environmental
Irdevonce for
which to
I consider regutolcry
1 action.
Remarks
Priorilization of
chemicals for
investigation &
development of
environmental and
health policies.
204 chemicals scored
by the EPA Office of
Toxic Substances (as
of 6/89).
Chemicals scored to
assess hazard and
identify data gaps.
_ .
The ranking process
establishes the
reporlable
quantities (ROs) for
CERCIA hazardous
substances that
triggers
Inolificalion
requirements to the
National Response
Center upon release
of the materials.
LJOssmyuim" I
-4-
Chemicals receive The
a score for '"d
ea
Ch
sel
on
sc
CO
fo
&
pt
:h paromeler. of
im|
emicols Pr
ecled based so
specific
ore Pe
mbinolions '«*
r exposure °'
effect b»
ramelers. Pr
&
or
S
d
f
Ihemicals
receive a score
for each
parameter.
End scores ore
calculated by
combining media
specific
exposure &
effect scores &
subtracting from
the maximum
score
attainable.
Lower scores
represent higher
Ssl
kjdes chemico's 1
industrial
portonce and (hose
Bsenl or probably
in the environment
slicides,
irgonic chemicals
nalurd origin. '
products of
eduction.
unstable chemicals
e excluded.
AR used to score
nernicols with
ogs.
_^— —?—.—— —^———1
^ change of 1 in a
score value reflects
a change of one
order of magnitude
n the exposure or
effect levels:
SAR used to estimate
criteria values.
Opportunity provided
to review exceptions.
hazard. 1
'
1 .
I Chemicals
receive a score
for each para-
meter.
Scores ore
SAR is used to
estimate paromeler
values and the
score is tagged.
1
rot weighed or 1
I combined. 1
1 1
i
Substances are
scored for
several
oaramelers. RO
|is assigned
based on the
lowest RO scored
in any
parameter. RO
con be adjusted
up one category
based on
degrodobiSly
characteristics.
1
Some RQs hove osen
assigned on Ihe
basis of chemica'
structure
similarity.
D-20
-------
hTable 3-2 Parameters and Types of Evidence Used to Evaluate 1
Chemicals in Existing Screening Programs.
Exposure
Acute Lethal Toxicity
Aquatic Animal
Terrestrial Animal
Other Acute
Terrestrial Animal
Chronic /Sub- chronic
ToxicqLty
Aquatic Animal
Terrestrial Animal
Aquatic Plant
Terrestrial Plant
Reproductive Toxicity
Developmental
Toxicity
Genetic Toxicity
Carcinogenicity
Chemical /Physical
Characteristics
Environmental
Transport
Bio -accumulation
Persistence
Flammability
Reactivity
Corrosivity
Tainting/Aesthetics
Production
Use
IJC
M,Q
P
P
P
P
P
W
P,W
W
W
M
MOE
M,Q
P
P
P,W
P
P
P
P
P,S
W
W
M,E
M,E
M
CMR
P
P
P,Q
P
P/Q/S
P
Q
P,W
P,W
W,S
W
M,E
M
M
M
M
P
BUA
M,Q
P
P
Q,W
W
M,E
M
M
Q
I?
P
P
P
P
p
W
W
M,E
M
M,E
TSCA
-
M,Q
P
P
P/S
P
P,S
P,S
W
w,s
W
M,E
M,E
M
.
JCERC
=T==i=^=
P
P
J
P/S
P,S
P,S
W.P
D-21
-------
Legend
M*-
£-
CAP-
e
Ministry of the Environment Scoring System.
Michigan Critical Materials Register. _
Existing Chemicals of Environmental Relevance,
Lists I & II.
Clean Air Program.
Xcance Act Scoring System .
- Comprehensive Environmental Response and Chemical
Liability Act.
M- Measurement
Q- Qualitative
p- potency
W- WOE
S- Severity of Effect
E- Estimate
D-22
-------
1
APPENDIX E
OVERHEADS FROM THE PRODUCT AND
PROCESS DESIGN PRESENTATION
-------
-------
1
!
8
8
o>
I
I
o
I
ft
s
L
CO
"D
CD
CD
CD
CO
g
"to
to
CD
CO
OJ
Q_
O)
'to
-------
E-2
-------
1
o
a:
oc
£
I
i
IS
g
o>
CO
0>
o-
E
0)
to
V)
05
C
O
"co
CO
CO
tr
CO I
3
CD
CO
13
CD
-C
CO
O
CO
T3
v_ C
CD *tf
CD
t'S
LJJ E
CD
E
CD
O)
''CO
OS
Q
I
E-3
-------
o
5
1
1
e
3
*
&
CD
CD
LL1
C
O)
O)
"o
o
CD~
JO
.25
o5
IS
E
CD
05
E
1
CD
o
Q. C^-
2 E
to
8
— O
O)
O
s
03
13
a
CD
CD
O o»
c S
g) CD
co ^;
CD ^
o»
-------
1
1
CO
% 1
$ i
^ I
o ! •
1 I c
5 1 °
1 I "co
i I <*>
8 I CD
S!
i
o
CO
CD
CD
X)
ol
IT
CO
-------
PC
8 1
c
2
I
o
1 1
I
s
e
I
i
o
o
C/)
T5
CD
CD
^••M
+-»
VilM
0
E
CD
§>
C
Cu
S
0-
"co
o
o
o
CD
"o*
V—
a.
"S
vu
HpV
^^^
CD
O
-3
n— *
0
Q.
CD
JQ
•o
O
^^
0-
.9?
•=
la
55
Q
^j
•^^j
i3
^^^
>•
*.-• i—
"m ^
CO "(3
•a ==
O CD
*— rn
o»
E-6
-------
f
°
o
S
I
i
4NUFACTUK
SI
5
i
3
0
K
r
®
s
1
i
o-
Q
O
o
V)
0
CO
Q
E
mm*m
ID
O
<
a
CO
CO
"•"•
o
§
i>
00
CD
C
D)
"co
CD
Q
•*- •
O
CO
05
CD
Z
n:
"CD
£Z
0
0
'ZL
e
0
05
a.
.i
"c6
"c
CD
o
c
CD
CD
Q
eg
O-
CD
"w J2
05 05
r—
& 0
O 00
-C 00
> (D
^ H—
o 2
co a.
w
C
05
O
**- 00
JT^» ^^P"W
O -Q
If
O Z3
"• O5
05 ^
5 -a
|£
-L_ -*—*
^- CD
« E
E g>
O)*=
2 to
,+_f
o o
Or-
"j~:
O ^
o >
00
t3 p?
13 c
o 2
Q.^?
< 2
w
T
C'
(
tarn
9
(D
•#-*
sz
(D
cz
co
o
IB
cz
JD
X
V.
0
"05
CI
0
|l
Q
CO
^* 0
CD
^iV
E-7
-------
CO
O CO
E-8
-------
1
f
§
§
i
tt
I
I
I
CD
15
CD
CD
C
C/D
CD
O
13
O
CO
0)
DC
o
•o
o
Q.
cd
cd £?
s-i
CD
E
CO
S «
O1 CD*
d. O
If
1!
"g
W .E
O F
E
CD
CD
n
+-«
D)
C
4—»
E
cd
td
~o ^j
w ^
co o
J? E
•n w
TR ^
m ®
2 tf>
C (0
= cd
"5S
ia-S.
O p
CD .E
g CD
cd
•-S CD
o
a 8
•^ a>
S'.»
c x
o CD
O co
O c
CD fc
-^ O
CO
c o co
£.<2 co
Vj *— 4—J
O :£ O
•£® a
g*.i E
vj •§—* •—
o
CD
*—* •_
CO CD
CD
<2 . O
E tS «°
E Z> CD
Q
cd
s .p-
CD .
i- i>
. CO
CO 0>
CD .2 CD
CD
ci *--
" O '>
1 *"- f") ^
CD ° —.
C~ T3 ^
TO C .£>
"55 ^ .5
a> co o
Q o o
co a 8
0
a
CL.i >,
8 £ o
CO ^ Q
CO -n ^
*=J (D S2
D)^: c
C CD E
"CO Q
•il.p
m o
g
O)
o
CO
E-9
-------
i
IS
CO
0
"cd
O)
CD
O
.
J
^
§
^
_x>
CO
£
o
cc
"•o
13
O
i
^
O
Q.
£
o
(/)
!2
"o
CD
Q.
CO
a.
E
1§
C c
E-B
CZ3
^^^^
O rS
CD
Q.
O
0
CO
CD
£
o>
t-
m*
0)
3
E-10
-------
APPENDIX F
OVERHEADS FROM THE RISK RANKING
AS A REGULATORY TOOL PRESENTATIONS:
(1) OVERVIEW OF REGULATORY ISSUES
(2) STATE REGULATORY USE
-------
-------
D)
C
CO
LU
D
CO
CO
cc
o
0>
3
LJL
0)
C
03
DC
J*
(0
£
Is
c:
(D
.0) CO =>
T: o. o
CD C CO
E §.2
50S
»\
Q.
O
.C
CO
o
Q>
E
c
2
LU
CO
08
c
o
O)
I
LJL
CO
(1)
Q.
UJ
F-l
-------
3
o
3
0)
I
O
o
•o
03
•3
i
•5
o*
TO
m.
(D
33
0
to
o
c
o
0
(D
g
O
CD
_
3
75
(D
CO
GO
O
O
o
CD
13
(Q
to
(D
O
O
-------
o>
_c
c
o
-^
CO
o:
g
"c
CD
E
c
o
•o
0)
0)
2:
^
35.
y Experience
•jj
"o
CO
uT
0)
H
#
w
ogist/Ecolog
o
ffi
«:*
,.
^
l
s
1
! . ;
CO
to
I
«r
•
c
5
e
o
"«5
"5
-------
I
(0
§
O
o
I
I
33
o'
ar
03
s.
o
F-4
-------
iVti
'////?,'
;{',s;~~',j
;-, / r''f
t>>'" , *i t
$'*i&
r^/'l
CO
o
CO
0
O
0
0
&
V)
O)
c
CO
CC
CO
CO
0
T5
T5
<
CO
C
O
"co
CO
0
O>
c
CO
^^^^
3
O
2
.2
CD
o
C/>
0)
§2
0)
Q
CD
O
O
CD
CD
Cfl
TJ
CD
CD
O C-
5«
c»
Q)
.2
^
•o
CD
CD
CD
"5
CO
0)
CO
o
>
OB
I
o
CO
o.
o
o
o
F-5
-------
o
(/>
D)
O
o
I
3
o"
CD
— t
CD
D CD
0
F-6
-------
0
D
Choice Of How We Produce
c:
o
C) O
x
8
"w o
c: o
S>LJ
tii3
±: c
a> <
75 «-
«J ^3
2: c
«»
Vil "i"
k. J.
V\f MWB
fli k-
st. CB
.. E
££
^
i«
|i
l-s
li. O
— o
§°
tl
o-l
O _co
4> O
JC DC
Hs
Kimerle
chard
(8
O.
£
o
o
(8
(0
O
F-7
-------
o
3
(A
D)
I
O
o
T3
D)
«<
i
3J
o'
Q)
=•
-------
CO
0)
CO
o
Q.
w
'S
UJ
to
a>
o>
'SI
o
o
(5
u>
CO
DC
CD
o>p> |
C .£ I
^T3
C 0
D 0
Di O
-v X
-co UJ
•iV
z
,>
MM
1
O
o
CD
CO
LL
"co
O
gle Tec
o
LL
x
UJ
o
CD
Peer Review
3
Q.
CD
CO
o
Must Be Scien
CL
UJ
(/>
CD
"o
CD
D)
- °.
d
oT j=
CD
m
0)
Single Framewor
^
5J
3 £
a.
UJ
CO
CD
"O
CO
-J
"2
CD
O
CD
0
jOJ
oii
5
P
(0
x:
o
£
i
>%
c
CO
Q.
O
o
o
**
(0
c
o
F-9
-------
3
o
I
I
O
o
I
«<
30
o'
M
a
Kimerle
Product Labelling
m
0
Q.
0
o
0
to
CD
C/>
C/>
CD
3
!-*•
CO
O
CD
F-10
-------
CO
£
CO
0
.1:
ods
CO
o
DQ
5* o>
CD O S
o S ^
S =» .0
^ Q. CO
g ® 15
a s -a
CO > a
CD O
CQ U> "m
co £
CD "S"
"43 C
1-3
Q- <
J> CO
.Q CD
CO O
D
O
CO
CD
n
CD
ill
±: 5 a
o -==: st
£ ^: o
^ CD Q
C? S O
c c I-
O O 0)
•9" •""' mm
S co E
CD 'O iZ
"S CD '
•§ Q 8
C CD "CO
IJJJ S H
5 c E
> ui :>
"^t* °^t ^r
T^ ^F '^
_£
I
chard
>%
CO
a
o
O
co
CO
o
F-ll
-------
I
w
05
I
O
o
T3
Q)
«<
i
33
o*
0)
s.
-t
(D
* #
30 m
(D 3
§ 5
C ^
5 «<
O
0)
D
(D
T3_
S
O
C
Z *
is
W jj)
I 1
w
m
X
0
(A
i
.
TJ
referred:
CO
7T
7D
Q
D
7T
D"
(Q
CO
CO
CD"
3
•o
^**<»
Q_
CO
O
.c
a
— i
(D
O
^^
CD
Q
DO
0
(
-------
CO
0
CO
CO
CO
c
0
O
C
O
O
s
o
"' f ;
/' ''''*
'/%
^:<3
/'/,^
/' ' ^
^ ?^
-------
o
CO
M
I
O
o
•o
B>
«<
i
30
o'
3-
J»
s.
Kimerle
=*
mi (/} *^^
(Q C C!
W 5) fi)
> 3 0
5 PT (Q
»s s
s<« g-
33 0) S
s H '
^^^ . ^"1*
•o ~
OfTt V
0 «.
•^ OCQ
"° St. Zi
=- 0) 3
^ 3"
-------
s
oncerns/lssue
O
CO
C
o
CO
0)
D
Q
„
o
o
_p_
o
vx
Q.
o-
CD
CO
CO
_Q
•••
CO
•*-*
CO
Q
J=
Minimui
^
CD
CD
CD
0
O
•
CO
T"
C-
Q.
CO
0
2
"5
Q
JO)
73
CO
X
i
O
Q
0
- _•
^j*
*"
a.
O
u.
<1> M
.N «
go
CD CD
Q- >
*- 5
O 3=
O) v
i §
| s
0 2Z
? c
*- CD
£°-
"S ^
i ^
0> (0
<= «
0 Q
s =
« >
i£
D
^rt '••
^
O
CO
"c
o
vt
^i
.2
CO
to
11.
0
o>
c
J)
"co
0)
2
0
O
• -
^•O
1r"'
CD
CO
O
CO
-Q
GL
•K .
3
Q.
—
CO
to
o
_CD
.£5
CO
jl
CD
5
Chemicals. Unify
*»—
O
CO
"F
CD
r
CD
Q.
O, .
i
£
+-*
c J=
CD £
£: o>
CD £
.c t—
Describing In
How We Use'
t Should Remain On
CD
O)
-o
: 3
"D
0
Scientifi
T5
Flexibility An<
.
js«
^"
Set Out Principles
c
o
1
LU
BCO
O
CD
Ekposure Sid
Of Exposure
To Next Generation
CO
CO
eo
CD
O
•D
O
fr
CQ
CD
O
to
i
i
•
00
Stressors
m
. Kimerle
•<
o Company - Richard
c
CO
CO
c
0
F-15
-------
s
o
3
(/>
0)
I
O
O
•o
0)
33
o'
n>
a
a.
0
30
to
o
F-16
-------
D
CO
Q.
o
1
o
o
0)
09
O
c:
"55
di
Regulato
TI
_
o
(0
o
(0
Q.
O
o
re
tn
o
F-17
-------
o
to
n>
O
o
I
«<
33
I
05
CD
m
<
I
(D
*-»•
CD
(D
(D
©
«
F-18
-------
STATE REGULATORY USE
Evaluating Cleaner
Technologies: User Needs in
a Toxics Use Reduction
Program
Mark Rossi
Massachusetts Toxics Use
Reduction Institute
F-19
-------
State Uses for a Relative Risk
Ranking Systems
To identify resource
allocation priorities —
research, technical
assistance, and education
To identify priority user
segments
To identify candidates for
chemical restrictions
F-20
-------
Resource Allocation
Research, education and
technical assistance priorities
need to be based on a better
understanding of the
comparative risks of using toxic
chemicals.
F-21
-------
Resource prortes are
currently based on:
• legal requirements
• amount of chemical
releases and transfers
"biggest bang for buck"
regional needs
alternatives to ozone
depleting chemicals
F-22
-------
1
User Segment
A set of no fewer than five
toxics users who employ a
similar production unit.
F-23
-------
Considerations for identifying
priority user segments include:
amounts of toxic
substances used by the
user segment in the
production units of
concern and their toxicity
amounts of toxic
substances disposed of,
discharged, or released to
water, land, air or
workplaces within facilities
the social, health, and
economic benefits and
costs of targeting the user
segment
F-24
-------
1
Chemical Restriction
Any policy that encourages or
directs a company to eliminate
production, uses, or distribution
of a chemical (or class of
chemicals) or product which
contains the chemical.
F-25
-------
Information Users
Massachusetts Department of
Environmental Protection
(DEP)
Massachusetts Office of
Technical Assistance (OTA)
Massachusetts Toxics Use
Reduction Institute (TURI)
Massachusetts Advisory
Board on Toxics Use
Reduction
Toxics Use Reduction
Administrative Council
F-26
-------
Priority User Segments
Administrative Council is
responsible for identifying
priority user segments
based on recommendations
from DEP, OTA, and
TURI.
Administrative Council
consists of "Secretaries"
and "Commissioners" from
state agencies such as
DEP, DPH, and Economic
Affairs.
F-27
-------
Advisory Board on Toxics Use
Reduction
15 individuals from
business, government,
environmental groups,
labor, health policy
groups, POTWs, and the
general public.
Provide a forum for
discussing direction of the
TURA program.
Develops recommendations
on the TUR program.
F-28
-------
Toxics Use Reduction Institute
Industry Advisory Board
Public Interest Advisory
Board
F-29
-------
Information Needs ~
Resource Allocation
How can the difference in
risks between use and
emissions be accounted
for?
F-30
-------
Massachusetts Toxic Chemical Use
Data (1990)
Total Use
Billion
Pounds
1.229
Manufactured 0.126 10.25%
Processed
Otherwise
Used
0.963 7836%
0.140 11.39%
F-31
-------
Information Needs —
User Segments
How can the difference in risks
between production processes
be accounted for?
F-32
-------
Information Needs —
Chemical Restrictions
How does changing the
threshold levels effect the
screening process? Need for
"scenario analysis".
How can a screening tool be
tailored to regional needs?
F-33
-------
-------
APPENDIX G
OVERHEADS FROM THE EPS SCORING SYSTEM PRESENTATIONS
-------
-------
The EPS System
Environmental Priority Strategies
in Product Design
Environmental Load Index x Quantity = Environmental Load Value
Product Ecology Project
Federation of Swedish Industries
G-l
-------
£
I
°
O CO
•- 3
— J?
> 0)
1
•62
•S? £ to
O -i S
"3 c ^
O o =
1
•S3
(C C „
1 2 8
^ •> 3
• «
c c
CO 0)
a> u
si tn
CO CD
Q.
f
"w
*o
o
a.
to
J>
CO
E
CD ^
1-1
E «
O
111
G-2
-------
pplementary regional
3
CO
g
Q.
CO
"co
.Q
O
D)
CO
TJ
CD
CO
CO
.Q
^
ocal adaptation)
^~
"O
c
CO
CO
o
ible to simply repeat calculati
CO
CO
o
Q.
oi
CO
"co
ling sensitivity and error anal
.a
CO
£=
CD
CO
judgements
TJ
orting validated and motivate
Q.
Q.
3
CO
*
n of new scientific findings
t environmental effects
3
O
CO
o
"to
3
O
c
CO
CO
CD
i
CO
C33
Q?
05
.O
"x
0)
10
-^^
CO
2
H
CD
C
TJ
C
co
possible an overview
O)
parent and consistent, makin
CO
c
CO
CO
separate calculation steps
-^
CO
0
G-3
-------
CD
Q.
"o
C.
a.
c
Valuatio
^— •
o
CD
•Cj
3
CO
"g
CO
3
O)
.CD
CO
^^
f!*
"s
!CD
>
'o
.Q
0)
•~
o
CD
2
Q.
,0
to
0
o
"CD
"o
o
CO
"55
CD
0
CD
CO
JE
•••St
•XD *w
^CJ ,Q CO
jg P. §)
XD '§ ^
*^x ^^L CO
i •— -
O T3 *O
m^ C
^» CD •«
•§ - °
^ CO 5** CO
C^ 5 ^* •""
CO CO Q "^3
^^ ^ ^^ f^
^W ^ ^ Q^
II ||
S CO '^ 3
CO o 5 co
^_
^^2
"CO
CD
co
z:
CO
CD
Q.
"CD
E
O
O
LU
O
c
.g
o
•a
2
D_
CO
•5
.•2,
3
•g
CO CD
3 0
D) ^
.2 0
CO CO
co o
II
0 0)
CD C
II
0 CO
^ 2
CO
CD
o
o
CO
CD
cc
9-
0
xb
CO
CD
0)
»•—
1
'CD
Q.
o
0
Q.
CO
CD
3
1-
0)
to
CD
G-4
-------
Goal
Definition
and
Scoping
Holistic environmental
impact assessment
Environmental
impact categories
Environ-
mental load
.CO
.>»
CO
CO
CD
•o
CO
CO
CD
CO
Inventory
Analysis
Classifi-
cation/
Charac-
terizatoin
Valuation
CD
CO
CO
0)
CO
C
CD
Q.
G-5
-------
Principal objective
for eco-design of products
Consequence
"to lower the overall
environmental impact
during a product's
entire life-cycle"
"to conduct an aggregated,
holistic environmental impact
assessment—from cradle
to grave"
-------
Calculation of eHVIFtinftteWtaf
i
f
f
i
: W
jBSOdSjp 9}SB/\A
^i^sn i^^ni^^^ i i
oaM 4*Jii|JUjQ
§
I
f
Materials'
uc
Processes
U4i/iL
Fastening
poj
Surface
d
treatment
O)
"5.
to
sz
CO
:« 0)
.JbtO
f i
g e
Si*
.. CO
i CD
l-i|»
|l§l
O E
(0
CO
O>
,.E
= W
t5 Q>
ffi CO
*r (o
Z3 O
C O
CO O
12 D.
C O)
CD O
£ .= CD ® 2
O = -cB 1 -I
*• "" ™ to •=
'> "% B £ •= ,i= "O
c 5 co co E < co
UJ O z DC UJ , ,
CO
CD
.0)
E ^
= 2 «
2"^— Ctl *^
m fl o
««« ^. •£:•*- s
iff -O tB 40 ^ c o
W -^ 5%.,0 .^2 = 0^
o CO CO 3
E CD CD IB
:"«:"!6 ^"5 ^ CO I Q- DC < >
> 0> CO CO i iii i
S.2
G-7
-------
'e
CD
>
b
o
.Q
5
t
o
•o
o
\
\
\
o
O
2
CO
§
•2
CO
CO
w
t
\
\
1 Natural resources
Ozone depletion
Global warming]
Acidification
Photo-oxidanttormation
Eutrophication
Exotoxicity
Human toxicityl
o
03
CD
CO
•o
2
\
\
\
\
\
\
\
\
\
\
~" /
O
i
I
i
i
i
i
i
i
I
i
t
\
CO
c c
'
I
G-8
O
CO
CD
I
CO
-------
EPS Valuation Principle
1. Define a "unit effect" on the five safeguard subjects:
• Human health
•' Biodiversity
• Production
• Resources
• Aesthetic values
2. Valuate the unit effects.
"willingness-to-pay"
3. Set a value to a change in the unit effects.
4.
Estimate the contribution a certain resource depletion, pollutant emission, or human
activity will give to this value of change in the unit effect (i.e., a change in the state
of the environment).
G-9
-------
The following valuations are used:
Impact type
Value in ELU (s.d.factor) Note
General and global impact on
biological diversity
5-1011 (5)
Decreased production of:
1kg of crop seed 0.2(2)
1 kg of meat or fish of economic value 1.0 (3)
1kg of wood 0.025(2)
1 kg of freshwater in areas with water 0.003 (4)
deficiency
Excess death, normalized
Painful morbidity and/or severe
suffering, per manyear
Other morbidity, per manyear
Severe nuisance, per manyear
Moderate nuisance, per
106(5)
10s (10)
104(10)
103 (10)
102 (10)
100 ELU/person
times 5-109
persons
equal to 50%
reduction in mean
life expectancy at
birth
G-10
-------
5
LU
2 ..... !£*— §
G-ll
-------
Input data to which the obtained priority has the greatest relative sensitivity
s,
0.2 T
0.15 -f-
I
!
0.1 t
0.05 t
£§ v I
1° f
| *«T
0
-0.1 -
10
PtobtWnty In % ttwt ttm diffwwre* In •nvlronnnmtrt load v«h» te taw thwi th» v«tu» gtvwi
'
G-12
-------
o
CO
o>
CO
c
CD
Q.
CO
O)
(0
00
\
13
_i-
ill
O
-o
-o
t
t
LU
LL
LU
b
O
CO
CO
o-
o
t
- L
co *•
o——
O
G-13
CD
O)
CD
O)
I
CD
DC
-o
CO
CD
u_ ^3
LLI JD
O |
IDQ =
OT
< O
O
-------
EPS Applications
• Strategic planning
• Product design
• Process waste minimization
• Waste recycling optimization
• Environmental auditing
G-14
-------
Example with Undercoating
Conventional Material
Use of material: 14.4 kg
Solvents to aJr: 2.2 kg
Waste: 0.4 kg
Hot-melt
Use of material: 6.0 kg hct-mett
+ 1.2 kgconv. mtrl.
Solvents to air: 0.26kg
Waste: 0.2kg
VOLVO
G-15
-------
Example with the Bonnet (Hood)
JUumtaun
Use of Material: 15.0 kg (5.0 kg loss)
Final weight: 10.0kg
Paint surface: 2.0 m2
Steel
Use of material: 26.0 kg (8.0 kg loss)
Final weight: 18.0kg
Paint surface: 2.0 m2
VOLVO
G-16
-------
APPENDIX H
OVERHEADS FROM THE MICHIGAN CRITICAL MATERIALS
REGISTER PRESENTATION
-------
-------
UJ
LJLJ
X
CO
UJ
CO
CO
UJ
CO
CO
<
a
cc
UJ
cc
o
c_>
CO
<
o
on
UJ
CQ
CO
<
O
< UJ
f-o:
o o
i-o
03 CO
CO
CO
UJ
o
DC
Uj
U_
Ul
o
u.
g
UJ
o
o
CC
UJ
CO
UJ
cc
CO
DC
UJ
cc
o
Q
UJ
<
< a:
o h-
*-• co
•Z. CQ
UJ <
CJ _J
o — •
O 3C
CJ 5J
u
M
X
p
V A A A
^s^
V A* ~ A*
*?» t i 25
• 8 T"
in m r« o in
V A A A
m J» K o «
V A A A
S^^S*
! >t 3 sea u.
s
s
£
>
iȣ s
1 5 a M.
S.TS I
is* si
S v s Ms
S W 9 ^
u
X
a
s-*£
»-s=f-
s s ts a
,il
£ £
>- •- la c
! S^aS
a *9* ^s
*v=5*I
i
II
asss5Z
ass s so
5
1
gs§iil
i
*3 si
» r* o •*
H-l
is_ a,
jd!
I;
la""-'
3 8
ii
§-j
9
- s
li
fe s: «
Ut uS
w S u
-------
SBS5
!P! 5
I
» tn — Cf c*J
s s 1 - s
al s ss
5 S g •*>
*~B §
2«
SE
: R; r1
!^ i
: Jg 2
VEL OF
m m
SS 2
uva
n '
? i;
* i
v s
y V A |A V A A
V V
is
53 v*
i
i
sss
= £
H-2
B
I
OTHER TOXIC
TERRESTRIAL
O
o
m
o
25
m
Til
mi
mi
ci
m
a»
c
c
M c
o-
oc
yo-
ma
r
-------
APPENDIX I
CRITICAL ISSUES IN THE DEVELOPMENT OF
HUMAN HEALTH AND ENVIRONMENTAL
RISK RANKING AND SCORING SYSTEMS
-------
-------
CRITICAL ISSUES IN THE DEVELOPMENT OF
HUMAN HEALTH AND ENVIRONMENTAL
RISK RANKING AND SCORING SYSTEMS
Prepared for the Workshop on Identifying the
Framework for the Future of Human Health and
Environmental Risk Ranking
Washington, D.C., June 30-July 1, 195)3
Gary A. Davis, Director
Sheila Jones, Research Assistant
Center for Clean Products and Clean Technologies
University of Tennessee
Knoxville, Tennessee
INTRODUCTION
The development of overall human health and environmental
risk ranking systems is in its infancy, although elements of such
ranking systems are being used in several fields today. Chemical
ranking has received the most attention, and there are several
systems in use for purposes such as setting regulatory priorities
and ranking contaminated sites for cleanup. Yet, there is still
no scientific consensus on this one area of overall ranking.
This paper discusses the important elements of human health
and environmental risk ranking systems, with a focus on chemical
ranking and scoring systems. It highlights important issues that
must be addressed in the development of any such system. These
issues include:
the purpose of the ranking and scoring system
the human health and environmental impacts included
whether potency and severity of impacts is taken into
account
whether measures of exposure are included
the use of aggregation and weighting of different
impacts
• how missing data is handled
There are two basic types of ranking and scoring systems
that are being used: chemical risk systems and life cycle impact
assessment systems. We have identified more than 40 chemical
ranking and scoring systems and discuss some of the elements of
these systems in this paper. Overall ranking and scoring
methodologies have been taken up by the field of life cycle
assessment. We have identified some of the methodologies being
developed for life cycle impact assessment and discuss elements
of these in this paper.
1-1
-------
PURPOSE OF THE RANKING AND SCORING SYSTEM
The purpose of the ranking and scoring system is the
principal issue in the development of a system. The intended use
of the system affects all of the aspects of the system, including
the human health and environmental impacts considered, the use of
aggregation and weighting, and how missing data is handled.'
Chemical Ranking and Scoring Systems
Most of the ranking and scoring systems that we have :
identified are for chemical risk. Chemical ranking and scoring
systems are typically used as screening tools, as a rapid
assessment of relative chemical hazards. They consider the toxic
effects of chemicals and some measure of exposure, but are not
intended to serve as a detailed risk assessment. They usually
function by assigning a score to each of several impact
parameters (human health and environmental endpoints) in
conjunction with potential exposure parameters. '
Chemical ranking and scoring systems have been.developed for
three broad purposes:
• Regulatory action
• Priority setting
• Impact evaluations
Regulatory Action
The need for the regulation of chemicals of concern has been
the driving force behind the development of many chemical ranking
and scoring systems. / •
[15]'
Examples include:
CERCLA Section 102 Reportable Quantity Ranking Process
Releases of certain chemicals at threshold quantities must
be reported to the Environmental Protection Agency. These
reportable quantities (RQ's) were established through a
scoring process taking into account the environmental and
human health hazards of the chemicals on the CERCLA list.
Michigan Critical Materials Register [103: The ranking
process results in a list of chemicals that may threaten
water quality in Michigan. Chemicals included in the
register are considered to pose a high degree of
environmental concern, and companies must report their use
and discharge of these chemicals.
* The superscript numbers refer to the numbered references at the end of this
paper. A more complete bibliography of chemical ranking and scoring system
references is also attached.
1-2
-------
. .Ontario Ministry of the Environment (MOE) Candidate
Substance List for Bans or Phase-outs I5]: This ranking
system identifies;chemical substances for release reduction,
bans, or phase outs.
George Washington University C7]: The scoring system
developed by GWU is intended to serve as a tool for
pollution prevention in the Great Lakes region through the
identification of chemical substances for Sunsetting (bans
or phase-outs) and other regulatory or non-regulatory
activities.'
.CERCLA Hazard. Ranking System tie]: EPA developed the HRS to
meet the requirements of CERCLA for listing priority
hazardous substance sites to be included on the National
Priority List for cleanup. The HRS is more of site-specific
risk assessment than a chemical ranking and scoring system.
Priority Setting
Several ranking and scoring systems have been developed
specifically for assessing chemical hazards for priority setting
for non-regulatory purposes. Examples include:
EPA Design for the Environment Program Use Cluster Scoring
System [17]: This system is aimed at measuring potential
.. .. health and environmental risks of chemicals that are used in
industry clusters (i.e., certain common industrial,
processes) as a way of evaluating the benefits of
substitutes for those chemicals. The relative risk scores
are provided to industry as a means of encouraging the
voluntary adoption of pollution prevention measures.
WMS Scoring System [9-14] : This system was developed by the
Directorate General for Environmental Protection of the
Netherlands Ministry of Housing, Physical Planning, and the
Environment. It was intended for use in industry,
government, and academia for the selection of a limited
number of chemicals as priorities for further investigation.
.Agency for Toxic Substances and Disease Registry (ATSDR) [2) :
The Superfund Amendments and Reauthorization Act (SARA) of
1986 required the ATSDR to prepare a list, in order of
priority, of the hazardous substances commonly found at
National Priority List sites that pose the most significant
potential human health threats. Substances on this priority
list become candidates for the preparation of toxicological
profile reports prepared by ATSDR.
University of Tennessee Chemical Ranking System U3]: The UT
Center for Clean Products and Clean Technologies has
developed a ranking system to prioritize chemicals for
1-3
-------
substitutes assessments. The system has also been used to
rank the overall hazards of state Toxic Release Inventory
releases.
linpact Evaluation
At least one scoring system is being developed to evaluate
the potential impacts of chemical releases:
Toxics Release Inventory Environmental Indicators
Methodology I1]: This system under development by Abt
Associates for the EPA Office of Pollution Prevention and
Toxics will be used to evaluate TRI releases and derive a
value to indicate the overall impacts.of those releases by
all facilities and to each environmental medium. Annual
calculations of the indicator numbers allow a comparison of
potential TRI impacts from year to year.
Life Cycle Impact Assessment
Life Cycle Assessment is a holistic approach to evaluating
the human health and environmental burdens associated with a
product or process life cycle. A full LCA includes a quantitative
inventory of resource and energy inputs and pollutant outputs and
some form of impact assessment. Life cycle impact assessment
attempts to assess the full range of environmental and health
impacts of product systems.
Life cycle impact assessment methodology involves three
steps:
• classification: the process of assignment and initial
aggregation of life cycle inventory data into
categories of impacts based upon impact chains
• characterization: the assessment of the potential
magnitude of the impadts within each impact category
• valuation: the process of assigning relative values
and/or weights to different impacts
LCA has several uses, although its uses are more limited if
some form of impact assessment is not included. LCA can be used
for internal product improvements, for designing new products,
for setting public policy on products and materials, and for
environmental labeling. Clearly, the different uses of LCA create
different needs for impact assessment. An LCA used for internal
product improvement, for example, might simply use the inventory
component and operate on a "less-is-best" approach. An LCA used
for setting public policy on materials or products would need to
include some framework for assessing and comparing the
significance of environmental releases and resource and energy,
use of the different products and materials being compared.
1-4
-------
Following are examples of some of the life cycle impact
assessment methodologies in development .or use:
EPS-Enviro-Accounting Method [123: The EPS method is being
developed by the Swedish Environmental Research Institute
and the Swedish Federation of Industries as a tool to assess
the health and ecological effects associated with the entire
life-cycle of a product, process, or activity. The main
objective of the EPS Method is to provide one overall
economic measure of resource depletion and potential health
and environmental impacts throughout the life cycle.
Critical Volumes Approach: This approach has been embodied
in some of the LCA software packages developed in Western
Europe, such as SIMAPRO, developed by the Centre of
Environmental Science at Leiden University in the
Netherlands, and Okobilanz, developed by the Swiss Federal
Environmental Agency. In this approach, specific pollutant
amounts from the inventory are aggregated by dividing by a
regulatory standard for that pollutant (expressed in mass of
pollutant per volume of air or water) and summing the
resulting volume figures and expressing the result as total
volume of air or water polluted at regulatory standards.
HUMAN HEALTH AND ENVIRONMENTAL IMPACTS CONSIDERED
Chemical Risk
In order to assess the potential or actual hazard associated
with a particular chemical substance, scoring and ranking schemes
include criteria for scoring the toxicity of a chemical to
terrestrial mammals, non-mammalian terrestrial species, non-
mammalian aquatic organisms, and/or plants. The toxicity of
chemicals to these organisms is used to evaluate potential
effects to human health and the environment. Several scoring
systems do not determine an overall score for human health
effects, but the data obtained for acute effects on mammalian
species, as well as chronic effects such as carcinogenicity and
mutagenicity, are used to assess human health effects. For the
purpose of this paper, acute and chronic mammalian toxicity data
are considered as surrogates for human health effects, as they
are in most ranking and scoring systems. Ecological effects
typically include acute toxicity to terrestrial mammals, non-
mammalian terrestrial species, aquatic organisms and plants.
Some scoring/ranking systems include bioaccumulation and/or
persistence in the environmental effects category, but these are
discussed as factors affecting exposure within the framework of
this paper. -.._. ;
1-5
-------
Human Health Effects
system are varied. Examples include:
°
wall
t
genera? toxicity, genotoxicity, carcinogeniczty and
reproduction damage/teratogenicity .
EPA Design for the Environment Program Use Cluster Scoring
.
potential score.
^•^ The hiaher of the two chronic toxicity scores is the
S 5S S?SS
reference dose nor the slope factor
Ministry of the Environment (MOB) Candidate
health effects.
scores .
1-6
-------
o
1
0)
to
*-*
c
'§_
u.
-D
1 1 1
-C
S
I
C
1
• I
"o
• JS
Q.
ra
T-
cu
_Q
f—
u
• 'E
2
^
u
a;
3
o
i_
x:
6
o
*- v
Si
z
~D ni O 15 .ti
2 > "a; c .y
&« fc 8 g
^ Q C (_
4^
i> S .-^
nj CJ w
3 >.'X
•^ £T O
•^ *u
0 >•
c ~
— u
U So
w
^r
O
(9
E
QJ
Q
c
.2
nj
"5
"2
0
o
jC
4M«
3
•>
">
^^
^
*v
•^
•«^
mu
"2
0
Z
*>
^^
•^
•-^
•6.^
^j
6
>
^
•-,.,,
t'
to
I
2
I
2!
§
o
a;
O)
1-7
-------
Environmental Impacts
Environmental impacts are those that occur in terrestrial
mammalian and non-mammalian species, aquatic organisms and plants
as the result of chemical exposure. Most ranking systems
evaluate effects through the use of toxicity test data.
"Schmallenberg" System t181: This system was developed by a
group of French and German officials and scientists and
includes many endpoints to assess ecological effects. These
are acute, subacute and chronic effects in species ranging
from algae and plants to earthworms, birds and fish.
George Washington University System [7]: The GWU system
includes acute and chronic toxicity to aquatic organisms as
well as terrestrial and aviari, non-mammalian species.
EPA Toxics Release Inventory Environmental Indicators
Methodology [1): This draft methodology determines a score
for ecological effects based'on a matrix which combines_
aquatic toxicity and bioaccumulation. Aquatic toxicity is
scored according to data on the LCSO or life cycle/chronic
NOAEL or other measures such as the Acute or Chronic Ambient
Water Quality Criteria (AWQC). Bioconcentration is based on
the water solubility, Log KOB/ or bioconcentration factor
(BCF) .
EPA Design for the Environment Use Cluster Scoring System
1171: scores ecological hazard on the basis of Aquatic Water
Qua'lity Criteria (acute and chronic) or Aquatic Toxicity
Reportable Quantities.
Michigan Critical Materials Register t101: scores acute and
chronic effects on aquatic organisms (fish, .invertebrates,
amphibians), acute and chronic effects on terrestrial
animals, and plants.
Table 3 summarizes endpoint selection for evaluating
environmental impacts in four scoring systems.
1-8
-------
1-9
-------
Life Cycle Impact Assessment
Life cycle impact assessment goes beyond direct chemical
toxicological impacts to include several other environmental
impacts as shown in Table 3. These impacts are often expressed in
terms of the quantity of certain environmental stressors that can
produce the impact of concern through an impact chain, such as
acidification chemicals, biological oxygen demand, and smog
producing chemicals. Energy use is often used as a surrogate for
the environmental impacts of resource depletion and environmental
impacts resulting from energy production.
J.CtiJJ.C J .
-=»——• —
Natural
Environment
~^^=====^=
Atmosphere
ozone depletion
crreenhouse effect
smog
Water
eutrophication
oxygen depletion
turbidity
Soil
erosion
salinity
Other
species extinction
habitat loss
=======^=^=^=:==^===
Human
Health
Acute Effects
accidents
noise
odor
1 s===
=r^=====^^=s=:=s=
Natural
Resources
==== =s=s=
Stock
fossil fuels
minerals
Flow
water
soil fertility
trees
land (space)
Following is an example of impacts included in a life cycle
impact assessment system:
Swedish EPS System [121: five "safeguard subjects" have been
identified in order to provide a base on which to describe
and value impact types. These include biodiversity,
production, human health, resources and aesthetics. "Unit
effects" is the term used to describe effects on these five
subjects. By assigning an "environmental load index"
(expressed in ELU per kg of the product or material being
considered) for use'of natural resources and energy, as well
1-10
-------
as for pollutant emissions, environmental indices for
materials and processes can be calculated. These indices
for materials and processes can then be used to calculate an
overall environmental load value (ELU) for the entire life
cycle of a particular product, process or activity.
ISSUES OP POTENCY AND SEVERITY
Chemical Risk
Potency
The dose required to elicit a toxic effect is referred to as
potency. Data regarding the potency of chemicals in test
organisms are frequently used to assess acute and chronic effects
to terrestrial and aquatic plant and animal species. Potency is
used in most scoring systems to evaluate acute effects in aquatic
and terrestrial species. Data regarding the potency of chemicals
required to elicit carcinogenic effects, as well as other chronic
effects, are often limited. Many scoring/ranking systems assign
scores for carcinogenicity based only on the weight of'tha
evidence that the chemical is a carcinogen. Examples include-
The Michigan Critical Materials Register t103
The German Existing Chemicals of Environmental Relevance
List I and II (DBA) [3-4]
The Ontario Ministry of the Environment (MOE) system t51
A few systems include both weight-of-the-evidence and
potency data. For example:
George Washington University m : assigns a .score according
to a matrix which includes a weight-of-the-evidence
classification as well as a potency factor (1/ED10) which
has been used to assign reportable quantities for
carcinogens by the CERCLA Section 102 program.
EPA DfE Use Cluster Scoring System I17]: weight-of-the-
evidence is combined with the reportable quantity potencv
factor or the qx*. *
CERCLA Hazard Ranking System [16]: assigns values based on
weight-of-evidence and the potency slope factor (a/) or an
estimate based on the ED10. '
Severity
How adverse is an adverse effect? Severity of effects, which
is the degree of a biological response, is not always considered
in ranking schemes. Within any human health effects category for
instance, there may be a range of severity. Neurol-oxicity for
Ml
-------
example, can include anything from a temporary narcotic effect to
long-term irreversible brain damage.
Some systems do include severity when assigning scores to
chemical substances. For example:
Toxic Substances Control Act Scoring System [11]: includes
severity in several parameters. For example, for the
nonlethal acute toxicity parameter, the dose score is
multiplied by the severity score to arrive at a final score
in this category. The severity of .effects are scored
according to whether the effect is life-threatening or
severe, moderately serious, mild or no effects are observed
at high doses.
Michigan Critical Materials Register tlo]: includes severity
in several parameters. For .example, a score for "other
toxicity" (chronic or subchronic) to terrestrial animals
includes severity ranging from adverse effects to severe
effects. Examples of each effect classification are_
provided to guide decisions iri response to the question of
how severe is the effect. For example, "moderate effects"
include effects such as degenerative or necrotic changes
with no apparent decrement of organ functions; or
reversible, slight changes in organ function.
Life Cvele Impact Assessment
The potency of human health and environmental impacts other
than those caused directly by chemicals is often taken into
account in life cycle impact assessment by the use of the impact
equivalency concept, where a reference substance is assigned a
unit impact of 1 and other substances are given a multiplier
based upon the potency of their impact as compared to the
reference substance. Thus, compounds that produce global warming
can be expressed and summed in terms of CO2 equivalents, which
gives an overall indication of the global warming potency.
Severity of impacts may be considered within impact
categories in deciding how to measure the impact. For instance,
impacts on endangered species can be measured by habitat
destruction or actual species loss.
EXPOSURE MEASURES
Chemical Risk
Any type of risk assessment includes an"assessment of both
toxicity and exposure. Factors affecting exposure to chemicals
released into the environment may include physicochemical
properties, environmental transport, persistence,
1-12
-------
bioaccumulation, actual measured releases, production and import
volume, use volume, probability and intensity of worker exposure
frequency and intensity of consumer exposure, and/or the
population in the area where releases occur. Since ranking and
scoring systems are usually not site-specific, they usually
include some type of surrogate for exposure, such as
environmental releases, when those are known, or a generic
exposure calculation. Examples include:
WMS- Scoring System [9-"': assigns scores for environmental
exposure according to use volume, percentage release to the
environment, degradation in air, soil and/or water, relative
occurrence in these media and bioconcentration. Exposure
via products is also scored, and this includes use patterns
exposure frequency and intensity of exposure.
George Washington University ™ : assesses exposure according
to the bioaccumulation, persistence, and release or
production volume. ,
Michigan Critical Materials Register fl«: scores chemical
exposure according to bioaccumulation, persistence, and a
few physicochemical properties such as flammabilitv
reactivity, and corrosivity.
The Ontario Ministry of the Environment (MOE) Scoring System
: includes environmental transport, environmental
persistence and bioaccumulation.
German (UBA) System »•«: includes bioaccumulation,
persistence and production volume.
EPA Toxics Release Inventory Environmental Indicators
Methodology ™ : attempts., to use facility-specific data and
generic models to estimate a, "surrogate dose" which is a
measure related to .amount of chemical an individual might be
exposed to in order to estimate human exposure,. A separate
evaluation is conducted for each release pathway allowing
comparisons across media. The level of uncertainty is
included in the^scoring for exposure. Exposure of aquatic
life is obtained by estimates.of the ambient water
concentration value. Table 4 summarizes surrogates for
exposure potential for six scoring systems
1-13
-------
1-14
-------
Life Cycle Impact Assessment
Life cycle impact assessment typically uses some
generic exposure assumption. The inventory usua ly dStene
whach environmental medium pollutants are released into and it
is assumed that they can exert their effects in that medium J
refinement of this approach for chemical releases would be to
incorporate modeling of chemical distribution af ler release such
a^ S%£5£2 ™dd ——ion after rife
AGGREGATION AND WEIGHTING OF HEALTH AND ENVIRONMENTAL IMPACTS
Chemical Risk
The chemical ranking and scoring schemes reviewed nrO
human health and environmental effecEs as wJ?l Is JxposSS
manner in which these parameters are combined to g ve an
ranking or risk categorization varies tremendousl? Fc?
Michigan Critical Materials Register t10' : chemical
selection is based on combinations of scores in the various
parameters. For example, selection for the Register may
occur if a chemical scores a ' 5 ' in two or more criteria
These crxteria include all of the health, environmental and
exposur,
carcinogenicity and c) mutagenicity.
Ontario Ministry of the Environment System ™ - substances
are selected on the basis of combinations of scores X
tOXICltV. r5(=T-HT Ol-on^a =»,^1 l^^ _________ -i . . • _ s=>*"^JJ-*=& J-J1
rZ • -I " . l-"e Dasis °r combinations of scores in
toxxdty, persistence and bioaccumulation. For exlmr>l£
substance is placed on the primary list if I- scopes a'
in any one of the seven toxicity categories mS has a
persistence of a half-life of 50 days and it has a
bioconcentration factor of more than 500.
Once scores are obtained for each of the various exposure
'10'
1-15
-------
it scores HIGH
and effects parameters, chemicals may be placed into categories,
such as selected or non- selected chemicals, or they may be ranked
in order of priority.
n^cro-rization may be accomplished by determining an overall
concern score of high, medium or low, or- by selecting chemicals
according ?o combinlt ions of numerical scores in the exposure and
effects categories. For example:
George Washington University Method [71 : assigns scores of ^
highfmedium or low in several toxicity categories, release
and production, persistence and bioaccumulation. ; A chemical
S sllected ae /candidate for Sunsetting i^l) it scores
HIGH in any toxicity category and HIGH in release and
production (excluding pesticides) OR if JO .J^scorea
in any acute or chronic toxicity category and HIGH in
persistence OR bioaccumulation (including pesticides)
WMS-Scoring System C9'141 : categorizes chemicals into three
areas of priority by plotting an image point . The image
poin? is determined after each chemical has been 'assigned a
rank for exposure and effects and these two are plotted in a
two-dimensional diagram.
Ontario Ministry of the Environment Candidate Substances
List fir Bans o? Phase-outs « , numerical scores are
assigned for environmental transport', persistence and
bioaccumulation as well as seven categories for toxicity.
Combinations of scores in these categories result in the
inclusion of the chemical on the primary or secondary lists.
A quantitative approach has been used by other ranking
schemes to prioritize chemicals in order of their potential
^to hSman and environmental health. For example:
University of Tennessee Method [13J : uses an algorithm which
includes both additive and multiplicative parameters to
determine an overall hazard value for each chemical. The
ctemicS rank indicates its hazard relative to the other
chemicals that are scored.
ATSDR Method [2] : places 275 substance's in order of hazard
potential based on the total score. The total score is
obtained by the following formula:
Total score = NPL frequency + Toxicity
(1800 max pts)
(600 pts)
(600 pts)
(600 total pts)
1-16
-------
t
Chemical score = Human risk reduction potential
+
Ecological risk reduction potential
1 ' ' ' • +• .
EPA interest
Chemical sc°re for the entire cluster is then added
P?evention Potential cluster score (the mean of
tia11 chemical scores and the higSer of
ecological risk reduction potential
1
Life Cycle Impact Asseggm<=>ni-
,.., As jf .ft were not difficult enough to weight and agqreaate the»
different direct chemical impact categories (eTg. , carcinoaeniSitv
versus acute LC50 for fish), in life cycle impact kssessmenf non/
^^^^^^
°1"°n
=urren*y for summing environmental releiee .
HOW MISSING DATA IS HANDLED
Chemical Risk .
1-17
-------
Examples of systems that use qualitative (SAR) and/or
ouantiSSvS (^MO structure-activity relationshxps for predxctxng
the effects of chemical exposure include:
"Schmallenberg" system [18] : states what minimum data set is
5s ?«£&£?.=£
sJmi-quantitative in nature, must be -considered.-
German (DBA) scoring system .»•«: allows the Assignment of
S^SrSS^S^ - ™-
use of negative scores was avoided in all but a few cases.
o
obtainSb?e through the use of QSARs for estimation.
by
concern .
OSARs when used appropriately, can be used to supplement
QSAKS, wnen ub FF £ purposes of screening chemxcals for
QSARs .
Other systems avoid the use of QSARs to fill data gaps by
can be lowered by filling data gaps.
CONCLUSIONS
Th^-re is currently no scientific consensus on human health and
environmental SsTranLng and scoring ^temSpedYSa?thouS the 9
=t,r) =<-or-ina chemical risk are much more developed, althougn tne
approac£esgvar? considerably. Given the interest in chemical rankxng
1-18
-------
and scoring, and the many potential users, an important first step in
filling in the framework for overall health and environmental ranking
and scoring would be to attempt to develop a consensus system for
ranking and scoring chemical risk.
Life cycle impact assessment methodologies are much less
developed at present, and the approaches being used vary dramatically
Some important policy questions about the uses of such methodologies
and the framewprk for decision-making must be resolved before a
scientific consensus can begin to be developed on the details of such
overall ranking and scoring systems.
1-19
-------
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Abt Associates, Inc. (1992). Toxics Release Inventory
Environmental Indicators Methodology. (Draft Report) Bethesda,
MD: U.S. Environmental Protection Agency Office of Pollution
Prevention and Toxics.
Aqency for Toxic Substances and Disease Registry. (1992).
Support Document: The CERCLA 104 Priority List of Hazardous
Substances That Will Be The Subject Of Toxicological Profiles.
Washington, DC: U.S. Public Health Service.
Behret, H. (Ed.). (1989). Existing Chemicals of Environmental
Relevance. GDCh-Advisory Committee on Existing Chemicals of
Environmental Relevance. New York: VCH.
Behret, H. (Ed.). (1989). Existing Chemicals of Environmental
Relevance II, Selection Criteria and Second Priority List. Gdch-
Advisory Committee on Existing Chemicals of Environmental
Relevance. New York: VCH.
Candidate Substances List for Bans or Phase-Outs. (1992).
Ontario Ministry of the Environment.
Criteria Working Group of A.R.E.T.S. (1992) . A Critique of the
Ontario Hazard Assessment System. Ottawa, Qnt: CLC Environment
Bureau.
Foran, J. A., Glenn, B. S. (1993). Criteria to Identify
Chemical Candidates for Sunset ting in the Great Lakes Basin.
Washington DC: The George Washington University, Environmental
Health and Policy Program, Department of Health Care Sciences.
Gios, N., Moller, M., Haegh, G. S., & Kolset, K.
Existing Chemicals: Systematic Data Collection and Handling for
Priority Setting. Oslo, Norway: Center for Industrial Research.
Konemann, H. & Visser, R. (1988). Selection of chemicals with
high hazard potential: Part 1: WMS-Scoring System.
Chemosphere, 17, 1905-1919.
Michigan Critical Materials Registry (Criteria and Support
Documents). (1987). Michigan Department of Natural Resources.
O'Bryan, T.R. & Ross, R.H. (1988).
Hazard and Exposure Identification.
119-34.
Steen, B. & Ryding, S. (1992). The> EPS Enviro-Accounting
Method. Goteborg, Sweden: Swedish Environmental Research
Institute (IVL).
Chemical Scoring System for
J. Toxicol. Env. Health, 1,
1-20
-------
13.
14.
15.
16.
17.
18.
Schultz, T. W., Bartmess, J. E., & Davis, G. A. (1993).
Characterization and Assessment of Chemical Pollutants: Exposure
and Impact Analyses. Knoxville, TN: University of Tennessee.
Timmer, M., Konemann, H. & Visser, R. (1988).
chemicals with high hazard potential: Part 2:
System. Chemosphere, 17, 1921-1934.
Selection of
WMS-Scoring
U.S. Environmental Protection Agency. (1985, 1986, and 1989).
Technical Background Document to Support Rule Making Pursuant to
CERCLA Section 102 (Vols. 1-3).
U.S. Environmental'Protection Agency. (Dec. 14, 1990). Hazard
Ranking System; Final Report. (40 CFR Part 300). Federal
Register (55:241, pp. 51532-51667). Washington, DC: U.S.
Government Printing Office.
U.S. Environmental Protection. Agency Office of Pollution
Prevention arid Toxiqs, Chemical Engineering Branch. (1993).
Chemical Use Clusters Scoring Methodology.
Weiss, M., Kordel, W., Kuhnen-Clausen, D., Lange, A. W., & Klein,
W. (1988) . Priority Setting of Existing Chemicals. Chemosphere,
17, 1419-1443. '
1-21
-------
APPENDIX A
SCORING AND RANKING SYSTEMS IDENTIFIED
Abt Associates, Inc. (1992). Toxics Release Inventory Environmental
Indicators Methodology. (Draft Report). Bethesda, MD: U.S.
Environmental Protection Agency Office of Pollution Prevention and
Toxics.
Agency for Toxic Substances and Disease Registry. (1992). Support
Document: The CERCLA 104 Priority List of Hazardous Substances That
Will Be The Subject Of Toxicological Profiles. Washington, DC: U.S.
Public Health Service.
Behret, H. (Ed.). (1989). Existing Chemicals of Environmental
Relevance. GDCh-Advisory Committee on Existing Chemicals of
Environmental Relevance. New York: VCH.
Behret, H. (Ed.). (1989). Existing Chemicals of Environmental
Relevance II, Selection Criteria and Second Priority List. GDch-
Advisory Committee on Existing Chemicals of Environmental Relevance.
New York: VCH.
Bouchard, D. (1991). Review of Region VII TRI Strategy. (Memo, EPA
Region VII).
Candidate Substances List for Bans or Phase-Outs.. (1992). Ontario
Ministry of the Environment.
Criteria Identifying High Risk Pollutants. (1991). Environmental
Reporter (pp. 463-465). Washington, DC: Bureau of National Affairs,
Inc.
Crutcher, M. R., & Parker, F. L. (1990). A Classification System f«^
Hazardous Chemical Wastes. Superfund 90, Hazardous Materials Control"*
Research Institutes, llth Annual National Conference, 222-225.
Environmental Protection Agency. (Dec. 14, 1990). Hazard Ranking
System; Final Rule. Federal Register (40 CFR Part 300).
Foran, J. A. & Glenn, B. S. (1993).. Criteria to Identify Chemical
Candidates for Sunsetting in the Great Lakes Basin. Washington, DC:
The George Washington University, Environmental Health and Policy
Program, Department of Health Care Sciences.
Gjos, N., Moller, M., Haegh, G. S., & Kolset, K. (1989). Existing
Chemicals: Systematic Data Collection and Handling for Priority
Setting. Oslo, Norway: Center for Industrial Research.
Gustafsson, L. & Ljung, E. (1990). Substances and Preparations
Dangerous for the Environment: A System for Classification, Labelling
1-22
-------
and Safety Data Sheets. Miljorapport: Nordic Council of Ministers.
Halfon, E., & Reggiani, M. G. (1986). Notes on Ranking Chemicals for
Environmental Hazard. Environ. Sci. Technol. 20, 1173-1179.
Hallstedt, P.A., Puskar, M. A., & Levine, S. P. (1986). Application
of the Hazard Ranking System to the Prioritization of Organic
Compounds Identified at Hazardous Waste Remedial Action Sites.
Hazardous Waste and Hazardous Materials, 3(2), 221-232.
Hawley, K. A., & Napier, B. A. (1985). A Ranking System for
Hazardous Sites & With Mixed Radioactive and Hazardous Wastes (Comment
Draft Report). Richland, WA: U.S. Department of Energy, DOS, PNL.
Hutchinson, W. R., & Hoffman, J. L. (1983). A Ground Water Pollution
Priority System (N.J. Geographical Open File, Report No. 83-4).
Trenton, NJ: Division of Water Resources.
International.Joint Commission's Binational Objective Development
Committee. (1989). The Great Lakes Water Quality Agreement Annex 1.
Lists 1, 2, 3.
Jones, T. D., Walsh, P. J., Watson, A. P., Owen, B. A., Barnthouse, L.
W. , Sanders, D. A. (1987). Chemical Scoring by a Rapid Screen of
Hazard (RASH) Method. Risk Analysis, 8(1), 99+
Klein, W., Kordel, W., Klein, A. W., Kuhnen-Clausen, D., & Weiss, M.
(1988) . Systematic Approach for Environmental Hazard Ranking of New
Chemicals. Chemosphere, 17, 1445-1462.
Konemann, H. & Visser, R. (1988). Selection of chemicals with high
hazard potential: Part 1: WMS-Scoring System. Chemosphere, 17,
1905-1919.
Laskowski, P. A., Goring, C. A. I., McCall, P.J., & Swann, R.L.
(1982). Principles of Environmental Risk Analysis: Terrestrial
Environment. R. Conway (Ed.), Environ Risk Analysis for Chemicals,
pp. 198-240. New York: Van Nostrand Reinhold.
Michigan Critical Materials Registry. (Criteria and Support
Documents). (1987). Michigan Department of Natural Resources.
Morgenstern, R., Clay, D., Emison, G., Hanmer, R., & Williams, M.
(1987). USEPA Unfinished Business Report: A Comparative Assessment
of Environmental Problems. Washington, DC: U.S. Environmental
Protection Agency.
O'Bryan, T.R. & Ross, R.H. (1988). Chemical Scoring System for
Hazard and Exposure Identification. J. Toxicol. Env., Health, 1, 119-
34.
Ontario Municipal Industrial Strategy for Abatement (MISA) Effluent
1-23
-------
Monitoring Priority Pollutants List.
Ministry of the Environment.
(1987, 1988, 1989). Ontario
Pavia, R., & Harris, L., (Eds.). (1985). Coastal Hazardous Waste
Site Review, 30 June 1985. Seattle, WA: Ocean Assessments Division,
Office of Oceanography and Marine Assessment, NOS, NOAA.
Rechard, R. P., Wilkinson, G. F., & Schreiber, J. D. (1991). User's
Manual for SRS88: 'Site Ranking System for Chemical and Radioactive
Waste (SAND87-2815, DOE/HWP-26). Albuquerque, NM: Sandia National
Laboratories.
Sampaolo, A. & Binetti, R. (1986). Elaboration of a Practical Method
for Priority Selections and Risk Assessment among Existing Chemicals.
Reg. Toxicol. & Pharmacol, 6, 129-154.
Schultz, T. W., Bartmess, J. E., & Davis, G. A. (1993).
Characterization and Assessment of Chemical Pollutants: Exposure and
Impact Analyses. Knoxville, TN: University of Tennessee.
Silka, L. R., & Swearingen, T. L. (1978). A Manual for Evaluating
Contamination Potential of Surface Impoundments. (EPA 570/9-78-003).
Washington, DC: Environmental Protection Agency.
Steen, B. & Ryding, S. (1992). The EPS Enviro-Accounting Method.
Goteborg, Sweden: Swedish Environmental Research Institute (IVL).
U.S. Department of Defense. (1991). User's Manual for the Defense
Priority Model (FY 1992 version). Office of Deputy Assistant
Secretary of Defense. (Environment).
U.S. Environmental Protection Agency. Hazardous Air Pollutants:
Proposed Regulations Governing Constructed, Reconstructed and Modified!
Major Sources (40 CFR.Part 63).
U.S. Environmental Protection Agency. (1985). , A Ranking System for
Clean Water Act Section 307 (a) List of Priority Pollutants.
U.S. Environmental Protection Agency. (1985, 1986, and 1989).
Technical Background Document to Support,Rule Making Pursuant to
CERCLA Section 102 (Vols. 1-3).
U.S. Environmental Protection Agency Office of Air Quality Planning
Standards, Chemicals and Petroleum Branch. (1990). The Source
Category Ranking System: Development and Methodology.
U.S. Environmental Protection Agency Office of Air Quality Planning
Standards. (1991). National Emissions Standards for Hazardous Air
Pollutants for Source Categories: Proposed Regulations Governing
Compliance Extensions for Early Reductions of Hazardous Air
Pollutants.
1-24
-------
U.S. Environmental Protection Agency Office of Environmental Criteria
and Assessment. (1986). Examination of the Severity of Toxic Effects
and Recommendation of a Systematic Approach to Rank Adverse Effects.
U.S. Environmental Protection Agency Office of Health and
Environmental Assessment. (1986). Screening Procedure for Chemicals
of Importance to the Office of Water.
U.S. Environmental Protection Agency Office of Policy Analysis.
(1978) . Measuring Air Quality: The New Pollutants Standards Index.
U.S. Environmental Protection Agency Office of Policy Analysis.
(1991). Ranking the Relative Hazards of Industrial Discharges to
POTWs and Surface Waters.
U.S. Environmental Protection Agency Office of Policy, Planning and
Evaluation, Pollution Prevention Division. (1990). Targeting
Pollution Prevention Opportunities Using the 1988 Toxics Release
Inventory.
U.S. Environmental Protection Agency Office of Pollution Prevention
and Toxics, Chemical Engineering Branch. (1993). Chemical Use
Clusters Scoring Methodology.
U.S. Environmental Protection Agency Office of Toxic Substances.
Screening Methodology for Pollution Prevention Targeting.
U.S. Environmental Protection Agency Office of Toxic Substances.
(1989) . Toxic Chemical Release Inventory Risk Screening Guide (Vol.
1) .
U.S. Environmental Protection Agency Office of Toxic Substances.
TSCA's TRI Chemical Risk.Assessment Pre-Screening Methodology.
Weiss, M., Kordel, W. , Kuhnen-Clausen, D. , Lange, A. W,, , & Klein, W.
(1988) . Priority Setting of Existing Chemicals. Chemosphere, 17,
1419-1443.
Whelan, G., et al. (1985). Development of The Remedial Action
Priority System: An Improved Risk Assessment Tool For Prioritizing
Hazardous and Radioactive-mixed Waste Disposal Sites. (PNL-SA-13385).
Richland, WA: Pacific Northwest Laboratory.
1-25
-------
-------
APPENDIX J
GRAMIC REFRESENTAHONS OF TWO RISK RANKING CTAMEWORKS
-------
-------
Framework for Assessment and Ranking
in
05
Crt
CD
W
O5
3
CD
Human
Impacts
Health
Descriptor
Purpose/
Scope
Ecological
Impacts
Resource and
other
Environmental
Impacts
Ranking
Systems
Social/
Economic
Impacts
np«rri JmPaCt Resource Depletion Social/Economic
Descriptor Descriptor Impact Descriptor
U)
Environmental
Valuation or Weighing
Involvement
of Stakeholders
CO
CD
3
CD
Improvement Opportunities
Options
Selection
j-i
-------
Proposed Framework for Environmental Valuation
Economic
Consequences
Prospective
Action
Performance
Consequences
Environmental
Consequences
•o
o
£1
4_>
0>
z:
Don't Know
cc.
Overall
Valuation
No
CD
<
CO
fD
>
n
r-c
O
implement Action
Evaluate Consequences
J-2
-------
APPENDIX K
HEALTH IMPACTS AND LIFE CTCLE ASSESSMENT
-------
-------
,992:9,
Health impacts and Life Cycle Assessment
Amte Schmidt®, Jens Erik ,e,nes», Lisbe* Enge, Ha^en' and Allan As.n.p
Summary
©: dk-TEKNIK, Gladsaxe M<,IIevej 15, 2860 S0borg, Denmark
'
14I,
-------
The
1. Introduction
T ife Cvcle Assessment (LCA) is rapidly becoming an important tool in environmental
S£S!fe last fel years has shown the need for a more m-depth hazard
evaluation of the data reported in the inventories.
from collecting data in relation to the impacts in a materials' life cycle (the
/^S ^o Evaluation of this information (the life-cycle assessment) has
obeen on vSy few occasions, although it is a great wish from potential users
of LC?s The majorlplanation is probably that this step today cannot be performed
on a folly scientific basis. No commonly accepted method has been developed, and
tofore more caution in defining system boundaries and stating omissions and
JSwe sourcl of misinterpretation must be observed when taking this step than
when presenting the life-cycle analysis.
1.1 Lack of scientific basis in health assessment
The lack of scientific basis may be divided into two categories, namely a lack of data
and A?subjSve evaluation of the available information. These two. issues are
discussed briefly below.
1J.1 Lack of data
Lack of data may be encountered both in the analysis and the assessment phase. With
S^cteLd. are without great importance with respect to *+™™^
energy and raw materials, but seen from the evaluation side they may be very
tmoortant These substances' may of course also prove to be of no importance, but if
S aTnot lisSl any where in u inventory the importance will never be determined
S^ack of Information is especially important when assessing the potential
occupational health impacts.
With respect to evaluation of the potential impacts on the health of the general
populatio^he inventories are not very specific as they are ^«*^*££
environmental impacts. For instance, many inventories accounts for the totol emission
of hydrocarbons per kg material, but there is no specification of these hydrocarbons.
As ther^S a very great difference between the potential impact of a carcinogenic
fub Sice e g Ae, and a relatively innocent compound like heptane. This lack
of s^ificity make any health evaluation impossible When assessing he : potent^
healSTimpacts it is therefore necessary to make a supplemented data collection to fill
in the information gaps of a "regular" inventory. The supplementary data are often
quaTi Ltive by Mature and will thus only provide sufficient information for a hazard
identification.
K-2
-------
Coverall t^c*vrf P™««- » insufficient when
A 2
Subjective evaluation
* Age of information,
* Test design and methods, and
* Reproducibility of results between laboratories
Which results
^
K-3
-------
2. Reports with an approach to health impact assessment
Although it is agreed upon by LCA-practitioners that a clear-cut evaluation of a Hfe-
c^cSSlysis STfaSte, some attempts to perform an evaluation have been mad^
The background for these attempts is that only in very rare cases is it possMe to
lonclude anything from a life-cycle analysis, and therefore * is a strong wish from aU
potential users of life-cycle assessments (authorities, consumers, producers) to have
a tool for interpretation of the life-cycle analyses.
The following discussion is mainly focused on the assessment of the impacts.
2.1 The BUS- and BUWAL-reportsIA3
These Swiss reports are commisioned by the Swiss Environmental Protection Agency
mundes^t rJ^Umweltsschutz) with the aim of producing a system where it is
possMe to choose the most environmental friendly packaging among a number-of
atenudves. The reports do not evaluate the potential health impacts of each emiss^n
fiTSaU, but aggregates'the relation between each emission quantity and he
corresponding MIC-value (Maximal Immission Concentration) according to the
following formula:
Emission Quantity (mg/feg)
Maximal Immision Concentration (mg/m3)
= Critical Volume (m3/kg)
The "Critical Volume" reached by the formula should thus represent the theoretical
amount of air necessary to dilute to a "safe" concentration the emissions to air from
one Wo of material throughout its life-cycle. Lack of a Swiss MIC-value4 on an
emisSon is at first handled by using German standards' (VDI-guidelmes). If no Swiss
o^rmaL MIC-value is found, a proportional interpolation of Occupational Exposure
Limits (MAK-Werte) is used.
A similar approach is used for evaluation of the emissions to water. In this context
however, the limit values are taken from Swiss legislation on waste water emissions
Together with aggregated data on energy consumption and generation_ of solid waste
these elements are considered to constitute an ecological profile per kilo of a given
material Specifying different products or, preferably, a functional unit then yields
Sr^ssibiSty of calculation of the overall impact this functional unit may have
LSout the life-cycle. In order to be more environmentally friendly a material must
perform better in all four elements (critical volumes in air and water, energy
consumption, generated volume of solid waste).
Use of legislative standards in the overall assessment of a material causes a number
of problems. As the methodology allegedly describes the environmental impacts it is
not very transparent how the effects are divided into health and environmental effects
and the methodology can thus not be used for a health assessment without a scrutiny
K-4
-------
2.7.7 Ecological scarcity and eco-points
the actual level of antropogenic emissions of this pollutant on the other hand'
The calculation of the "Ecofactor" is done according to, the formula
Ecofactor = 1/Fk * F/Fk * c
K-5
-------
- emissions to air,
- emissions to water,
- energy consumption and
- waste generation.
The final result is a number of Ecopoints that should be comparable between different
materials and products.
'suf a nail L the methodology as a whole or as a basis for assessing human
Impacts is not suitable for the Nordic (or any scientifically based) study.
2.2 The DTI-report on PVC and selected alternative materials
» ss
a voluntary reduction of the consumption of PVC.
S Furthermore the Danish authorities underlines the necessity -of
occupational health assessment in the overall evaluation and ^ was thus
to develop an integrated method for assessing the environmental impact of
the following elements:
* Consumption of resources (including energy)
* Potential exposure in the work environment
* Potential effects in the work environment
* Potential exposure in the natural environment
* Potential effects in the natural environment
* Risk of accidents.
The overall methodology has been summarized elsewhere9, and the present review
therefore focuses on the assessment of potential health impacts.
As no "regular" life-cycle analysis applies to the work environment the basis chosen
for the assessment is qualitative to a large extent. The principle in the asses ™nti
that a consultation of experts and review of literature sources is performed in order
to creTn the Materials for possible impacts. In those cases where the occupational
hea^mTmpact T were considered to be of possible importance a more extensive
K-6
-------
documentation of the potential impacts was established. Finely, both exposure and
effects were scored in one of the four categories:
Potential for severe impact
Potential for some impact
No potential for impact
Insufficient data for evaluation
The authors of the report are aware that this methodology has a number of serious
limitations of which the major considerations are: Problem areas may be forgotten or
maybe no longer exists, only available (= old) literature is used, variations between
plants and countries are not documented, and focus is on materials and not on
products.
As the methodology is expert-based the subjectivity of the assessment may become a
dominant feature of this methodology. This subjectivity will be more pronounced with
decreasuig amounts and quality of the data. However, the documentation of the
potential exposure and effects in critical elements is very transparent and therefore a
suitable basis for discussion. If more detailed and up-to-date inventories including the
work environment, could be produced some of the subjectivity associated with the
me hodology would disappear. The remaining subjectivity which is related to the effect
evaluation would not be so important as there is full transparency in the documentation
tor the assessment
10
2.3 The EPS-system
Z?e,?PSQ~SySjem (Environmental Priority Strategies in Product Design) was developed
by the Swedish Environmental Research Institute in order to give designers and
•££? T£ the P^1^ °f asse^"g the total impacts of a product fronf cradle to
grave The model is based on aggregation of four impact elements: Resource (raw
Sh H enefgy' } SCardty Md th£ CffeCt °f emissions to Jiir' soil a"d *ater °"
health and environment. One emission or situation may have several impacts, e.g on
the foCwTng^a acidification- Each of these situations are treatedI seperatlly in
For the effect-indices a combination of several figures is used: Intensity or frequency
of an effect, extent and duration of the effect, contribution to the total effect in
fac'torTh;^ •? drTg thC emiSSi°n by °ne Wei*ht unit' ^ an a^essment
factor that considers the degree of inconvenience of a negative effect with a given
intensity and limits in time and space.
If a substance is contributing to several problems an index is calculated for each
problem and these indexes are added to get the total index for the substance. The
1S r!^ by the emitted am°unt ^ M "Environmental
a materia1' a Pr°duet' °r Jl process'the
K-7
-------
the conclusion.
24 The Danish Materials Technology Development Programme"
^necessary. Such a development is presented in the present report.
2.5 Development of environmentally friendly products12
The Technical University of Denmark, Institute for Product Development and the
critical elements in the life-cycle of the chosen alternative.
The life
inventory should include both raw materials, additives, auxiliaries, and
the relevant processes. There is therefore a ne^ for more mform^on
than normally generated in a life cycle analysis.
The inventory screening is based on Danish regulations for the external environment
t "All compounds and their potential effects, i.e., carcmogemci-
K-8
-------
Al'J^h gyi' ne"rotoxlclty' ^production toxicity, corrosion and irritation are listed
Also the classification and R-sentences are noted in the list. The effects are graded in
three classes with corrosion and irritation in the lowest category, allergy neurotoxici-
ty, genotoxicity, and systemic effects in the middle category and c^cer
reproduction toxicity in the highest and most serious category6 do an^ng ™
HoweveTfhi^ *" — ^ *"» * ^^ ™nt of **
However, this screening step is not meant for decision-making,,
^ i
hygielts
Pr0file' ** P016"^ effects m more thoroughly evaluated
^fential eXP°SUre (eXp°SUre Wa*S' ma£nitud*> is «nS
this step is performed by toxicologists and industrial
In step 4, the number and exposure level of chemical compunds in different effect
categories are counted, and the best alternative should be the one wi*£ ^smallS
number of compounds in the highest category.
The final step in the methodology is rather superficial as the number of chemicals in
each category is the only parameter used for choosing the right: alternative No
consideration is given to e.g., the potency of the carcinogenic cLpounds a^d £e
potent^ for exposure is only described very briefly. An experienced LcologS may
6 " tive' but U is ^ doub^ul Aether a de%neT
***** '
2.6 Environmental Risk Management
The Danish company COWI Consult has developed a framework for environmental
nsk management13. This framework is based on a life cycle analysis aTd XTbo^
IT?Sf oS relT t? lmPaCt im° aCCOUnt' ^ P^"- ^V°lves ^k
1) A list of all relations between a given industrial activity and the environment fan
inventory) is made, and 2) The inventory is elaborated into a
environmental and human health relations including priority
ranking is performed on the basis of the following matrix-
K-9
-------
Assessment
parameters.
No protection
Very good
protection
Dispersion in
working
environment
Occupational
health/safety
Reproduction
and cancer risks
Irritation,
illness
Effect on,
health
Dispersion
cale
Reproduction
and cancer risks
Irritation,
illness
Effect on
health
As for "Amount" under health impact, other
parameters unchanged
Amount x
probability
Each of the three relation to human health impact is assigned three assessment
posters, each of which may have the score 1, 2 or 3. Reasons should be given
Sv and briefly for each score assigned, and the total assessment score for each
Son * obSned by multiplying the three assessment scores. If more than one score
Is oSed Tm a given activity, i.e., there are more than one emission, the worst
score is taken as the indicator of the impact.
The priority ranking system is basically very similar to the screening
will be described later. As the system is meant as a tool for enviromental risk
in industries the author emphasize that it is not an exact system, but
ogic^l framework for the assessment proces. In this context the absence of
criSria for exposure and effects assessment is of less importance. However,
Lessment in lifecycle context the development of such catena is crucial.
2.7 Other methods
As life-cycle assessment is a fairly new discipline
health assessment may exist. One such method will soon be published by the Tellus
Sfutet i uTA. Their method has been developed in a study on the environmental
impacts of packaging materials, but no information is available at present.
K-10
-------
2.6 Conclusion , .
Under all circumstances should the goal definition of each of the above mentioned
reports be kept in mind when discussing the methodology and the results. The reports
discussed above have all been criticized because of the methodology, the results, or
both. Some of the criticism has forgotten to take the goal definition into account, and
it must be realized that a 100% objective assessment methodology cannot be
developed. It is however necessary to make attempts to assess and compare the
potential impacts of a material or product, as the life-cycle concept otherwise lose its
value. If the "hired-gun"-criticism shall be avoided, it is necessjiry that objectivity is
maintained as far as possible and that it is clearly stated whenever the evaluation
becomes subjective.
3. Description of a new method for screening and priority setting
3.1 Introduction
In the following a new method is presented. The method does not pretend to give ah
unequivocal assessment of materials but is rather a method which gives information
on the seriousness of the potential health impacts of a material. It must be stated that
this method also calls for subjective evaluations, but we think that the subjectivity is
minimal in relation to the efforts that a purely objective assessment, if possible, would
demand. In order to facilitate the evaluation, the methodology follows the normal way
of making a risk assessment as far as possible without pretending to be an actual risk
assessment. In the screening procedure, guidelines on how to h,andle lack of data is
incorporated.
3.1.1 Prerequisites
One of the basic prerequisites for a thorough assessment of the potential health
impacts of a material is full information on exposure to and effects of all emissions.
Exposure data must relate to conditions in the work environment, in the immediate
vicinity of the production plant, regionally and globally. Exposure data should
consider possible uptake via the respiratory tract and the skin arid,, orally, by drinking
water and eating contaminated food. The effect data should include acute toxicity,
irritation and, more important, long-term effects like cancer, reproduction toxicity,
neurotoxicity and chronic organ damages.
Full information is rarely found even for major chemicals. Generation of the data will
be extremely time-consuming and therefore it is necessary to make a number of
assumptions and simplifications, if health impacts are to be integrated in Life Cycle
Assessments. In the making of these assumptions and simplifications a degree of
subjectivity is introduced, and a methodology based on this should be regarded as a
screening tool or a tool for making priorities for a more in-depth evaluation. It is
indicated that the following method can be used for these purposes. It is also possible
that the degree of subjectivity is so small that comparisons between emissions and
materials may be possible. However, only testing of the method will show its
limitations.
-------
3.2 Exposure
The exposure assessment should whereever possible be related to operational units.
As basic operational units occupational exposure limits and environmental quality
standards are suggested. In doing so it is recognized that these are the maximum
values occurring at normal circumstances. The values are not used to assess the effects
but only to relate exposure to official policy, and a division into three exposure
categories will not cause any scientific "damage" in a priority system. In the priority
system the possible exponential relation between exposure and effect is reflected by
the rating ofthe three categories. The ratings of 1, 2, and 4 are thus to be regarded
as indicators of potential for impact rather than actual scores. When combining the
rating for exposure and effect this will point to the most critical situations, where an
in-depth evaluation is necessary.
"Low" exposure:
"Medium" exposure:
"High" exposure:
Rating = 1
Rating = 2
Rating = 4
As the exposure situations varies widely according to geographic distribution, different
categories are made for four situations:
Wnrlf environment:
Low exposure:
Medium exposure:
High exposure:
Detected
Generally < 1/lOofOEL
Frequently > 1/10 of OEL
Immediate vicinity (r < 1 km)
Low exposure:
Medium exposure:
High exposure:
Regional (r < 25 km)
Low exposure:
Medium exposure:
High exposure:
Global
Low exposure:
Medium exposure:
High exposure:
Always < 1/10 of quality standard
Seldom > 1/10 of quality standard
Generally > 1/10 of quality standard
Detected
Generally < 1/10 of quality standard
Frequently > 1/10 or quality standard
Trace amounts
< 1/10 of quality standard
> 1/10 of quality standard
The suggested criteria for both the geographical and exposure division are not based
on scientific findings but are set arbitrarily. The choice of 1/10 of the OEL or the
quality standard can be seen as a "safety" factor in the priority system which indicates
K-12
-------
that these values not always are set on a purely scientific basis. The use of a safety
factor of 10 is thus justified in a screening procedure, and only a more thorough
exposure assessment can estimate the real situation. In this context it is important that
all possible exposure routes are taken into consideration and thus an evaluation of
distribution and fate of the emissions is an important part of the: exposure assessment.
As for the use of the terms "seldom", "generally" and "frequently" in the exposure
assessment it should be pointed out that they are intended to encompass possible
exposure under normal conditions as well as during e.g., uncommon weather
conditions or accidents. In the work environment the exposure from a well-controlled
process should generally be below 1/10 of the OEL and only exceed this value in case
of poor process management or accidents. However, if the process for some reason
is difficult to control, there may be frequent excursions above 1/1.0 of the OEL and
the exposure will accordingly be rated "high". In the regional environment the term
"seldom" applies to situations like inversion or atypical operating conditions, e.g.,
cleaning operations, starting procedures and minor accidents, occurring at most a few
times per year. In the context of regional exposure, "generally > 1/10 of quality
standard" means that there is a general impact on the environment and that this impact
can be observed at almost all occasions.
It can already be foreseen that occupational exposure limits or quality standards cannot
be found for all of the emissions described in the life cycle analysis. As a preliminary
approach to this situation it is suggested that the exposure level for such compounds
are rated as "medium". When combining exposure and effect ratings it should then
become evident if these situations are regarded as critical.
3.3
Effect criteria
The criteria for evaluation of the effects are developed from ideas based on Gj0s et
al. (1989)14 and Schmidt et al (1990)"..The major change relative to the references
is that all categories and effects have been filled in.
Where internationally accepted systems for labelling or classification exist these have
been incorporated. Thus for the effect irritation, substances classifiable with R-phrases
according to the EEC labelling guide have been grouped according to seriousness of
the effects, ie irritating substances are in category 2 and corrosive substances in
category 3. Similarly, the criteria of IARC has been included with modifications in
the categorization for carcinogenic effect. The lowest score (Category 1) is given to
substances with inadequate evidence for carcinogenic effect (IARC group 3), and the
highest score (Category 3) is given to substances with animal or human evidence for
carcinogenic effect (IARC Group 1, 2A and, 2B). Substances, where no studies on
carcinogenic effect can be found, are placed in category 2 (sees below).
In the evaluation of substances for effects on irritation, sensitisation, and carcinoge-
nicity it should be considered if the substance fulfils the criteria for classification.
Therefore it is not enough to check in the published lists, if the chemical is on the lists
or is to be marked with an R-phrase.
K-13
-------
For most of the other effects no internationally accepted criteria exists. The cut-off
values presented in the table have been set partly from unofficial criteria used by the
Danish Environmental Protection Agency (Milj0styrelsen) in the evaluation of
pesticides.
If an effect of a compound cannot be assessed as belonging to one of the three
categories because of lack of data, it is preliminary put in category 2.
In the final evaluation of the health effects of a chemical information on all the effects
listed in the table should be included, and the substances are grouped as listed below.
"Low" effect (score 1):
"Medium" effect (score 2):
"High" effect (score 4):
At most one effect from Category 2
At least two effects from Category 2
At least one effect from Category 3
The exponential scoring (1, 2, and 4) will, when combining exposure and effect,
clarify the final evaluation and indicate if an emission is critical in the life cycle
assessment.
Table 1. Effect categories in health assessment.
Element
Acute toxicity
LDM, oral, rat, mg/kg
LDM> skin, rat/rabbit, mg/kg
LCM, inhalation, 4 h, rat, mg/1
Irritation
Sensitisalion
Organ toxicity
Genotoxicily
Carcinogenicity
Category 1
LDM > 2000
LDM > 2000
LCM > 20
No corrosive or irritative
effects
•Negative result in the
GPM test or equivalent
No organ effects seen at
1/10 of lowest dose level
for category 1 acute toxici-
ty in studies lasting at least
28 days
No evidence of mutation or
chromosome damage in in
vivo or in vitro test systems
IARC group 3 or equivalent
data
Category 2
25 < LDjo <. 2000
50 < to* <. 2000
0,5 < LCM <_ 20
Irritating to skin (R38), eyes
(R36) or respiratory system
(R37)
Positive result in GPM test
or equivalent, but no human
evidence
Organ effects seen at dose
levels of 1/10 for category 2
high limit but above the
category 3 limit in acute
toxicity in studies lasting at
least 28 days
Up to 3 nonsignificantly
positive results for mutation
and or chromosome damage
in in vivo or in vitro systems
IARC group 2B. but no R45
Category 3
LDjo <. 25
LDM <. 50
LCM\<.0,5
Corrosive (R34, R35), seri-
ous eye damage (R41)
Human evidence for allergy
(inhalation or skin contact
(R42, R43). On more than
2 country lists for allergens
Human organ effects are
known. Organ effects seen
at dose levels below the
category 3 high limit for
acute toxicity in studies las-
ting at least 28 days
At least one significantly
positive result for mutation
and or chromosome damage
in in vivo or in vitro sys-
tems
IARC group 1 , 2A or 2B
with R45 or equivalent data.
K.-14
-------
Element
Neurotoxicity
Reproduction toxicity/
Teratogenicity
Category 1
No neurotoxic effects seen
in studies lasting at least 28
days at lowest dose level
given under category 1
acute toxicity
No maternal toxicity and no
effects seen on fertility or
foetus at 1/10 of lowest
dose level for category 1
acute toxicity
Category 2
Neurotoxic effects reported
in studies lasting at least 28
days at dose levels for cate-
gory 2 acute toxicity
No maternal toxicity but ef-
fects seen on fertility or
foetus at dose levels below
1/10 of highest level for
category 2 acute toxicity but
not within limit of category 3
acute toxicity
Category 3
Hunuin evidence for neuro-
toxic effects. Neurotoxic
effects reported in studies
lasting at least 28 days at
dose levels for category 3
acute toxicity
No maternal toxicity but
effects seen on fertility or
foetus at the dose level for
category 3 acute toxicity
GPM test: Guinea pig maximisation test
IARC : International Agency for Research on Cancer
Note: For effects where internationally accepted criteria exist, the chemical under
concern should be evaluated to see, if it fulfils these criteria.
3.4 Description of effects
Regulatory agencies in different countries have set criteria for when to classify a
chemical as for instance a poison, a carcinogen or a teratogen. These criteria are
mainly based on data from animal, bacterial or in vitro experiments, since experience
of human exposure to one certain chemical is rarely available. Exceptions to this gene-
ral observation are the medicines, which have been through clinical trials before they
are marketed, and toxic agents such as asbestos, mercury, lead, vinyl chloride, and
alcohol, which have shown clear human effects.
In general, most of the information which can be gathered on toxicity of a chemical
is based on animal studies using mammalian species such as mice, rats, rabbits, guinea
pigs or hamsters. As a response to the increasing pressure against such animal experi-
ments, development of alternative test methods using cell cultures or non-mammalian
species is in progress. At present the greatest success within this field has been
obtained in the area of genotoxicity testing.
In the following a general and brief overview of the lexicological effects selected as
criteria in the screening of chemicals for human health risk in connection with life
cycle assessment will be presented.
3.4.1 Acute toxicity
Testing for acute toxicity in animals (rodents) is rapid and fairly inexpensive. For
almost all chemicals data on acute toxicity via some route of administration can be
searched and most often found. In the predictive toxicology administration via the oral
and dermal route and via inhalation can be used to group the chemicals into categories
according to the dose levels causing lethality in 50% of the animals after a single dose
(LDsoor LC50). Due to differences in absorption, metabolism and sensitivity it is not
K-15
-------
uncommon that the LD50- or LC50-values vary by a factor of ten using the same route
of administration when different species are investigated.
Regulatory agencies and OECD have established guidelines for this and other types
of testing. They group the chemicals into categories according to results obtained
following these guidelines. In dosing orally and dermally a fixed amount of chemical
is given to the animals at a specific time. However, when using the inhalation route
of exposure the duration of exposure in hours is also important. Most guidelines use
a 4-hour exposure for LC50-value classification. For any LC-value the duration of ex-
posure should be indicated.
Sometimes only LDLO-values are available. LDLO-values give the lowest concentration
at which lethality has been observed. The LDLO-values can be used to give a very
rough estimate of the LD50-values. As LD50 values always are higher than LDLO
values, a LDLO value can be used in classification, if it is above the high limit for
LD50 of a category in table 1.
3.4.2 Irritation
Gasses, most liquids and quite a number of solid chemicals have a sufficient vapour
pressure at room temperature to cause activation of the sensory system of humans, ie
they can be smelled. If this sensory activation is regarded as unpleasant or adverse to
the eyes or respiratory tract the substance is regarded as irritating. Furthermore, dust
from solid chemicals may cause irritation, if the dust is inhaled or deposited in the
eyes.
Testing for irritative properties, including corrosion, can be carried out in laboratory
animals, and guidelines for testing eye and skin irritation in animals exist (eg OECD
Guidelines 404 and 405). Most often rabbits are used as test animals for this type of
test.
Testing for respiratory tract irritation can be carried out in rodents by measuring their
respiratory frequency and respiratory volume. A substance irritating to the respiratory
tract will reduce the respiratory frequency and/or respiratory volume of-the animals.
It is thus possible to test a variety of irritating effects in laboratory animals, but in
general there is a lack of data, as the tests have only been performed on a limited
number of chemicals, or results of the tests have not been published in open literature.
Irritation is the effect most often observed at the lowest concentration. As chemicals
in general can be smelled before they are found irri.ating by humans, smell is a
warning sign of the presence of chemical. For most chemicals the occupational
exposure limits are based on no-effect level for irritation.
3.4.3 Sensitisation
The ability of chemicals to cause sensitisation can be tested in a variety of laboratory
tests. The best known is the "Guinea Pig Maximisation Test", which uses dermal
exposure, but further tests are described in OECD Guideline 406. The methods
K-16
-------
include a sensitisation with the chemical followed by later challenge to the chemical.
If the animal reacts to the challenge dose, it is said to be sensitised.
Hypersensitivity in an individual is seen as a violent reaction on contact with small
doses of the chemical, which is without such an effect in other individuals. If the
immune system is involved the effects are called allergenic, but general toxic
hypersensitivity is also known.
Quite a lot of information on sensitisation and hypersensitivity comes from human
evidence, ie when people have reacted violently to a chemical or a natural agent such
as fungal spores or animal proteins; The effects may involve the sldn (eczema) or the
respiratory tract (rhinitis, asthma, or allergic pneumonia). The chemicals causing
allergenic reactions are mostly known to produce a skin reaction, but some chemicals
are known to result in allergenic reactions of the respiratory tract.
For chemicals causing allergy there is no threshold. This means that once an
individual has been sensitised to a chemical, subsequent exposures to only very small
amounts of the chemical may induce a violent allergenic reaction. As there is no cure
for allergenic diseases, this may lead to change of occupation or retirement, if the
exposure is occupational.
The criteria for EEC labelling with the R-phrases 42 and 43 are semiquantitative in
as much as labelling of a chemical with these phrases can be carried out, if human
evidence have shown that persons exposed to the chemical have a. high frequency of
allergy.
3.4.4 Organ toxicity
Most information on organ toxicity is derived from short-term (28 or 90 days) or
long-term (1-2 years) studies in laboratory animals. The doses used in these studies
are always lower than those used in acute studies. The OECD has published guidelines
(nos. 407-413, 452,453) for the conduct of these types of study and, depending on the
duration of the study, different numbers of organs are studied histopathologically at
the end of the study. In general, organs such as lungs, liver, kidneys, heart, and
spleen are included in all these studies. The aim of studies on organ toxicity is to de-
termine the possible target organ for the chemical. The target organ is the organ "being
affected at the lower dose. Once the target organ has been established one might
conduct further studies to establish the detailed effects and mechanism of action for
the chemical, and one can establish the no-effect level (NOEL), if one or more of the
lowest doses are without effect.
As there may be differences in metabolism and sensitivity to chemicals between
laboratory animals and man, it is usual to include safety factors when relating the
observations in laboratory animals to possible effects in man.
The final evidence for an effect of a chemical in man, however, comes from
observations in man, so if a chemical has shown an effect on an organ in laboratory
animals, then studies on intoxications or on workers occupalionally exposed to the
K-17
-------
chemical can be conducted in order to confirm if the effect occurs under normal
exposure conditions.
Prior to the introduction of animal testing for regulatory purposes most evidence on
toxicity of chemicals were derived from human experience, ie workers occupationally
exposed to hazardous chemicals.
3.4.5 Genotoxicity
Agents that affects the DNA or the chromosomes are considered genotoxic. A wide
variety of bacterial tests systems has been developed to test if chemicals can induce
changes in DNA, i.e. they are mutagenic. The most commonly used of the bacterial
tests for mutagenicity is the Ames test using Salmonella typhimurium. Investigations
of chromosome damage involves either in vivo dosing or the use of in vitro cultivation
of mammalian cell cultures, followed by a suitable staining and or scoring procedure.
Such test procedures may detect chromosome aberration, sister chromatid exchanges
or unscheduled DNA synthesis. The OECD Guidelines 471-485 describe these various
genotoxicity tests.
In general genotoxic agents are very likely to be carcinogenic, therefore chemical
companies use the bacterial mutagenicity tests to screen new chemicals. Most often
the presence of mutagenic activity in a bacterial test will stop further development of
a new chemical. -
3.4.6 Cardnogenicity
A normal cell has to undergo a number of changes before it becomes a malignous
cancer cell. The process can be divided into initiation and promotion. The process of
initiation usually involves changes in the DNA, and is thus often caused by mutagenic
substances. The process of promotion does not involve changes of the DNA, and
therefore an additional number of substances can act as promoters of the carcinogenic
process. Some substances can act as both initiator and promoter, these are called
complete carcinogens.
In animals, testing for carcinogenicity, eg according to OECD Guideines 45-1, 453,
is among the most expensive tests as such a study normally has a duration of two
years and involves quite a large number of experimental animals. The rationale behind
conducting animal carcinogenicity studies is that cancer is considered as one of the
most serious effects, due to its severity and poor prospects of cure. Therefore there
is a general interest in preventing exposure to carcinogenic substances in order to
reduce the risks of developing cancer.
It is generally accepted that for most carcinogenic substances there is no threshold
value i.e. exposure to even low concentrations of carcinogenic substances might lead
to the development of cancer. For some carcinogenic substances (promoters), such as
hormones and substances affecting balance of hormones, there might be a threshold.
Whether or not a threshold exists has to be clarified for the chemical of concern. For
the purpose of screening in relation to life cycle assessment it appears to be
K-18
-------
appropriate to regard all substances, where positive carcinogenicity data exist, as
carcinogenic even if a threshold may exist.
Evaluation of chemicals for carcinogenic effects is carried out by a number of
agencies including the International Agency for Research on Cancer (IARC). This
agency evaluates the available epidemiological and animal data on carcinogenicity of
selected possible carcinogens and places the chemicals into groups 1, 2A, 2B or 3
according to, the evidence available. IARC has a further group 4 (non-carcinogens),
this group includes a very limited number of chemicals. In all IARC has evalutaed
around 200 chemicals, most of which had been suspeceted to have carcinogenic
efffects. For by far most of the industrial chemicals no information on carcinogenicity
exist, wherefore it is in table 1 conservatively considered to rejgarcl them as suspected
carcinogens and place them in category 2. The group 2B of IARC has been divided
into those marked with R45 (can cause cancer), which are placed in category 3, and
those not marked with R45, which are placed in category 2.
3.4.7 Neurotoxlcity
Chemicals, which affect the nervous system, are regarded neurotoxic. There may,
however, be a difference between the part of the nervous, system, which is affected.'
Some chemicals, eg. 2,5-hexanedione and n-hexane, have their most serious effect on
the peripheral nervous system (PNS), whereas most of the other organic solvents and
the heavy,metals manganese and lead have their main effect on the central nervous
system (CNS).
Acute, high exposure to neurbtoxicants often results in reversible effects, whereas
long-term, medium exposure may lead to irreversible effects. The irreversible effects
are due to the fact that nerve cells do not multiply, and thus once the cell is dead, it
is lost. It often takes a long time before the irreversible effects show up, as the
nervous system has an overcapacity of cells, that is, the function of a dead cell will
be taken over by other cells present.
There are no detailed guidelines for general studies on neurotoxicants, and no
quantitative criteria exist for their classification.
3.4.8 Reproductive toxicity
A chemical, which specifically affects the fertility or reproductive outcome, is
considered a reproductive toxicant. The effects seen can be divided into 1) effects' on
fertility and 2) effects on pregnancy and the foetus. Effects on fertility includes
damage to the sperm cells or ovum, and disturbances of the menstrual cycle in
women. Effects on pregnancy and the foetus include spontaneous abortion, congenital
malformations, low birth weight, and premature birth.
A number of chemicals are known to produce effects on either fertility or pregnancy
The most well known reproductive toxicants are.the teratogens, causing malfor-
mations. However, of at least equal importance are the chemicals causing low birth
weight and/or premature birth, as studies have shown that individuals with a low birth
K-19
-------
weight might lag behind in growth. Also important are the
on fertility ie the capacity to fertilise and obtain pregnancy. Some chemicals have
c^use^nfertiUty orTJduced semen quality in men occupationally exposed without any
o^siS^ffiS^^™ there ^ numer°US deSCripti°nS °f diStUfb ^
cycles due to an occupational exposure to chemicals.
most information on reproductive toxicants have been obtained from
^-SS are generally accepted guidelines for conducting studies on
rnd foetal toxicity in laboratory animals. However when going into more
no guidelines exist for studies on menstrual cycle disturbances and effects on
semen quality.
Several chemicals have shown an effect on the foetus in the
weight, but simultaneously there have been a general effect on the mothe 'i
of sL ficantly reduced body weight as a result of dosing with high Concentrations of
l Itappears that if high enough doses are used, nearly every chemical will
eec oSmother and the tetu. in the form of reduced weight At such
a chemical can not be regarded as a reproductive toxicant, because the effect
fefoetus is secondary to the general toxicity induced in the mother.
Three OECD Guidelines (nos. 414-416) describe how to study effects on fertility and
teratogenicity, but no quantitative criteria for classification of reproductive toxicants
have been published by OECD or by regulatory agencies.
4. Relation between exposure and effect
The exponential relation between exposure and effect is combined in the following
way:
Exposure
••—
Effect
Low
Medium
Low
Medium
High
One of the major questions that arise in connection with this system is what is
acceptable? According to legislation even the rating "8" W*»<"^»«?
medium exposure criteria are set well below regulative standards. However it is
TuSested that all "8" and "16"-ratings are assessed closely with respect to health
mpacs. This assessment should preferably be performed on the basis of actual
exposure data in a given process. Only on this basis will it be possible to estimate the
K-20
-------
actual local and work environmental health effects. For the regional and global effects,
consideration must be given to other, concomitant emission sources.
In this system it is assumed that the health impacts connected with energy production
are not taken into consideration. These impacts are to a large degree reflected in the
figure for energy consumption, although it is realised that by doing so no attention is
given to the energy source. The health evaluation will, on the basis of this, only
reflect those situations that are associated with the specific material.
The suggested scores might be added, yielding an overall health score. However, it
seems more important that it is possible to present information on the situations that
causes the actual scores. An optional possibility might be to be able to "peel" off the
situations one by one so that the second worst situation in a given exposure area will
be presented next and so on. By doing so, a constructor or designer may be able to
choose between raw materials in order to obtain situations with as little health impact
as possible. This option calls for the design of a database with a relatively large
amount of information. It should however be possible to gradually build this database
in order to meet the needs of the users.
5.
Thorough evaluation of selected situations
The intention of the screening procedure is to choose the most important situations of
the life cycle for a closer assessment of the potential health impacts. The reason for
this is that a full lexicological evaluation of all the situations will be extremely time
consuming and impossible to include in the scope of most LCA's. However, the most
important situations should be assessed more closely. In the following such a
methodology is presented. The method is based on some of the ideas discussed at the
SETAC workshop in Sandestin, Florida in February 1992. The scientific basis for
these ideas are to a great extent developed and used by the US EPA, but it is the first
time that these ideas are. used in the context of LCAs.
5.1 Reference dose (RfD)
A reference dose is an estimate of a daily exposure to the human population (including
sensitive subgroups) that is not likely to present an appreciable; risk of deleterious
effects during 70 years of exposure. The RfD approach is developed by a work group
under the US EPA15 and is based on the assumption that an exposure level can be
identified below which no adverse effects will result. For situations where dosimetry
or pharmacokinetic data are available to adjust exposure levels between experimental
animals and humans, the NOAEL (No Observable Adverse Effect Level) or LOAEL
(Lowest Observable Adverse Effect Level) is selected after these dose adjustments are
performed.
The RfD is derived from the NOAEL through the application of uncertainty factors
(UF) that reflect various recognized uncertainties in extrapolating from animal to
human studies. An additional modifying factor (MF) is also applied to reflect the
professional judgement of the entire data available on the specific: chemical
K-21
-------
The reference dose is derived from the NOAEL as follows:
PfD =
NOAEL
UFxMF
general, a 10-fold uncertainty factor is selected as die default value when
^^SA^^SSK
variation in sensitivity among humans.
An additional 10-fold factor is usually applied when extrapolating from valid I results
of lone-term animal studies when results of human exposure are not availab e or are
?nade?uaS ms fector is intended to account for the uncertainty in extrapolatmg from
animal data to humans.
Finally, an additional factor may be selected when there are '>^^J™*£™
dakThls factor is intended to account for the uncertainty m extrapolatmg from short-
term to long-term NOAELs. This factor is usually 10.
Moreover, additional factors may be selected when deriving a RfD from a LOAEL
tasSdrf a NOAEL. These factors are intended to account for the uncertainty m
extrapolating from LOAELs to NOAELs.
The modifying factors (MF) which usually are less than 10 are used to indicate the
of the scientific uncertainties of the studies and the data base
es ug
explicitly trLed above, e.g., the completeness of the overall data base and the
number of species tested. The default value for the MF is 1.
In addition to human and animal studies, cell culture tests and structure-activity data
can be used as supporting information when estimating acceptable levels of human
e. Therefore validated short-term predictive tests (in vtvo and m yam) should
included in the evaluation. These may include predictive tests for car
may evaluate the following endpomts, point mutations, e.g. , Ames
*lffects, e.g., SCE (Sister Chromatid Exchange), and other
endpoints such as cell transformation.
5.2 Characterisation of the exposure
To characterize the potential exposure to the chemical, the objective is to gathei -and
critically review the demographics and activity patterns to determine the total inttte
the potentially exposed population may receive by direct and indirect exposure from
multiple exposure pathways.
K-22
-------
Exposure data may be obtained by a combination of the following methods:
*
*
Direct measurement by analytical instruments which is the preferred method,
indirect measurement, which uses mass balances,
* models, which are mathematically derived, and
* expert evaluation/statements.
If it is impossible to make an exposure estimation, threshold limit values and/or
quality standards should be used as worst-case exposure indicators. Under all
circumstances the inherent uncertainties and assumptions used for monitoring data or
model assumptions should be clearly defined and characterized.
In general, three exposure pathways should be considered, i.e., inhalation, dermal,
and ingestion, although the latter one seldom is relevant in occupational exposure. The
potentially exposed population should also be characterized by demographics, e.g.,
age, sex, and health status as well as activity patterns.
For all potential exposure pathways, population/site specific exposure data should be
used to characterize potential exposure. Standard default assumptions may be used in
the absence of population/site specific data, e.g., average drinking water consumption
of 2 liters/day, average inhalation rate of 20 nWday, and soil intake is 0.2 g/day for
a 10 kg infant. The duration of exposure should be characterized as acute (single
exposure), short-term (only a limited fraction of lifetime), or long-term (a major
portion of lifetime). In addition, the frequency of exposure should also be taken into
account and characterized as continuous (daily) or intermittent (less than daily).
Subsequent to the exposure characterization, the total chemical intake for each specific
exposure pathway can be estimated. The exposure intake from each of the specific
pathways are then summed to estimate the total chemical intake. Usually, only the
applied dose, i.e. the exposure at the point of contact, is available to calculate the
intake of the chemical but if possible, the absorbed dose should be used. The total
intake of a chemical can be calculated with the following equation:
C x CRx EFD 1
BW
AT
where
I = intake,
C = average concentration over exposure period,
CR = amount of chemical in media,
EFD = exposure frequency and duration,
BW = body weight over exposure period, and
AT = averaging time, period over which exposure is aiveraged.
K-23
-------
5.3 Combining Reference Dose and exposure
The first step when combining the reference dose and the total intake of a chemical
is of course to make them comparable in terms of time, i.e., the two factors must be,
expressed in the same time horizon (days, life-time). Having done this, it is possible
to estimate the impact by dividing the total intake with the reference dose:
Impact -
Total intake
Reference dose
The total contribution of several chemicals (several "situations" in the life cycle) to
a given effect can then be summed. However, there will still be a number of factors
to consider in doing so.
Firstly, potential cancer effects from exposure to carcinogens without an effect
threshold cannot be assessed by the reference dose model. Therefore, another way of
assessing the risk of cancer must be employed and no matter which is being used the
results will not be comparable.
Secondly, the critical effects on which the reference dose is based are not equally
serious, neither by expert opinion nor by perception of the general population. Some
factor should be applied to account for the severity of the effect of concern.
Thirdly, the magnitude of the exposed population should be taken into consideration
when adding the impacts. It is obvious that emission of a potential toxicant into
density populated areas is more serious than when only few people are exposed to the
same toxicant. However, no guidelines on how to weigh these factors can be given on
a scientific basis. Indications on how non-fatal effects Can be weighed ban be found
in guidelined for workmen's compensation, but if mortality impacts dominate in an
LCA, these are of little utility! Ryding and Steen10 have in the EPS-system given an
example on how they rate the above mentioned factors (and others), but no,
explanation of how these factors are derived is given.
A final calculation of the total impact cannot be justified scientifically today. A rough
approximate could be achieved by summing the single (dimensionless) impacts into
one overall impact score. Interpretation of this figure will however demand that all
assumptions and weighing factors are well characterized and explicitly described.
There will thus be a need for further development of health assessment in the context
of life cycle assessment. Major problems will be aggregation of many contributions
to a given effect, i.e. a potency ranking of chemicals with respect to a given effect,
and weighing and aggregating different effect categories. The framework outlined
above may prove to be a valuable starting point for the discussions necessary to
continue the work already going on in many LCA projects of both theoretical and
practical nature.
K-24
-------
5.4 Conclusion
The screening procedure described in Chapter 3 aims at pointing out the most critical
situations in the life cycle. The necessity of such a screening procedure becomes
obvious when a subsequent detailed assessment of these situations is performed. In the
procedure for such an assessment, which is outlined in Chapter 5, many resources are
used for assessment of just one situation. If hundreds, or maybe: even thousands, of
situations in the whole life-cycle should be assessed this would probably rriean that the
assessment would be of very limited value to the end-user, simply because it has taken
so long time to perform.
The screening procedure may also be helpful by being a tool for a sensitivity analysis.
The thorough assessment should of course start with the most critical situations and
should be continued until no significant impacts are found and added to the total
impact. As a minimum requirement the situations rated "16" and "8" by the screening
procedure should be assessed. Also several (preferably all) of the situations rated "4"
by the screening procedure should be assessed j but if none of these contributes to the
overall assessment in a significant way, it should not be necessary to assess the rest
of the situations.
It seems possible to describe and maybe even quantitate both the poten':al and actual
effects of, a given situation in the life-cycle rather precisely if the data base is
adequate. Some iteration between the description of effects and collection of data on
actual exposure and complementary lexicological information will however be
necessary in many LCAs as experience shows that only limited data sets are available
for most chemicals and "situations". ' ,,
A scientific aggregation of health effects is not possible, but a subjective weighing
may be justified, for some LCA applications, e.g., internal compiany priority setting.
When performing a subjective weighing, the implications of the subjectivity should be
characterized as well as possible. By; doing this, other users of the data sets in the
assessment have the possibility of assigning Other weight factors and maybe reach a
different result in another context. Transparency is thus one of the key issues, also in
the health assessment; • ' "
K-25
-------
6. References
1. BUS. Okobilanzen von Packstoffen. Schriftenreihe Umweltsschutz 24. Heraus-
gegeben von Bundesamt fur Umweltsschutz. Bern, 1984.
2. BUWAL. Oekobilanz von Packstoffen Stand 1990. Schriftenreihe Umwelt Nr. 132.
Herausgegeben vom Bundesamt fur Umwelt, Wald und Landschaft. Bern, 1991.
3 BUWAL. Methodik fur Oekobilanzen auf der Basis okologischer Optimierung.
Schriftenreihe Umwelt Nr. 133. Herausgegeben vom Bundesamt fur Umwelt, Wald
und Landschaft. Bern, 1990.
4. Luftreinhalte-Verordnung. Stand am 1. Oktober 1987. Umweltschuthgesetz,
Switzerland, 1987.
5. Verein Deutscher Ingenieure. Maximale Immisions-Werte. VDI 2306 (March 1966)
and VDI 2310 (September 1974).
6. Verordnung uber Abwassereinleitungen. Stand am 1. Januar 1988. Gewasser-
schutzgesetz, Switzerland, 1988.
7. Braunschweg A. The new ecological valuation method based on "ecological
scarcity" and its application in Switzerland. Presented at the SETAC-workshop on
Environmental Life-Cycle Analysis of Products, Dec. 2-3, 1991, Leiden, Holland.
8. Christiansen K, Grove A, Hansen LE, Hoffmann L, Jensen AA, Pommer P,
Schmidt A. Environmental Assessment of PVC and Selected Alternative Materials.
Miljastyrelsen, Copenhagen. Danish version 1990, English version in press.
9. Christiansen K, Grove A, Hansen LE, Hoffmann L, Jensen AA, Pommer P,
Schmidt A. LCA of PVC and selected alternatives. Integrated Environmental
Managements: 21-23, 1991.
10. Ryding S-O, Steen B. The EPS-system. A PC-based system for development and
application of environmental priority strategies in product design - from cradle to
grave. Institut for Vatten- och Luftvardsforskning. Stockholm, 1991.
11. Schmidt A, Christiansen K, Jensen AA, Nielsen K, Lange M, Andersen O,
Grandjean P, Rasmussen B, Sortkjaer O, L0kkegaard K. Integreret milj0- og
arbejdsmilj0vurdering af nye materialer. Dansk Teknologisk Institut, Milj0teknik,
1990.
12. Jensen AH, Winge U, Broberg O. Vejledning til forskere, udviklere og kon-
strukt0rer i milj0- og arbejdsmilj0venligt materialevalg. Institute for Product
Development, DTK, 1991.
K-26
-------
13. Kaergaard H. Environmental risk management - trends in knowledge and techno-
logy. Paper presented af Risk Management Forum AWAI/RIMS Conference, October
13-16, 1991, Monte Carlo.
14. Gj0s N, M011er M, Hasgh GS, Kolset K. Existing chemicals: Systematic data
collection and handling for priority setting. Nordic Council of Ministers. NORD 1989:
87.
15. Barnes DG, Dourson M. Reference Dose (RfD): Description and use in health risk
assessment. Regulatory Toxicology and Pharmacology 8: 471-486; 1988.
K-27
-------
Annex 1.
Data sources
Many data sources are available when assessing possible health impacts. Handbooks,
criteria documents and scientific papers on specific issues have traditionally been used
for documentation. With the rapid development in toxicological research this kind of
documentation will tend to be insufficient within a relatively short period and hence
of limited value.
During the last years on-line search in central databases or databases on CD-ROM has
caused a great progress in documentation. The number of accessible databases is
rapidly increasing but this raises another problem, namely the costs of searching these
sources. Subscription to one of the databases on CD-ROM costs 10 - 15,000 DKr
annually so a careful choice must be made.
For the outlined screening procedure the documentation could be based on RTECS
(Registry of Toxic Effects of Chemical Substances) which is a database issued by the
U.S. National Institute of Occupational Health and Safety. It is available on CD-ROM
for a reasonable price and is updated four times a year. The amount of information
is comprehensive, but no evaluation of the data has been performed. Another source
of information is ECDIN (Environmental Chemicals Data and Information Network),
a database issued and maintained by the CEC. Sixty thousand chemicals are
registered in this database, but for the majority of chemicals only sparse information
can be found. '' ? -.-'"-;, ' • ; • •• -'•• •
Official lists (National Work Inspectorates and Environmental Protection Agencies)
holds information on legal implications of chemical substances and should be used as
basic documentation e.g. on chemicals that are considered to be carcinogenic. Besides
the Nordic lists also German,(MAK-Werte) and American (American .Conference of
Governmental and Industrial Hygienists, ACGIH) documentation of Threshold Limit
Values may be useful. For the assessment of carcinogenic properties valuable
information can be found in the monographs from IARC (International Agency for
Research on Cancer) (WHO, Lyon).; , *'• „ '-. •>, ' ,,f « , ^ ',:>•• ;'' '• •
Other written sources are issued by various expert groUpS:rNordiska Expertgruppen
for Gransvardedokumentation; issue a series of documents containing toxicology
reviews for the setting of an Occupational Exposure Limit. The IPCS (International
Programme on Chemical Safety (WHOS Geneva)) series on Environmental Health
Criteria also contains evaluation on anumber of chemicals. Ira this series, the possible
effect on the general population has also beeti evaluated, ;B0th of these series
comprises more than 100 chemical substances., The'International' Register, of
Potentially Toxic Chemicals (IRPTC) under the United Nations Environmental
Programme collects regulations and recommendations related to control of a substance
in various media in the IRPTC Legal Files. ;•• . •*•> •
The American Agency for Toxic Substances and Disease Registry (ATSDR) under the
U.S. Department of Health and Human Services issues a series of toxicological
profiles of toxic substances which may be useful when making a hazard assessment
and eventually a risk assessment. The same department also issues a- series of
K-28
-------
toxicology and carcinogenesis studies from the National Toxicology Programme
(NTP). Both of these series are available from National Technical Information Service
(NTIS). ._•'. •..'••' ' '" ' •- '• _ ' ':-'•' • .';'•
The U.S. Environmental Protection Agency (USEPA) has made IRIS, a data system
which contains summaries of health risk and EPA regulatory information on about 500
chemicals. The core of the system is a collection of computer files that contain
descriptive and quantitative information to be used in setting the reference dose RfD.
The data are publicly available as an electronic data base, but also paper versions of
the documents are available on request. The US EPA is currently making a further
development of IRIS. This development aims at making risk assessment of specific
situations, i.e., situations with information oh number of exposed people, site of
exposure, concentration, duration etc. This data base, allegedly to be called "RISK"
at the moment only contain information on a limited number of chemicals.
A number of handbooks may also be used in the screening procedure. Sax & Lewis:
Dangerous properties of Industrial Chemicals (Van Nostrand Reinhold, New York,
1989) gives an overview of possible health effects,1 but the documentation is rather
limited. Also Patty's Industrial Hygiene and Toxicology (Johm Wiley & Sons, New
York, 1982) has information on a large number of chemical!;. The latest edition is
however out-dated and should be used with caution.
Working with the screening procedure will probably point to other sources that may
be used. Dutch and Swedish criteria documents are being constantly issued and may
provide information on chemicals other than those normally seen.
K-29
-------
-------
APPENDIX L
COMPUTERIZED DATA AS A TOOL FOR SUBSTITUTION ANALYSIS
-------
-------
Computerized data as a tool for substitution analysis;.
j. Niemela, Danish Environmental Protection Agency
L. Seedorff, COWlconsult, Denmark.
Introduction
The Danish Environmental Protection Agency (DBPA) is often confronted with
<|uestions about the possibilities to substitute hazardous chemical* by !•••
hazardous one». To help meet this need, DSPA has initiated a project
concerning the development of a computerized model for substitution analysis.
The model has been deaigned for systemically screening of onvironmental and
human health data in order to identify - from a large number of chemicals -
the moat hazardous ones.
The basis of the computer model ia data from the Danish Product Register Data
Base (PROBAS). The Danish Product Register i» a governmental data base for
information concerning substances, materials and products us«d in Denmark. The
development of PROBAS started in 1980. The data base wa« entablisheti by the
Ministry of the Environment and the Ministry of Labour. Ais of May 1993, PROHAS
contained information on about 67,000 chemical product*. Information on the
composition of about 48,000 products is computerized. Data on the products
are submitted by Danish manufactures, importers, agents and their
international suppliers according to notification rules, surveys and research
projects and from ether relevant sources. Notification rule* cover hazardous
products, classified as such according to BBC rules, or by special order from
the DBPA and the Danish Labour Inspection, {ref. 1.)
As an example of how the Product Register can be u«ed, we present the
preliminary results of a pilot study on cutting fluids. The project is still
open to further development>
'The data-sources • • '
391 cutting fluids are registered in PROBAS. Thii cutting fluids
contain a total of 431 different substances.
Based on the registered products a data file is prepared containing
CAS-RN, chemical names, percent-content (highest iind lowest) and codes
for the technical functions .of all chemical constituents.
The data file is transformed to a database in DEPA (dBase IV
format).
The following data are automatically transferred to the DSPA-base via the
CAS-RNi
toxicologieal data (from, for example, RTECS and lAltC)
risk phrases (according to EEC rules)
data on aquatic .toxieity (from "Katalog wassergiefahrdenden stoffe",
BRD)
- data on biodegradation, including the Japanese M£TI database.
L-l
-------
physical/chemical data as LogP, vapour pressure etc. (measured ,or
calculated by tha U.S. EPA QSAR program)
Selection criteria
After running the DEPA-base, where (data are added to the individual sub-
stances, the following criteria are used to select the most hazardous ones:
Carcinogens or suspected carcinogens
substances with risk phrases for being very toxic, toxic, allergenic,
carcinogenic, teratogenic, mutagenic, or causing cumulative or
irreversible effects
substances with moderate or serious aquatic toxicity
non-biodegradable substances
substances with high predicted aquatic effects, for example, chemicals
With toP > 3-
Results
The 391 cutting fluids contained 431 different substances with CAS-RN.
Of these, 90 substances were selected as the most hazardous by using the
selection criteria. In table 1 the 90 substances are subdivided 'after their
function in the cutting fluids.
Table 1. Technical functions of the selected substances.
Technical function
Lubricity agents
Bioeides
Smulsifiera
Corrosion inhibitors
Organic solvents
Extreme pressure agents
Antioxidants
Pigments
coupling agents
Deodorants
Propellant
Impur it ies /monomers
Number of
substances
29
16
11
7
4
3
3
3
2
1
1
10
L-2
-------
Alternatives for the unwanted substances are searched by looking into the
group of non-selected substances with the same function cod*.
Lubricity agents:
As examples of selected lubricity agents pan be mentioned 13 petroleum oils
(carcinogenic effect) and one vegetable I oil: tall oil (aquatic toxicity,
allergenic).
By substitution of paraffinic oil or other low aromatic and FAH-free oils for
the petroleum oils the cancer risk can be reduced or eliminated. A suitable
alternative for tall oil is soya oil (low aquatic toxisity, no allergenic
effect).
Biocides:
The selected biocides - primarily selected because of aquatic toxicity and
allergenic effects - represents nine phenolic derivatives and four
formaldehyde releasers.
No suitable alternatives were found among the non-selected Substances in this
group. In general, the best solution concerning the biocides in cutting
fluids, is - if possible - to avoid them. This can among oth»r things be done
by using physical methods such as centrifugation, filtration, pasteurisation,
ultra sound or radiation (UV).
Emulsffiers:
Among the selected emulsifiers we find a nonylphenoi derivative (non-
biodegradable, log p > 3). This can be replaced with «.g. an alcohol
ethoxylate (biodegradable).
Corrosion inhibitors:
One of the selected inhibitors is sodium nitrite (aquatic «md human toxicity).
This and other selected inhibitors (amines, benzotriazol«s) can possibly be
replaced by ethanol amines (low aquatic toxicity).
Extreme pressure agents:
Here we find three chlorinated paraffins (aquatic toxicity, bioacumulative and
possible carcinogenicity). Possible alternatives could b« sulphur powder or
sulphonated vegetable oils.
Pigments and deodorants:
The three pigments are selected because of aquatic effects, the deodorant -
pine oil - because of the same and because of allerganic effect.
Figments and deodorants should be eliminated from the cutting fluids as they
only are used of aesthetic reasons.
Coupling agents:
The two selected coupling agents are an EDTA sodium salt (acpiatic toxicity)
and nitrilotriacetic acid (non-biodegradable). Perhaps these substances can
be substituted by tartarlc acid or oxalic acid.
conclusion
The mentioned substitution possibilities are an example of the outcome of
L-3
-------
u.ing the computeritcd modal to screen a large number of chemicals with
readily available information in a minimal ampunt of time.
Subjective evaluation of the outcome is nevertheless necessary for analysis
and evaluation of chemical hazard and technical utility.
The greatest'limitation in using the model is the lack of available data on
which to base effects assessments. In this pilot study there was no data on
41 percent of the 431 chemical constituents in the cutting fluids.
Continuing development of the model to include additional data sources is
therefore necessary to further enhance its value as a tool for substitution
analysis.
Three additional studies are under preparation. These cover automobile under-
carriage agents, sealants and fillers, and chalk- and rust-removers. When all
of these pilot studies are complete, it is intended to establish a dialogue
with relevant industrial organizations, with the aim of enhancing the
substitution process.
References*
1. PROBASt The Danish Product Register Data Base - a national register of
chemical substances and products, Journal of Hazardous Materials. 30 (1992),
pp. S9-69. [included]
2. Working Report, DEPA, No. 5, part 1., Cutting Fluid* (available in Danish).
3. Working Report, DEPA, No. 7, part 2., Automobile Undercarriage Agents, (in
press, Danish). •',..'
4. Working Report, DSPA, No. xx, part 3., Chalk and Rust Removers, (under
final preparation, Danish).
4. working Report, DSPA, No. xx, part 4., Sealara and Fillers, (under
preparation, Danish).
L-4
-------
APPENDIX M
PROBAS: THE DANISH PRODUCT REGISTER DATA BASE—
A NATIONAL REGISTER OF CHEMICAL SUBSTANCES AND PRODUCTS
-------
-------
Journal of Hazardous MateriaU, 30 (1992) 59-69
Els«vj«r Science Publishers B.V., [Amsterdam
59
PROBAS: The Danish Product Register Data Basse — A
national register of chemical substances and products
Mari-Ann Flyvhoim, Poul Andersen, Inge D. Beck and Nanna P. Brandorff
The Product Register Department Danish National InstiKse of Cccipcsono/ Health, Lino
Purkalli 105, DK-2100 Coptnhagtn (Denmark)
(Received July 7,1991; accepwd August 5.1991)
Abstract
A description of the structure the content, and the purpose cf the Danish Product R«fist«
DataBase (PROBAS } is given. The PROBAS data baw is a national data b*s« started in 1930. Data
sources have been notifications of hazardous chemical products and data from investigations o=
occurrence or use of chemicals nude by. for .example, the Danish National Institute of Occupa-
tional Health. Examples of the content of ths register ar« presentee, luch as tna mem frequently
registered product categories, or the most common substances. As of January 1991, PROJIAS con-
tained information on about 53.300 chemical products. Information a.-, the composition of about
30.000 chemical products was computerizsd. Th« possibilities and limitations of this use of a na-
tional data base on chemical products are discussed
1. Introduction
Information systems on chemicals have become important tools for moni-
toring and controlling emission from and occupational exposure to <±eBiicals.
and for preventing adverse.effects on environment and health caused by chem-
icals. From the very few product data bases described in literature, it seems as
if data bases are often established for selected product categories or industrial
areas of use, where data on products are collected as part of investigations on
occupational or environmental hazards [1,2]. A reason for the very few pub-
lications on product data bases could be the difficult task of obtaining data on
product ingredients caused by the confidentiality of composition of commer-
cial products [3], or the explanation could be that such databases mainly serve
administrative or poison control centre purposes [4], making them inappro-
priate for scientific investigations and publications. Most of the few chemical
and occupational data bases reported hi the literature are on hazardous (pure)
substances like RTECS, on material safety data sheets, or on selected groups of
commercial products, e.g. cosmetics or Pharmaceuticals. Nation-wide surveys
like National Occupational Hazard Survey (NOHS) and National, Occupa-
Comspondence tc: Dr. M.-A. Fiyvhobn. The Product Register -Apartment. Danisih National
Institute'of Occupational Health, LersB Parkalle lOo. DK-2100 Cop«:ihagen (Denmark).
0204-3894/92/305.00 £ 1992 Els«vier Science Publishers B.V. All rights rw«rved
M-l
-------
Bonal Exposure Survey (NOES) [5] have provided data base* on trade name
products, their composition and -potential exposure to the ingredient sub-
stances, and a model for the identification of high risk occupational groups by
linkage with RTECS data [6J. In Sweden and Norway national product infor-
mation systems like the Danish Product Register have been developed (7,8).
In this paper we describe the structure, the purpose, and the content of the
Danish Product Register Data Base (PROBAS), and discuss the experiences
with and the use of a national data base on chemical substances and products.
1.1 The Danish Product Register Data Base -
The Danish Product Register is a governmental data base for information
and evaluations concerning substances, materials and products used in Den-
mark. The development^pprtPft? j^rtPf1;" 1Qftrt T>"> 4atfl frf«* »«« estab-
lished by thefMinistry of the Environment and the Ministry ofLabour^wfaich
administer parts of the Danish legislation concerning chemicals. The"Product
Register is located at the Danish National Institute of Occupational Health
(DNIOH), which is pan of the Danish Labour Inspection.
The purpose of PROBAS is to. collect and utilize information on the use and
the adverse effects on health and environment of substances and products used
in Denmark. ; - •".."• ; "• -..'''.
2.Methods . '''> ••' •'• • •• -• ,•" •- .' •
: ' / - '' t ",'*'•' • ~ ' ' ' .
2.1 The data base structure
The technical configuration of PROBAS is a Local Area VAX-cluster. The
data base management system is ADABAS using inverted lists for data access
with applications written in NATURAL, a programming language designed for
AD ABAS. These tools provide practically unlimited searching and reporting fa-
cilities for the many different situations where information retrieval IB re-
quired. Data and conversion tables are stored in 30 files. An overview of the
main files is given in Fig. 1,
The data base has been designed with three interconnected subsystems, i.e.
fit. 1. An ovtrvitw of th« f »<" files in PROBAS and their inurconnictton.
M-2
-------
,-chunic^^^^
products aw preparations (m^ti^)^^^^^^1^^'^^
Intfae subsystems each er
3. Material
3.1 Data sources
M-3
-------
• lamens aideline documents, and nocifications of substances
ftcts, criteria *SSdSte «»«««• are names, CAS BN. other registry
user companies).
4. Results
TABLE!
Chemical composition
Product caujtoiy (field of application)
Industrial »rea of u»e
Physical and chemical properties
Toxicfcydau
Cl»»»ific«tion «nd labelling
Sautymeajurts
Transport classification
Product Rftgistration numbers
awisntd to notiflcn
30000
31900
23400
22700
13100
17100
16700
5000
11800
M4
-------
TABLE 2
Nimberof conputeiyed products divided into degr^of knowledge product composition.
ot January '.991 i ba*»neti i iKKfja ft.
Ccnavooi may
Hf*. tnnrj i vaatn)
Qtmtnant toaitat
Buaan. .-tatoi, h««;
WM M>W« t umn
Nunatrd craduai:
woo JOGS 4000 icao woo
11,800 of the notified products PR NOS were assigned to the notifying coiapanies
ifflplying that the Product Register has information on the complete compo-
aition of the products and that the notifying companies are obligjsd to report
alterations of the submitted information.
The information on product composition has been evaluated anil a code for
quality has been assigned. The numbers of fully computerized products (divided
into degrees of knowledge on product composition are given in Table 2. For
more than half of the products the complete composition was known. For 70%
of the products the information on composition was sufficient to evaluate the
tone effects of the product or the suggested classifications and safety measures.
The most frequently recorded industrial areas of use, based on 23,400 prod-
M-5
-------
M
vfth this information, January 1991
ucts registered with this information, January 1991, were the iron and rneud
Ssw chemical industry, and repair and personal services, followed by the
^S^S^^^f^V^^^^^S^^^
of *
counted for nearly the same number --
most frequently recorded product categories based on ol.900 products reps
SSSUormat^
product categories were cleaning agents and paints/lacquers which together
ces^^
including more than half a million names. The number of ^^ffl|°^f'
reprewntedby these records ^unwrtam bemuse Cheim<^Abst^t^rv^ce
igned generic inder name, and registry numbers to man y h°»°logou«
ounds, mixtures of isomers, natural products and chfferentfrac.
l. The number of CAS BN for which certain information cat-
TABLES •
130.000 CAS RN)
Information category
^^^BMH|^^B^M^^_^B«HIH^^H^
Classification «nd Ubillinf
Toxicitydaw
Transport clauiHcation
Phytlcaland chemical properties
Occupational eipotuit level
Safaiy meuurte
Number of CAS RN
' ~
5840
2379
2 122
2004
1544;
252
M-6
-------
65
TABLE4
Most frequently recorded subitanws in PROBAS as of Januarv 1991 b«.rf „„ k. . ,« *~
ucts rtgisttred with information on product composition ^ 3MO° prod-
CAdRN Number of
product*
7732-18-5 7878
1330-20-7 4 060
13463-67-7 2496
7631-86-9 2208
108-88-3 1 906
1309-37-1 1 618
106-89-8 1 582
1333-86-4 1 417
1675-54-3 1 412
123-86-4 1 377
67-43-0 1 375
50.00-0 1 368
14S07-96-6 1 241
25085-99-8 1191
471-34-1 1103
71-36-3 1098
, 57-S5-6 ,1093 '
111-15-9 1057 - ' , '
7727-43-7 ,1030
64-17-5"' 1016
Substanct aom« ' , . , ~
• " ' ' , ,- i
v^ ' " : '", •• — :
Xyi«n« ,
Titanium dioxide
Silica
Toluene
Iron oxide
-------
66
nroduct groups, and furthermore, the amount of information.giver, may also
SSSfor some registration systems. Moreover, the demand for submission
of updated information only includes products with PR No officially assigned
to rjotifiers. . .
Hhe degree of coverage varies according to product category- and ;raae pi use.
Estimates derived from investigations in manufacturing and service industries
[16],and in the wood and furniture industry [17,18] showed that PROBAS had
a coverage of 34-50% of products used in Danish workplaces Since certain
product categories were covered by special notification rules or nan oeen sub-
ject to special investigations, the information in PROBAS on such categories of
products is considered complete for the Danish market, e.g. epoxy and iao-
cvanate products, asbestos-containing products, pesticides, cleaning agents (&ii
by special order) and cutting fluids, which have been the subject o: a national
study. The Danish product registration has an overrepresentation of nazard-
ous products due to the notification rules, which means that the degree of cov-
erage for hazardous products is higher than the estimated 34-oCi;, oaaed on
both hazardous and non-classified products. The most frequently recorded
trades of use and the most frequently recorded product category reflect the
selection of products for inclusion in the data base, since the notmcation sys-
Save^^
8gThe'amount of information on chemical substances is very spare, mainly
because only limited resources have been allocated to the update of this part
of the register, but also due to the fact that only a minor fraction of substance*
has bin investigated for physical and tozicologica properties -The count,
listed in Table 3 can therefore be considered a* ambiguous, e.g. physical and
chemici parties can be anything- ranging fror* a single boiling point to
eSuve information on density, solubility, partition coefficients, vapour
pressure at many different temperatures, etc. .
The fact that certain product categories have been revered|» nfl-
reflected in the list of the most frequently recorded •*•»"••" p
Thus, epichlorohydrin, bisphenol-A-diglycidylether and its powers,
fillers and pigments are typical constituents of epoxy-based products.
SiaughSormation derived from ?ROBAS cannot provide complete esti-
mation of exposure, PROBAS constitute* a very useful system for surveillance
of exposure to chemicals in Denmark, providing qualitative and semquanti-
tative surveys On occurrence or amount of substances used, product categories,
industrial area of use, enterprises involved, etc. As the need for classification
of the thousands of chemicals in workplaces and environment has become ev-
ident, the limited resources for testing physical chemical and lexicological
properties are heavily dependent on priority settings. Lists of substances with
high exposure potentials derived from PROBAS and similar registers can be
important tools for the determination of these priorities.
M-8
-------
67
5.1 Examples of surveys and research based on PROBA* data
.
consumption of carcinogens in Denmark [19]
6. CondosioQ , -
Reference*
M-9
-------
68
2 M. JKarsudt and R. Bob«l. Ac«*s to data on chemical composition of products used in au-
tortpair and body shops. Aa.J.Ind.Med.. 6 (1984)339-372.
3 D S. Sundin. The national occupational hazard survey: a difficult quest for a reliable data
base. Oceup. Health Saf.. May/June (1978) 21-23. .
4 K.ErikWn.LBondessonandH.PerSSon,Compu«t^p^u«^or™t,or. ^pr.hmintrj
report from the Swedish poison information centre. Hum. ToxicoL, 2 U983) 279-280.
5 J.A. Seta. D.5. Sundin and D.H. Pedersen, National Occupational Exposure Survey. Field
Guideline DHHS (NIOSH) Publication. No. 88-016, U.S. Department of Health and Hu-
6 DH 5dS'J?R!a52 aad D.S. Sundin. A Model for the Identification of High Bisk
OccuSS Groups u,5ng RTECS and NOHS Data. DHHS (NIOSH) Technical lUpcrt.
PubuSnNo. 88-117. U A Departm.nt of Health and Hua*n Service*. Cuicumati. OH.
7 PrSuktkontrolL Kundjorelw om produktanmalan. SNFS 179: 2 PK (in Swedish). Statens
naturvardtvwksfbTfattningssamline. Sweden. 1979. .
8 DeklaMrtosav kjenisk* staffer ogprodukter. Veikdninj (in Norwegan), Produtaegistret.
9 Da^hMfaistTy of Labour, Order on the roister of Substance and^teriah. Order No.
466 of 14 September 1981, Danish Ministry of Labour, Copenhagen. 1984.
10 Danish MinLy of Labour, Order on Subrtanc** wd Material*. Ord»r No. 540 of 2 S«ptem-
b«r 1982, Danish Ministry of Labour, Coptnhagen, 1984.
15
16
19
20
21
5n? environment (in Danish; sununary in English). AMI-IUport No- 35/1931.
tionallnstitutt of Occupational Health. Copenhagen, 1991- . w«,«.i,i
K ft ThoM«n. AUeraens in the Working Environment (in Danish; susmary in English)
AMf SS !3 S, DaSh N^Sal Institute of Occupational HMO. Copenhagen,
l!sLn»tn and S.P. Lund. A strattfy for risks due to «po»ur« to neurotic chemicals, Am.
*^
ucts. Int. Arch. Occup. Environ. Haahfa.. (submitted).
J Holnanard. The industry uw 16 carcinogenic substances IB l»amcu«ir In: C. Falhne
tE?.)IwSEnvlronme«.SunimryofartklesPubli*h.din Aibejcsnil^. The Danish
M J??yvholm and T. Menne, Sensitizing risk of butylated hydroxytoluene based on exposure
and effect data. Contact Dermatitis, 23 (1990) 341-345.
M-10
-------
69
21
24
in
M-ll
-------
-------
-------
-------
-------
-------
IKD
*fc»-ltf Hft.fefeHE.'-frfc f»
„«,«», •ml^,,«:«,,.,«,^,.™
------- |