UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

                          REGION IV

                      345 COURTUAND STREET
                      ATLANTA, GEORGIA 30365
                The Use Of Risk Assessment
                in EPA Regional Operations
This report examines the nature and use of risk assessment
in U.S. EPA Regional operations.  The report was done to
clarify what the term "risk assessment" means in the context
of current Regional policies and procedures.  It focuses on
the use of risk assessment to estimate human health impacts
from toxic chemical exposures in site-specific situations.
It is based on a survey of regulatory programs at a "typical"
Regional office - Region IV, located in Atlanta, Georgia.
From the survey results, the step-by-step process involved in
site-specific risk assessment is defined.  The report also
presents recommendations for ensuring the scientific quality
and consistency of site-specific risk assessments.

Please forward your comments to:

                        Susan Deihl
                      Project Officer
             Office of Policy and Management
                       EPA-Region IV
                345 Courtland Street, N.E.
                 Atlanta, Georgia  30365
                       404/881-3776
                                               NOV 2 l 1985
Jack E. Ravan
Regional Administrator
Date

-------
                        -11-
                 ACKNOWLEDGEMENTS






We would like to thank the staff of EPA - Region IV



who gave their time, effort and valuable insights



to this project.  These contributions were instrumental



to the development of this work.  We hope that this



report serves to improve Agency understanding of



Regional risk assessment processes.

-------
                             Ill
                      List of Preparers
Project Director
Principal Investigator
Advisory Committee
Reviewers
Robert B. Howard
Chief, NEPA Compliance Section
Environmental "Assessment Branch
US EPA, Region IV

Susan M. Deihl
Risk Assessment Coordinator
NEPA Compliance Section
Environmental Assessment Branch
US EPA, Region IV

Dr. Frederick A. Rossini
Associate Professor, Technology
  and Science Policy Program
Georgia Institute of Technology

Dr. Alan L. Porter
Associate Professor, Industrial
  and Systems Engineering
Georgia Institute of Technology

Dr. Jay A. Weinstein
Professor, Technology and
  Science Policy Program
Georgia Institute of Technology

The Office of Research and
Development, U.S. EPA

Regions I - X, U.S. EPA

-------
                                                           IV
                      TABLE OF CONTENTS
                                                         Page

ACKNOWLEDGEMENTS	  i i

LIST OF PREPARERS	  iii

LIST OF TABLES	  vi

LIST OF ILLUSTRATIONS	  viii

GLOSSARY	  ix

SUMMARY	  xi

CHAPTER

I .    INTRODUCTION	 1

     Historical Background

     Terminology

     EPA and Risk Assessment

     The Study

II.  RISK ASSESSMENT IN REGIONAL OPERATIONS	 13

     Characteristics of Risk Assessment Users and Work Activities

     The Role of Risk Assessment in Regional Operations

     The Steps of Risk Assessment

     The Site-Specific Risk Assessment Process:  A Summary Model

     Staff Recommendations for Improving the Process

-------
III.   CONCLUSIONS AND RECOMMENDATIONS	53



      Conclusions




      Recommendat ions




 APPENDICES	58




 A-l.  Region IV Organizational Chart




 A-2.  Protocol




 A-3.  Survey Form




 A-4.  Analytic Strategy




 A-5.  Major Activities at Region IV




 BIBLIOGRAPHY	100

-------
                                                          VI

                        LIST OF TABLES


TABLE                                                   Page

1.    Region IV Organizational Units Studied              11

2.    Human Health Risk Assessment                        12

3.    Types of Activities Studied                         15

4.    Characteristics of Activities
      Studied                                             17

5.    Risk Assessment:  Some Examples                     18

6.    Number of Risk Assessment Steps Performed           22

7.    Regional Risk Assessment Responsibilities           24

8.    Key Characteristics of Risk Assessment Steps        25

9.    Hazard Identification Responsibility                29

10.   Crosstabulation of Hazard Identification
      Responsibility by Basis                             30

11.   Crosstabulation of Hazard Identification
      Responsibility by Approach                          31

12.   Risk Assessment Sources of Information              32

13.   Dose Response Responsibility                        35

14.   A General Description of CDC's Health Assessment
      Protocol                                            36

15.   Exposure Assessment Responsibility                  38

16.   Crosstabulation of Exposure Evaluation
      Responsibility by Basis                             40

17.   Crosstabulation of Exposure Evaluation
      Responsibility by Approach                          41

-------
                                                          Vll
TABLE                                                    Page

18.   Risk Characterization Responsibility                43

19.   Crosstabulation of Risk Characterization
      Responsibility by Basis                             44

20.   Crosstabulation of Risk Characterization
      Responsibility by Approach                          45

21.   Nature of Risk Characterization:
      Quantitative vs Qualitative                         46

22.   Staff Recommendations                               51

-------
                                                        viii

                    LIST OF ILLUSTRATIONS


FIGURE                                                   Page

1     Elements of Risk Assessment and Risk                6
      Management

2     Time and Resources for Risk Assessment             20

3     Index of Risk Assessment Activity                  23

4     Region IV Risk Assessment Operations               48

-------
                                                        IX
CAA
CAG
CDC
CERCLA

CPSC
CWA

EA
EIS
EPA
ERT

FIFRA

FR

NAAQS
NCP
NEICC
NEPA
NESHAP

NIOSH

NPL
NRT

OAQPS
ODW
OHMTADS

0PM
OPTS
ORD
OSC
OSHA
OTS

PMN
PSD

RAMP
RCP
RCRA
RI/FS
             GLOSSARY

Clean Air Act
Carcinogen Assessment Group
Centers for Disease Control
Comprehensive Environmental Response,
  Compensation & Liability Act of 1980
Consumer Product Safety Commission
Clean Water Act

Endangerment Assessment
Environmental Impact Statement
Environmental Protection Agency
Emergency Response Team

Federal Insecticide, Fungicide, and
  Rodenticide Act
Federal Register

National Ambient Air Quality Standard
National Contingency Plan
National Enforcement Investigations Center
National Environmental Policy Act
National Emission Standards for
  Hazardous Air Pollutants
National Institute of Occupational
  Safety and Health
National Priority List (CERCLA)
National Response Team

Office of Air Quality Planning and Standards
Office of Drinking Water
Oil & Hazardous Materials Technical Assistance
  Data System
Office of Policy and Management
Office of Pesticides and Toxic Substances
Office of Research and Development
On-Scene Coordinator
Occupational Safety and Health Administration
Office of Toxic Substances

Premanufacture Notice
Prevention of Significant Deterioration

Remedial Action Master Plan
Regional Contingency Plan
Resource Conservation and Recovery Act
Remedial Investigation/Feasibility Study

-------
ROD           Record of Decision
RRT           Regional Response Team
RTF           Research Triangle Park,  North Carolina

SDWA          Safe Drinking Water Act
SIP           State Implementation Plan

SNARLS        Suggested No Adverse Reaction Levels
STORET        Storage and Retrieval for Water Quality Data

TSCA          Toxic Substances Control Act

-------
                                                         XI
                         SUMMARY
The United States Environmental Protection Agency recently
initiated several efforts to ensure the quality and consistency
of its scientific data base for regulatory decision making.
One of the areas of intense investigation has been the risk
assessment process.  Although risk assessment in the context
of regulatory development has been investigated, little
attention has been directed toward the role of risk assessment
in the site-specific implementation of EPA's laws, regulations,
and policies.  This study examines the use of risk assessment
in EPA Regional operations, where much of that implementation
occurs.  EPA's Regional offices are known to play a more
significant role in Agency activities than field offices of
the typical federal agency, and are frequently called upon to
perform site-specific risk assessments.

The objectives of this study were to determine what Regional
programs were using risk assessment, how, and what Regional
decisions it was useful for.  The study focuses on the use of
risk assessment to estimate human health impacts from toxic
chemical exposures in site-specific situations.  It involved
a survey of regulatory programs at a "typical" Regional
office: EPA Region IV, Atlanta, Georgia.  Interviews were
conducted with 39 staff from Region IV - these staff described
the activities of 80% of the professional staff at the Regional
office.

       The major findings are:

0 Risk assessment is used within a number of different program
contexts at the Regional level.  Some uses for risk assessment
include establishing priorities for remedial action and
determining control strategies at toxic waste dumps under the
Superfund program; permitting hazardous waste treatment or
disposal facilities, establishing alternate concentration
levels, and supporting administrative actions when" permit
violations occur under the RCRA program; assisting Regional
programs and States in developing acceptable control strategies
for toxic air emissions; and, estimating public health impacts
likely to occur as a result of proposed actions through the
Environmental Impact Statement process.

-------
                                                        Xll
0 Staff involvement in risk assessment is from one of three
roles: assessment, management, or overview.  The majority of
Regional work involves managing or overviewing risk assessments
performed by other parties, e.g. States, permittees, contractors,
Approximately 70% of Regional activities require performance,
management, or overview of parts of the risk assessment
process.  The process is sometimes done implicitly and on an
informal basis; more recently formal risk assessments have
been initiated in EPA's CERCLA and Air programs.  The recent
reauthorization of RCRA also provides for risk assessments to
be performed in hazardous waste facility permitting.

0 The level of risk assessment activity varies across Regional
programs.  Currently CERCLA regulatory programs, and technical
support staff to CERCLA, show the highest levels of activity.
The average amount of time Sections spend on risk assessment
activities is 30%.

0 The site-specific risk assessment process involves some or
all of the following steps:  (1) hazard identification,
(2) toxics evaluation, (3) exposure assessment, and (4) risk
characterization.  While the Region often has primary
responsibility to perform steps 1, 3, and 4, staff depend on
EPA Headquarters offices, other federal agencies and external
sources for the dose-response information needed to perform
step 2.

0 As risk assessment is relatively new to Regional operations,
systems and procedures are not yet in place to facilitate use
of this approach.  For example, staff have difficulty accessing
current dose-response information available within the Agency.
In addition, exposure assessment protocols are not always
designed to estimate impacts on receptor populations.  Improve-
ments in Agency data bases, information access, and exposure
assessment protocols could help ensure the quality of the
toxics evaluation and exposure assessment steps.

0 The Centers for Disease Control play a major role in risk
assessments performed in CERCLA programs due to the inter-
agency agreement between the two agencies for health assess-
ments.  EPA and CDC currently have different philosophies
regarding the use of risk assessment.  While EPA is taking

-------
                                                      Xlll
steps to ensure consistent use of risk assessment to the
extent possible across its regulatory programs, CDC has
stayed away from quantitative risk assessment techniques and
currently has no policy regarding their use.  The consistency
of site-specific risk assessments could be increased by improved
interaction between the two agencies.  An interagency task
force should be set up to examine philosophical differences
and working relationships between these two Agencies.

The most striking feature of risk assessments performed in
Regional operations is the high degree of information transfer
and coordination they require among Regional program offices,
and between Regional programs and Headquarters.  For example,
the Regional office depends on many sources for dose-response
data, then performs the exposure assessment and risk charac-
terization itself to complete the risk assessment process.
In addition, although risk assessment requires a multi-media
analysis, EPA's programs often focus on a single media.

Although the Region does have much of the expertise required
to perform the different steps of a risk assessment, organiza-
tional structure can be a barrier to integrating this expertise
in problems impacting several media.  Regional managers
pointed out that information exchange and coordination between
key actors is often less than adequate.  It appears the
output of the risk assessment process to Regional decision
makers could be improved through better utilization of existing
resources, especially Regional subject matter specialists and
computerized data bases.  To meet these needs and facilitate
internal consistency, we recommend an inter-program group with
expertise in specific areas of risk assessment be established.
The group would work towards achieving scientific consensus on
the health risk at specific sites, integrate EPA programs to
develop multi-media work plans for toxic chemical exposures at
specific sites, and serve as a peer review mechanism for risk
assessments performed and managed within the Region.

-------
                          CHAPTER I

                         INTRODUCTION

This report is about the use of risk assessment in EPA's
Regional operations.  The Agency has recently initiated
several efforts to ensure the quality and consistency of its
scientific data base for regulatory decision making.  One of
the areas of intense investigation has been the risk assessment
process.  These efforts have focused on risk assessments
performed as part of the development of EPA regulations and
standards.  Little attention has been given to the site-
specific assessments performed to implement and enforce these
regulations and standards.

The purpose of this study is to explore the role of risk
assessment in EPA Regional operations, where the implementation
of regulations and standards occurs.  The objectives are:

          (1) To determine the nature and function of risk
              assessment in the context of EPA regulatory
              programs at the Regional level,

          (2) To define how and in what programs risk assess-
              ment is performed in Regional operations; and

          (3) To develop conclusions and recommendations on
              the use and effectiveness of risk assessment
              techniques in Regional regulatory activities.

Historical Background

The scientific study of risk and the technology for its
control are not new.  Risk assessment has been practiced
since the formation of insurance companies in 14th century
Spain.   Risks in the modern business world have been
estimated for years by both formal and informal methods.

-------
                             -2-
The newly emerging field of environmental risk assessment
gathered momentum during the early days of the nuclear power
industry.  One small section of the National Environmental
Policy Act (NEPA) required an environmental impact statement
on all federal projects - an empirical assessment of the
possible impacts (outcomes) and their relative likelihood
(Kates, 1977). Confronted with this new requirement, the
nuclear power industry was facing the task of proving the
benefits from nuclear power outweighed the health risks.  Due
to limited experience with nuclear power, very little hard
evidence was available upon which to base empirical risk
assessment.  The nuclear industry responded with the develop-
ment of several hypothetical risk models that have influenced
the development of risk assessment to this day (Chestnutt,
1983).  Risk assessment of human exposure to radioactivity
offered the conceptual framework for the assessment of risk
from exposure to chemicals.

The evidence for health effects of a few chemicals, such as
asbestos, has been clear; in many more cases the evidence is
meager and indirect.  To aid decision making, agencies have
used risk assessment to identify chronic health hazards and
estimate the risks to human health posed by toxic chemicals.
But rather than alleviating the controversy over regulatory
decisions, risk assessment procedures have become a focus of
criticism by scientists, industry, and the public.  Criticisms
of risk assessment have ranged broadly from details of the
process to administrative management to statutory authority
(National Academy of Sciences, 1983).

Several years ago the National Research Council formed the
Committee on Institutional Means for Assessment of Risks to
Public Health.  The Committee was directed to study the issue
of increased risk of cancer resulting from exposure to chemi-
cals in the environment.  They examined whether changing
institutional arrangements or procedures for risk assessment
could  improve the regulatory performance of federal agencies.
In their book Risk Assessment in the Federal Government;
Managing the Process, the Committee reviewed organizational
structures and processes of risk assessment at EPA and the
three  other agencies who regulate exposure to carcinogens.
They concluded that guidelines which clearly state the in-
ferences to be made in each step of risk assessment would be
useful, but that the controversy over risk assessment is
largely due to the sparseness and uncertainty of the scien-
tific  data base  it is dependent upon.

-------
                            -5-
evidence.  Both scientific judgments and policy choices may
be involved in selecting from among possible inferences, such
as the judgment that animal bioassays are indicators of
probable human response, or that no threshold of response to
carcinogens can be confidently defined.  NAS coined the term
"risk assessment policy" to differentiate those judgments and
choices from the broader social and economic policy issues
that are factored into risk management decisions.  NAS found
at least some of the controversy surrounding regulatory
actions has resulted from a blurring of the distinction
between risk assessment policy and risk management policy
(see Figure 1).  Risk management is the process of weighing
alternatives and selecting the most appropriate regulatory
action.  It involves integrating the results of risk assess-
ment with engineering feasibility, social, economic, and
political concerns to reach a decision (Ruckelshaus, 1984).

The term "risk assessment" is often given narrower and
broader meanings than described in this report.  For some,
the term is synonymous with quantitative risk assessment and
emphasizes reliance on numerical results.   For our purposes,
we studied risk assessments that involved both quantitative
and qualitative expressions of risk.

EPA and Risk Assessment

EPA administers nine comprehensive environmental protection
lawss the Clean Air Act (CAA); the Clean Water Act (CWA); the
Safe Drinking Water Act (SDWA); the Comprehensive Environ-
mental Response, Compensation, and Liability Act (CERCLA,
or, "Superfund"); the Resource Conservation and Recovery
Act (RCRA); the Federal Insecticide, Fungicide, and Roden-
ticide Act (FIFRA); the Toxics Substances Control Act
(TSCA); and the Uranium Mill Tailings Radiation Control Act
(U.S. EPA, 1984a).   These laws all contain different
provisions for protecting the public from the health risks
of toxic chemical exposures.  The history of these laws
reflects a maturation of the nation's goals for protecting
human health.  Earlier laws, such as the Clean Water and
Clean Air Acts, had the ambitious goal of reducing to zero
any risks to human health in the environment.  The goal of
zero risk over time proved to be unachievable with current
scientific, technological, and economic resources.  Later

-------
                                                    -6-
          RESEARCH
                                               RISK ASSESSMENT
                                                                                      RISK MANAGEMENT
     Laboratory and fwld
     observations of advene
     health effects and ex-
     posures to particular
     •BOTH
     Information on
     txtrapolatian methods
     for high to low dota
     and animal to human
    Field measurements,
    estimated exposures,
    characterization of
    populations
Hazard Identification
(Does the agent cause
the advene effect?)
(Wtet is *» relationship
between dose and inci-
dence in humane?)
Expoaura As*fwmnt
(What e»{Meures an
currently experionead
or antieipetad under
difterant conditions?}
Development of
regulatory options
Evaluation of public
health, economic,
social, politieal
consequences of
regulatory options
                             Risk Characterization
                             (What is to estimated
                             incidence of the ad-
                             verse effect in a
                             0vwi population?)
                                                                                   and actions
                                             FIGURE  1
               ELEMENTS  OF RISK  ASSESSMENT AND  RISK MANAGEMENT
Reference:   Risk  Assessment  in  the  Federal  Government:
                  Managing  the  Process

-------
                             -7-
laws such as the Toxic Substances Control Act  (TSCA), reflect
this realization.  TSCA calls for a balance between risks to
human health and the benefits of a chemical to society.  The
later laws are also more vague with respect to how this
balance should be obtained, but the following national envi-
ronmental goals can be identified:

1)  the public should be protected from unreasonably high
    risks to their health from chemical exposure.

2)  protection should be determined in light of the potential
    benefits to society from use of these chemicals.

Recently, the public has expressed concern that toxic chemi-
cals in the environment are affecting their health, and that
political considerations have taken precedence in the Agency's
decisions about how to reduce these health risks.  In response
to this public concern, EPA took several actions to ensure
quality and consistency in its scientific basis for decision
making.

In 1983, the Agency formed a Toxics Integration Task Force
to suggest practical improvements to risk assessment and risk
management that could be implemented within EPA's existing
organizational and statutory limits.  The Task Force investi-
gated risk assessment and risk management practices in the
context of regulatory development decision making at Head-
quarters.  They found that although regulatory programs do
risk assessment and risk management, they are not all doing
it in the same way.  This finding was not surprising, as
EPA's organizational structure has been referred to as a
"patchwork" of regulatory programs which function autonomously
to administer and implement the nine environmental laws.
These offices have carried out their mandates with a history
of considerable independence -- EPA is considered to be the
most decentralized among U.S. regulatory agencies (National
Academy of Sciences, 1977).

As EPA concentrated more attention on risk assessment and
risk management,  it became clear that these activities may
be construed as being nearly the whole of EPA's mission; that
is, most of what EPA does involves finding out what harmful
effects pollutants have on people and the environment ("risk
assessment," broadly speaking) and deciding on the best way of
mitigating those effects i.e., risk management (Risk Manage-

-------
                             -8-
ment Subcommittee, 1984).  Furthermore, the distinction
between risk assessment and risk management decisions often
become rather fuzzy under such broad definitions.

The central issue EPA is addressing is the extent to which
risk assessment judgments can be made more consistent and
more reflective of the state of scientific understanding.
The Agency recognizes that there can be no constant formula
for conducting a risk assessment.  Because it is an analytical
tool, it can be argued that risk assessment must be tailored
to the needs of the program in which it is used (EPA, 1984b).
The Task Force recommended EPA develop intra-Agency risk
assessment guidelines and establish a forum of Agency scien-
tists to review and reach scientific consensus on risk assess-
ment issues.

Five sets of guidelines were recently published in the Federal
Register (U.S. EPA 1984 c, d, e, f and 1985) and the Ris~k
Assessment Forum was formed in 1984.  The guidelines are
statements of EPA's risk assessment policy augmented by more
specific technical guidance appropriate to regulatory devel-
opment and standard setting.  The guidelines are intended for
use by scientists throughout the agency to assess the risk of
carcinogens (cancer causing effects), mutagens, (genetic
damage), developmental toxicants (birth defects or terato-
genicity), chemical mixtures, and exposure.  As they were
developed primarily by workgroups of EPA scientists who perform
risk assessments for standard-setting, they are tailored for
this end use.

The Study

Although risk assessment in the context of regulatory devel-
opment has been investigated, little attention has been
directed toward the role of risk assessment in the site-
specific implementation of EPA's laws, regulations, and
policies.  Thus, this study was conceived of to define the
nature and function of risk assessment in a "typical"
Regional office.

EPA has ten Regional offices throughout the U.S. that are
similar in form and scale.  Although program decisions can
and do vary widely, Regional offices work closely with state
and local governments to make sure Regional needs- are considered
and that federal environmental laws and regulations are
properly implemented. (U.S. EPA, 1984a).

-------
The study examines the current and potential uses for risk
assessment within the context of Regional program activities.
Our focus is on risk assessment as an input to risk management
and any other Regional office decisions, although risk assess-
ment is only one factor in the process by which the Agency
controls hazardous substances.  The social, political, and
economic factors affecting risk management decisions were
outside the scope of this initial effort.

EPA's Regional offices play a more significant role in
Agency activities than do field offices for the typical
federal agency.  41% of EPA's employees are located in
Regional offices, nearly matching the 42% who are in Head-
quarters (the remaining 17% work in research laboratories).
Regional offices have extensive authority in many decision
areas: Superfund cleanup actions, the review of State stan-
dards and plans, the monitoring and enforcement of discharge
standards, and the granting of substantial operating funds to
State agencies (NAS, 1977).  Many of these decisions require
knowledge of the potential health risks of substances in
order to make balanced decisions about how to reduce these
risks.

The office selected for intensive study was the Region IV
office in Atlanta, Georgia.  Region IV's organizational
structure at the time of the study is presented in Appendix
A-l.  At Region IV, regulatory program offices were grouped
under four divisions:  two divisions were organized according
to program areas (air and waste, water) and two dealt with
functional areas (management, technical support).  [Note:
Since that time, the office has been reorganized into five
divisions - air and waste management have been separated.  We
did not feel this subsequent change had any major effect on
our findings.]

Many studies have been performed on the use of one sort of
analysis or another in organizational decision-making.  Two
common research methods employed to study the use of analysis
in organizations are case study and sample survey.  In our
study of risk assessment at Region IV, we wanted to produce
data that both described the frequency of risk assessment
activities and offered insights into how these risk assessments
were done.  Given this situation, a research method was designed
that incorporated both qualitative and quantitative strategies
(Yin and Heald, 1975) .

-------
                            -10-

Th e research strategy employed in this study was a survey
augmented with qualitative case study techniques.  We collected
qualitative evidence on the use of risk assessment across
programs to produce a "snapshot" of the Regional Office.  The
snapshot was sketched with baseline information on each
program's operations, and focused in with more detail in
offices where risk assessment activity was greater.  As this
was a pilot study, we were unsure which offices were involved
in risk assessment.  There was no information available on
what Regional programs had used this type of analysis in the
past.  Thus, we wanted to first identify those programs which
were involved in the risk assessment process, and from those
programs gather as much information as possible on useful
approaches and sources of information worth sharing across
programs.

The survey involved taking a census of the various risk
assessment activities in each Program Office.  A  list of
the organizational units studied is presented in Table 1.
A census of activities was chosen over more popular sample
strategies to allow us to better define the Regional risk
assessment process, and to prevent our completely over-
looking risk assessment techniques that were worth sharing.

The definition of risk assessment used in the study is the
4-step process outlined in Table 2 (adapted from NAS, 1983;
U.S. EPA, 1984 g).  A detailed discussion of the survey
protocol, data collection and analytic strategy can be found
in Appendices A-2, A-3, and A-4 respectively.  In Chapter II,
we present the results of the study.

-------
                            -11-
Regulatory
Program
CWA 404,
EIS Review
EIS Prep.

Building
Mgmt.
NPDES
 Permit
201 Grants
SDWA, UIC
               Table 1

Region IV Organizational. Units Studied


  OFFICE OF ARA FOR POLICY AND MANAGEMENT
    Environmental Assessment Branch
      Environmental Review Section
      NEPA Compliance Section

    Management Branch
      Management Services Section

  WATER MANAGEMENT DIVISION
    Facilities Performance Branch
      Permits Section
      Industrial Operations Section

    Facilities Construction Branch
      Applied Technology Section
      South Area Grants Management Section

    Water Supply Branch
      Drinking Water Section
      Groundwater Section
Program
 Support

CERCLA, RCRA
Air, Water
 Support

Air, Water
CWA 404
CERCLA,
Superfund
RCRA
FIFRA,
TSCA
Clean
 Air Act
  ENVIRONMENTAL SERVICES DIVISION
    Analytical Support Branch
      Chemistry Section

    Engineering Support Branch
      Hazardous Waste Section
      Municipal & Industrial Section

    Ecological Support Branch
      Air and Water Monitoring Section
      Environmental Biology Section

  AIR AND WASTE MANAGEMENT DIVISION
    Emergency and Remedial Response Branch
      Site Screening and Engineering Section
      Emergency Response and Control Section
      Remedial Action Section

    Residuals Management Branch
      Waste Planning Section
      Waste Engineering Section
      Waste Compliance Section

    Pesticides and Toxic Branch
      Pesticides Section
      Toxics Section

    Air Management Branch
      Air Planning Section
      Air Engineering Section
      Air Compliance Section

-------
                                   -12-


                                 TABLE 2

                       HUMAN HEALTH RISK ASSESSMENT

In the simplest sense, risks to human health are a function of two factors:
hazard (i.e., toxicity) and exposure.  It is thought that assessments of
these health risks contain some of the following steps.  THE SCOPE OF
THESE STEPS MAY VARY MARKEDLY, FROM BACK OF THE ENVELOPE TO A DETAILED
ANALYSIS, DEPENDING ON THE PURPOSE OF THE RISK ASSESSMENT.

Steps in Risk Assessment

1  HAZARD IDENTIFICATION;  weighing the available evidence to determine
whether a particular site or substance presents a hazard to human health
and identifying the general nature of that hazard (e.g.,  cancer, birth
defects, fire).  Hazard ID helps one decide whether to take action in a
particular situation; it involves a "preliminary assessment" and results
in a list of human health effects caused by the hazard.  This step also
identifies known effects and relevant routes of exposure.

2  DOSE-RESPONSE EVALUATION;  determining how strongly a hazard elicits
identified responses at various levels of exposure (e.g., saccharin and
dioxin both cause cancer in animals, but it takes a million times more
saccharin than dioxin to produce equivalent effects in the lab) to determine
an acceptable intake level for humans.

3  EXPOSURE EVALUATION;  estimating the likely degree of human exposure
to a hazard of concern; methods may involve ambient monitoring or estimates
of environmental releases that predict ambient concentrations.

4  RISK CHARACTERIZATION;  combining information from steps 2 and 3 above
to determine the risk to human health associated with the particular
exposures in the situation being considered.  This step may be done
quantitatively (e.g., 1/1000 cancers) or qualitatively (e.g., a substantial
risk).

Examples of Risk Assessment Activities

1  RISK SCREENING METHODS;  used to set priorities or assess the danger
inherent in broad ranges of substances.  Examples include the Waste/
Environment/Technology method used to screen hazardous waste streams in
the office of Solid Waste, and the Superfund hazard ranking system.

2  RISK ASSESSMENTS FOR EMERGENCY AND CLEAN-UP OPERATIONS;  an assessment
of the immediate danger to people in areas that have been contaminated
with toxics and, in some cases, an estimate of the risks associated with
various clean-up options.  Included in this category are the Health
Advisories put out by ODW in response to requests by local water authorities.

-------
                         CHAPTER II

             RISK ASSESSMENT IN REGIONAL OPERATIONS


In this Chapter a "snapshot" of the role of risk assessment
in Regional operations is developed based on the results
of the organizational survey.  The first part of the Chapter
describes some general characteristics of Regional risk
assessment users and work activities.  Next, the steps of
the risk assessment process are discussed in detail, and the
key actors involved in each identified.  Finally, the risk
assessment process as practiced in Regional operations is
summarized, and staff recommendations are presented for
improving the quality and consistency of risk assessments.

Characteristics of Regional Risk Assessment Users and
Work Activities

A risk assessment user was defined for purposes of this
study as a person who has occasion to perform, or oversee
the performance of, risk assessments.  The population of
possible risk assessment users represented all four Divi-
sions and 26 Sections at Region IV.  Where a Section's
operations were more related to risk assessment, a larger
number of staff were interviewed.   A total of 39 supervisory
and project staff described 80% of the professional staff
activities at the Regional office.  This included EPA
program support activities of the Centers for Disease
Control (CDC).  The majority of staff interviewed were
engineers; some were environmental scientists, and a smaller
percentage were specialists in other scientific fields.

Each Section served as a unit of analysis for the study.
Sections were responsible for implementation of from one
to six programs (refer to Table 1  for a list of the Sections
and their corresponding programs).  A program was defined
as two or more decisions or activities required to implement
the same regulatory program within a Section.   65% of
Sections were responsible for implementation of a single
program; the mean number of programs per Section was 1.4.

-------
                             •14-
Information was collected on a total of 130 program decisions
and activities.  These 130 activities define the entire
context within which risk assessment is done at Region IV.  A
complete list of these program activities can be found in
Appendix A-5.  Each of these activities were performed within
one of three roles: assessment, management, and oversight.

Assessment activities are defined as those which require
staff to collect and analyze scientific data on a site-specific
problem.  Also included in this category are the various
types of technical assistance the Region provides upon request
to States and other external parties.  Management activities
were those where Regional staff were directly responsible for
implementation and administration of EPA's regulatory programs,
and made decisions about controlling environmental pollutants.
These activities often involved managing research investiga-
tions to estimate the actual or potential magnitude and extent
of environmental contaminants at specific sites.  Oversight
activities were those where regulatory program implementation
had been delegated to a third party, often the States or
industry, and the Regional office had an indirect role in
assuring compliance with these programs.  These activities
typically involved evaluation of completed environmental
studies (including risk assessments) evaluation of record-
keeping practices, and inspection of facilities.

Regional activities are almost equally distributed among
these three roles: 38% assessment, 34% management, and 28%
oversight.  Table 3 summarizes the frequency of assessment,
management and overview activities.  Comparison between
programs shows this distribution does vary from one program
to the next.  As one might expect, the percent of assessment
activity is greatest (79%) in the Environmental Services
Division (ESD), which provides technical support to a number
of regulatory programs.  The Office of Policy and Management
(OPM) would be expected to manage and oversee regulatory
programs, rather than directly perform assessment activities.
Note that none of OPM's activities are in the "assess" category,
The rest of the programs possess a more even distribution of
these three types of activities, with the exception of RCRA.
Because many RCRA programs are delegated to the States,
RCRA's Regional activities are primarily in oversight and
management roles.

-------
                -15-
              TABLE 3
    Types of Activities Studied
(absolute and relative frequencies)

DIVISION AND
PROGRAM
OPM

NEPA, EIS Review,
Ocean Disposal,
Building ^Services
WATER

SDWA, CW&,
Grcundwater
BSD

CERCLA, RCRA,
Air, Water
Support
AWMD

CERCLA
RCRA

TSCA, FIFRA

CAA

TOTAL
(Mean Percent)
ASSESS
0
(00)



6
(27)


23
(79)



10
(36)

1
(10)
4
(50)
5
(24)
49
(38%)
FUNCTION

MANAGE
7
(58)



9
(41)


1
(3)



15
(54)

3
(30)
4
(50)
6
(28)
45
(34%)


OVERSEE
5
(42)



7
(32)


5
(18)



3
(10)

6
(60)
0
(00)
10
(48)
36
(28%)
TOTAL
ACTIVITIES
12
(100%)



22
(100%)


29
(100%)



28
(100%)

10
(100%)
8
(100%)
21
(100%)
130
(100%)

-------
                            -16-
Th e majority of Regional activities (81%) are perceived
by staff as having direct impacts on human health (See Table
4).  This finding was consistent across programs, as responses
ranged from 60% in RCRA to 100% in Air programs.  The latter
was high due to this office's role in assisting States and
other Regional programs in assessing and regulating sources of
toxic air pollution.

The majority of Regional activities currently, or have
the potential to, involve risk assessment.  68% of activities
currently involve performing, managing, or overviewing parts of
the risk assessment process.  Table 4 displays the percentage
of Regional activities that are related to risk assessment.
The percent of risk assessment related activities varied
across regulatory programs, from 40 to 80%;  BSD, CERCLA, CAA,
and OPM Programs were all clustered at the higher end of this
range.  RCRA programs showed the greatest potential to expand
into this area; risk assessments are soon to be required in
authorizing hazardous waste facilities, in issuing certain
types of administrative orders, and in setting alternate
concentration levels.  The finding that the majority of Re-
gional activities involve risk assessment parallels previous
studies (Risk Management Subcommittee, 1984) done at EPA that
found "risk assessment" and "risk management" appropriately
defined nearly the whole of EPA's regulatory development
activities.

The Role of Risk Assessment in Regional Operations

   Examples of Risk Assessment Use

Risk assessment  is used in Regional operations within a
number of different decision-making contexts and  for a number
of different purposes.  The risk assessment  is often done
implicitly and on an informal basis; more  recently formal
risk assessments have been  initiated under the Region's
CERCLA and Air programs.  There  is also  a  potential for
greater use of risk  assessment  in the  near future; for
example, in last year's reauthorization  of RCRA,  Congress
mandated that  risk  assessment be performed in permitting of
hazardous waste  facilities  (Public Law 98-616, Nov. 8, 1984).

At least seven formal  risk  assessments were  in various stages
of completion  at the Regional office during  1984;  these
examples are presented  alongside the context in  which  they
are  used in Table  5  [Note:   This estimate  does not  include
assessments performed  in  emergency  response  and  control
actions].  Air programs perform risk assessments  to assist
other Regional programs  and States  in  developing  acceptable
control  strategies  for  toxic  air emission  sources.  They

-------
                 —17—
               TABLE  4
Characteristics of Activities Studied
 (absolute and relative frequencies)
DIVISION AND
PROGRAM
0PM

NEPA, EIS
Review,
Ocean Dis-
posal, Bldg.
Services
WATER

SDWA, CWA,
Groundwater
BSD

CERCLA,
RCRA, Air,
Water
Support
AWMD
CERCLA

RCRA

TSCA,
FIFRA
CAA

TOTAL

Perceived Human Health Impact
None Indirect Direct
2
(17)





3
(14)


0
(00)





7
(25)
0
(00)
1
(13)
0
(00)
14
(11%)
1
(8)





2
(9)


4
(14)





0
(00)
4
(40)
0
(00)
0
(00)
11
(8%)
9
(75)





17
(77)


25
(86)





21
(75)
6
(60)
7
(87)
21
(100)
105
(81%)
Risk Assessment
None Potential
3
(25)





5
(23)


3
(10)





7
(25)
4
(40)
2
(25)
6
(28)
30
(23%)
0
(00)





3
(14)


3
(10)





1
(4)
2
(20)
2
(25)
0
(00)
11
(9%)

USE
9
(75)





14
(63)


23
(80)





20
(71)
4
(40)
4
(50)
15
(72)
89
(68%)
TOTAL
ACTIVITIES
12
(100%)





22
(100%)

/
29
(100%)





28
(100%)
10
(100%)
8
(100%)
21
(100%)
130
(100%)

-------
                                                 -18-
                                               Table 5
                                   Risk Assessments;   Sane Exanples
                                          (Region IV - 1984)
Legislative
Act	

CAA
CAA
CAA
CERCLA
Regulatory
Program

State Air
Toxics
State Imple-
mentation
Plan (SIP)
State Imple-
mentation
Plan (SIP)

Investigation
and Compliance
CERCIA
Superfund
Remedial Action
   Exanple

Fairbanks Sand Pit
Gainesville, Florida
Savannah River Plant
Savannah, Georgia
Formaldehyde
Region IV Office
Brown Wood Preserving
Live Oak, Florida
Biscayne Aquifer
Miami, Florida
         Contexts	

Perform exposure assessments
for states to develop compliance
strategies for particular sources.

Provision of technical assistance
to states and Region IV programs
upon request; in this case South
Carolina and Region IVs Ground-
water program.

Provision of technical assistance
to Region IV1 s Office of Policy
and Management upon request.-

Research NPL sites to determine
responsible parties and docunent
on a preliminary basis, the existance
of a threat to himan health and the
environment? part of this process is
an endangerment assessment.

Perform remedial investigations and
feasibility studies to obtain more
information on NPL sites; part of
this process  is an endangerment
assessment.
CERCLA


CERCLA
Superfund
Remedial Action

Superfund
Remedial Action
Gold Coast Oil
Miami, Florida

Vliite House Oil Pits
Jacksonville, Florida
                                                              Same as above.
 Same  as above.

-------
                            -19-
 typically  depend  on  the  Environmental  Criteria  and  Assessment
 Office  (ECAO-RTP)  for  dose/response  data  and  couple this  with
 exposure estimates obtained  from dispersion modeling to
 estimate risk.  CERCLA programs perform risk  assessment to
 help  set priorities  for  remedial action and determine control
 options at Superfund toxic waste sites.   In both  emergency
 and planned  removals,  a  Regional Response  Team  (RRT)  is
 formed  to  determine  actual risks as  well  as resolve risk
 management issues  such as "how clean is clean?"   NEPA and
 environmental review staff are called  upon to evaluate risk
 assessments  performed  by other agencies in federal  activities
 programs.

  The Level  of Risk  Assessment Activity

 Several measures were  used to describe the level  of  risk
 assessment activity  in the Regional  office.   Besides  the
 percent of activities  requiring use  of risk assessment (see
 discussion on p.  16),  the time and resources  spent  and the
 number of  risk assessment steps performed  by  each Section
 were  investigated  to estimate the level of risk assessment
 activity.

 The time and resources spent on risk assessment were  30%  in
 relation to other Regional activities.  Time  and  resources
 varied over  a considerable range, from 2 to 95%,  across
 regulatory programss (See Figure 2).  Higher  levels  of time
 and resources are currently being expended in CERCLA,  and the
 BSD activities in support of CERCLA, than  in  other  regulatory
 programs to perform  formal risk assessments.  This  is  because
 the Office of Solid  Waste and Emergency Response  (OSWER) has
 directed Regions to  perform endangerment assessments  to assess
 health risks from chemical exposures at toxic waste  sites.
 OSWER has  also developed guidance for use at  CERCLA  sites and
 RCRA facilities that provides a conceptual framework  for
 performing risk assessments.

 The time and resources spent on risk assessment were  difficult
 to estimate for several  reasons.  First, staff often  felt
 risk assessment as performed for regulatory development did
 not accurately describe  the Region's site-specific  risk
 assessment process.  Therefore, there was a reluctance to
 estimate time and resources spent on the type of risk  assess-
ment defined in Table  2  (we did note patterns in these
 differences that were useful  to develop a working definition
 of the typical site-specific  risk assessment).  Second, the
 risk assessment process  is often initiated by the spontaneous
 discovery of a potential public health problem.   This  makes
 time and resources spent highly irregular and difficult to
 estimate.   Third,  risk assessment is often performed  informally
on a routine basis.

-------
                                                -20-
1UU -
90 -
80 -

70 -
Ul
2 60 -
F
Z 50 -
o
ft:
u 40 -
0.
30 -
20 -
10 -
0 -















pn
/I M
?•
^
^
/
/





12345
0 F M W A














_
^
/
/

6
















m
7-
^
^
/

^
^
/
7

/
r
/
/
/
j
f
/
/
/
/
/
/
/
^
/
/
/
/
/
/
/
y
y
/
/
/
/
/
/
/
/













P
^
<

7 8 9 10 11 12
TEH ESI)












-7i



M
^ • — i — • 	 1 • r ' i i • i
/•
$
^
<
rr-|
/
/
/
/
/
/
f
>
r
>
/
/
/
/
/
/
/
















•7-
/
>
r
/
/
^
/
<
7

/
/
/
/
/


/
/
^
^
^













MR R
—
/
/
^
/
;/





f/l !/

13 14 15 16 17 18 19 20 21 22 23 24 25 26
WAiTE TOXICS AIR
Mean - 30.41
Standard Deviation- 28.6
         Range - 93 (2 to 95X)
         Median - 21.51

             FIGURE 2
I Section time for illsk Assessment

-------
                           -21-
Although risk assessment is defined as a four-step process, a
Section is typically responsible for only one or two of these
steps.  Note the mean number of steps is only 1.5 per section
(See Table 6).  In addition, only 23% indicated their Section
had primary responsibility for performing or managing the
entire site-specific risk assessment.  Thus, the site-specific
risk assessment typically requires the coordinated effort of
several regulatory programs and a variety of scientific
disciplines.

Each of the measures discussed so far provided different
information about the level of risk assessment activity in
Regional operations, and were fairly independent of one
another.  For example, the time and resources spent on risk
assessment did not increase with the number of steps of risk
assessment performed or the percent of risk assessment activi-
ties in a section, as one might expect.  To determine which
Regional programs were most active in risk assessment, stand-
ardized scores were developed to depict a Section's level of
risk assessment activity on all three of these variables:
number of decisions requiring risk assessment, time and
resources spent, and number of steps performed.  Figure 3
ranks Sections according to how they scored on all three
measures.  From these results, it appears that CERCLA programs,
and the technical support functions to CERCLA, (i.e., two
Sections of BSD and CDC) currently are most heavily involved
in risk assessment activities.

The Steps of Risk Assessment

The Regional office is clearly more involved in certain steps
of the risk assessment process than others (See Table 7).
Regional office responsiblity is highest in hazard identifi-
cation (55%), exposure assessment (50%), and risk characteri-
zation (38%).  None of the Sections actually perform dose-
response assessment; however, dose-response information was
needed to perform site-specific assessments.  Information on
dose-response was sought out from a variety of sources,
especially other federal agencies (CDC, NIOSH, OSHA) ORD, and
Program offices at Headquarters.

Some key characteristics of each risk assessment step the
Regional office was responsible for were examined.  These
characteristics were the statutory basis for risk assessment,
whether standardized approaches were employed, and what key
sources of information and assistance were used.  The first
two characteristics give an idea of how institutionalized risk
assessment processes are in the Regional office.  A summary
matrix of our findings on these three characteristics for each
step of risk assessment is presented in Table 8.

-------
                          -22-
                          TABLE 6



                       Number of Steps
Portion of One



1 - 1.5 Steps



2 - 2.5 Steps



3 Steps
of Risk Assessment
No. of
Sections
(Absolute
Freq.)
6
7
7
6
Performed
Pet. of
Sections
(Relative
Freq.)
23.1
26.9
26.9
23.1
Pet. of Risk
Assessment Users
(Adjusted
Freq.)
23.1
26.9
26.9
23.1
                  TOTAL    26             100.0            100.0

-------
-1.5
                              -23-
                            FIGURE 3

               INDEX OF RISK ASSESSMENT ACTIVITY
                     (Standardized Scores)
 2.0
              1  Air and Toxics
 1.5          &  CDC
              t  CERCLA
              x  BSD
              -  OPM
              *  RCRA
 1.0          °  Water
 0.5

                                         O _ I
                                       o
                                     I

                                 t X


                               X
                           O „


-0.5
                     _ * o
                 I  _
               I
           * I

-1.0     *
     Lowest                                            Highest

                              Sections

-------
                                    -24-
                                  Table 7

                 Regional Risk Assessment Responsibilities
Category Label
Hazard ID

Dose Response
Evaluation

Exposure
Assessment

Risk
Characterization

TOTAL
NO. Of
Sections
(Absolute
Frequency)
14
0
13
10
37
Percent of
Sections
(Relative
Frequency)
37.8
0
35.2
27.0
100
Percent of Risk
Assessment Users
(Percent of
Population)
53.8
0
50.0
38.5
	 *
*Total exceeds 100% since sections may do more than 1 risk assessment step

-------
                                                  -25-
                                                Table  8
                              Key Characteristics of Risk Assessment Steps
Risk Assessment
Step
Key Characteristics
KEY (1)
ACTORS*
(2)
Statutory 1
BASIS
APPROACH 2
SOURCES/ (1)
ASSISTANCE*
(2)
Hazard
Identification
Region
States
STRONG
50% written protocol
Other Federal
Agencies
The Literature
Dose-Response
Assessment
HQ Program
or ORD
Office
CDC
-
-
Other Federal
Agencies
ORD
Exposure
Evaluation
Region
States
STRONG
Majority
written
protocols
Other Federal
Agencies
Field Studies
Risk
Characterization
Region
No one
WEAK
Few written
protocols
Other Federal
Agency experts
Literature
*  Key actors and sources ate named in order of occurrence.

-  Not applicable.  Regional office is not key actor in toxics evaluation.

1  Statutory bases for risk assessment were categorized as:
   Strong - required to perform risk assessment by law, regulation, and guidance
   Medium - required to perform risk assessment by regulation
   Weaker - directed to perform risk assessment by guidance

2  Standardized approaches for risk assessment were categorized as:
   Written protocol - documented standard operating procedures
   Some protocol - received training or followed standard procedures that weren't documented
   No protocol - procedures decided on a case-by-case basis

-------
                            -26-
We observed that procedures for risk assessment became more
institutionalized within the Regional office as regulations,
standards, and policy guidance were written directing all
parties on what such assessments should cover.  We found
procedures for analyzing health and environmental impacts
were more institutionalized in programs where there was a
strong statutory base for doing so (i.e., stated in law,
regulation, and guidance).

To better define these differences from one program to the
next, each Section's statutory base for risk assessment was
classified into strong, medium, or weak categories.  A
"strong" basis occurred where those Section's directions were
set out in law, regulation, and policy guidance; in these
cases risk assessment procedures were strongly institutiona-
lized.  A "medium" basis occurred where explicit regulation
and/or standards exist.  A "weak" basis meant that all that
existed was the general statement of law or policy; this can
be taken to indicate that little institutionalization had
occured.  We found the practice of performing hazard identi-
fication and exposure assessment generally had strong statutory
bases across the Region, while risk characterization had a
relatively weak basis.  This may in part explain why Regional
staff were not performing this step of risk assessment in a
"definitive" manner (see Risk Characterization discussion,
page 41).

The existence of standardized approaches for performing or
managing the performance of risk assessment was also explored.
Besides indicating institutionalization had taken place, the
degree of standardization can be taken to indicate the extent
to which risk assessments are done in a consistent fashion
within a program.  "Written protocols" included guidance or
standard operating procedures; they were often provided by
Headquarters Program offices or ORD. (Note: One section had
used the Agency's risk assessment guidelines for carcinogens.)
"Some protocol" included those who had received training or
followed a standard procedure that was not in written form.
"No protocol" described those Sections which performed or
managed risk assessments on a case-by-case basis.

Across the Region, standardized protocols were most abundant
for collecting the data used to perform exposure assessment.
However, these protocols were often not designed to collect
the types of data necessary to estimate impacts to human
receptor populations.  Approximately half of the Regional
staff had written protocols for hazard identification, and
only a small percentage had protocols for risk characterization.

-------
                            -27-
Participants were asked what key sources of  information and
assistance they  relied upon to do each step  of the risk
assessment process.  These responses were reduced to seven
general categories  (references named only once were excluded
to simplify the  classification scheme):

0 "External agencies/parties" included other federal agencies
(CDC, NIOSH, OSHA), States, permittees, permit applicants, and
manufacturers.

0 "The literature"  referred to primary and secondary literature
sources; the latter were more often utilized.

0 "Professional  experts" referred to toxicologists,
epidemiologists, and medical experts.

0 "Environmental Services Division", part of Region IV (but
located in Athens, Georgia) which serves as  the Region's
technical support arm.

0 "ORD Research  Offices" included the Offices of Research and
Development at Cincinnati, Research Triangle Park, the Carcinogen
Assessment Group, National Enforcement Investigations Center,
and the National Response Center.

0 "Computerized  data bases included Oil and  Hazardous Materials
Technical Assistance Data System (OHMTADS),  Storage and
Retrieval for Water Quality Data (STORET), the Center for
Disease Control's Hazardline,  and population data bases
available from the Office of Toxic Substances (OTS).

0 "Field studies" refer to monitoring, media (and occasionally
body burden)  sampling, and epidemiological investigations.

An interesting finding was that external agencies were
the most frequently named sources of information for all the
steps of risk assessment.  We  also noted that only a small
percent of staff were utilizing computerized data bases (e.g.,
10% used computerized sources  for hazard identification; even
fewer reported using those sources for dose/response or exposure
assessment)  to obtain the information needed to perform risk
assessment.   As  the quality of a risk assessment is highly
dependent on access to current toxicological and exposure
data, these findings suggest there may be a need to assess
current institutional arrangements and computerized systems
for information transfer to the Regions.

-------
                            -28-
  Hazard Identification

54% of the participants reported that their Section had
primary responsibility for the first step of the risk
assessment process, hazard identification (see Table 9).
Regional activities in hazard identification are initiated
in response to public concern, on the basis of a discharge
monitoring report or inspection, upon discovery of aban-
doned waste sites, or in emergency situations.  For some
TSCA, FIFRA, and Groundwater activities, hazard identifi-
cation was the terminal step of risk assessment.  In these
cases, institutional factors or the unavailability of
resources prevented the Region from estimating the risk to
public health.  TSCA staff indicated there was a need to do
more, e.g., estimate exposure from asbestos in schools.
Groundwater staff also noted many potential uses for risk
assessment, for example, to estimate impacts from contami-
nation of sole-source drinking water aquifers.

31% of the Sections responded that the Regional office had
secondary responsibility for hazard identification, with the
primary responsibility delegated to States (19%) or shared
with Headquarters Program offices (12%).  In cases where
Headquarters was designated as primarily responsible, it was
because some regulatory programs have pre-identified some chem-
icals in priority lists (e.g., Water's priority pollutants).
The Region then determines whether these are present at a
particular site.  These lists vary a great deal in context
and comprehensiveness from one program to the next.

All of those Sections with responsibilities in hazard iden-
tification had a statutory basis for doing so.  The largest
percentage (43%) fell into the "medium" category (See Tables
10 and 11).  71% of those with hazard identification
responsibilities had a standardized approach to perform or
manage risk assessments:  57% followed written protocols,
while 14% followed some protocol, usually learned through
training.  29% had no regular protocol they followed for
hazard identification.

Table 12 presents the sources of information and parties
assisting the Region in hazard identification.  54% of the
sections reported that parties external to the Regional
office assisted in and provided information used to perform
this step.  25% consulted the literature and computerized
data bases for preliminary information.  Also note that only
8% of the respondents utilize computerized data bases for
hazard identification.  Many reported that these data bases
were outdated, high in quantity but lacking in quality and
utility.  Others commented that staff as a whole could benefit
from a training course on use of computerized literature
search and data base capabilities.

-------
                               -29-
                            TABLE 9

               Hazard Identification Responsibility
HQ Media Office
or ORD

Region

States

Shared

External Party

Not Applicable
No. of
Sections
(Absolute
Freq.)
1
14
5
3
2
1
Pet. of
Sections
(Relative
Freq. )
3.8
53.8
19.2
11.5
7.8
3.9
Pet. of Risk
Assessment Users
(Adjusted
Freq.)
4.0
56.0
20.0
12.0
8.0
-
               TOTAL      26           100.0        100.0

-------
                                -30-



                              TABLE 10

                         Crosstabulation of


Hazard ID
Basis*


Stronger

Medium

Weaker

Not App.
COLUMN
TOTAL
Hazard Identification Responsibility
Hazard ID Responsibility

by


Not App HQ Media Region States
or ORD
3
21.4
1 6
100.0 42.9 60.
5
35.7 20.
1
100.0 20.
1 1 14
3.8 3.8 53.8 19.



3
0
1
0
1
0
5
2
Basis


Shared

1
33.3
1
33.3
1
33.3
«.

3
11.5



External ROW
Agency TOTAL
4
15.4
1 12
50.0 46.2
1 8
50.0 30.8
2
7.7
2 26
7.7 100.0
* Statutory bases for risk assessment were categorized as:
  Stronger - required to perform risk assessment by law, regulation, and
             guidance
  Medium   - required to perform risk assessment by regulation
  Weaker   - directed to perform risk assessment by guidance

-------
                                -31-


                              TABLE 11

                         Crosstabulation of

Hazard ID
Approach*
Written
Protocol
Some
Protocol
No
Protocol
Not App.
COLUMN
TOTAL
Hazard Identification Responsibility
Hazard ID Responsibility
by Approach

Not App HQ Media Region States Shared
or ORD
8
57.1 20
2
14.3 40
4
28.6 20
1 1 -
100.0 100.0 20
1 1 14
3.8 3.8 53.8 19
1 3
.0 100.0
2
.0
1
.0
1
.0
5 3
.2 11.5

External ROW
Agency TOTAL
12
46.2
1 5
50.0 19.2
5
19.2
1 4
50.0 15.4
2 26
7.7 100.0
* Standardized approaches for risk assessment were categorized as:
  Written protocol  - documented standard operating procedures
  Some protocol     - received training or followed standard procedures
                      that weren't documented
  No protocol       - directed to perform risk assessment by guidance

-------
         —32—






       Table 12



Sources of Information
HAZARD IDENTIFICATION
Rank
1
2
3
4
5
6
7
TOTAL
Category Label
External Agency/Party
The literature
Field Studies
Professional Experts
ORD Research Office
Conputer Data Bases
Env. Services Division
No. of
Sections
(Absolute
Frequency)
14
7
5
5
3
3
2
39
Percent of
Sections
(Relative
Frequency)
35.9
18.0
12.8
12.8
7.7
7.7
5.1
100
Percent of Risk
Assessment Users
(Percent of
Population)
53.8
26.9
19.2
19.2
11.5
7.7
7.7
	 *
DOSE RESPONSE
Rank

1
2
3
4
5
TOTAL
Category Label

External Agency/Party
ORD Research Office
Professional Experts
Env. Services Division
The literature
No. of
Sections
(Absolute
Frequency)
14
8
3
2
2
29
Percent of
Sections
(Relative
Frequency)
48.3
27.6
10.3
6.9
6.9
100
Percent of Risk
Assessment Users
(Percent of
Population)
53.8
30.8
11.5
7.7
7.7
*

-------
                                   -33-



                            Table 12 (cont'd)

                         Sources of Information
EXPOSURE ASSESSMENT
Rank
1
2
3
4
5
6
TOTAL
Category Label

External Agency/Party
Field studies
Models
ORD Research Office
Professional Experts
Env. Services Division
No. of
Sections
(Absolute
Frequency)
7
6
4
4
4
3
28
Percent of
Sections
(Relative
Frequency)
25.0
21.4
14.3
14.3
14.3
10.7
100
Percent of Risk
Assessment Users
(Percent of
Population)
26.9
23.1
15.4
15.4
15.4
11.5
	 *
RISK CHARACTERIZATION
Rank
1
2
3
TOTAL
Category Label

Professional Experts
No. of
Sections
(Absolute
Frequency)
3
External Agencies/Parties 2
The literature
2
7
Percent of
Sections
(Relative
Frequency)
43.0
28.5
28.5
100
Percent of Risk
Assessment Users
(Percent of
Population)
11.5
7.7
7.7
*TOTAL exceeds 100% since sections may have more than one source of
 information for each step

-------
                            -34-
  Dose-Response Assessment

None of the staff at the Regional office perform, or manage
the performance of, dose-response assessments for threshold
pollutants, or estimate the risk per unit exposure for non-
threshold pollutants such as carcinogens.  62% answered that
Headquarters Program or ORD Offices were responsible for
performing this step of the risk assessment process (See Table
13).  Although Regional staff do not perform dose-response
calculations, they often evaluate how toxic environmental
contaminants are and use "risk numbers" to write permits.  To
do so, they rely on estimates calculated in previous studies
by other agencies, EPA's health advisories, maximum concen-
tration levels, preliminary protective concentration levels,
and unit cancer risks.  Therefore, this step of risk assess-
ment as performed in Regional operations is more accurately
described as "toxics evaluation."

One might reason that since Regions recognized that dose-
response assessment was a Headquarters Media or ORD Office
responsibility, they would go to these sources most often for
dose-response information.  This was not the case; dose-
response information is often obtained from other sources.
Some reasons were the turn-around time to get this type of
information from Program offices, and the belief that ORD-CAG
assessments were only useful for rule-making and produced
estimates that were difficult to defend with CDC, environmental
groups, and responsible parties.

12% said the Centers for Disease Control (CDC) were responsible
for dose-response assessment; EPA has an interagency agreement
with CDC for health assessments at CERCLA sites.  CDC was
also named  frequently as a primary source of  information for
this step.  A general description of CDC's Health Assessment
protocol  is described in Table  14 in the context of EPA's
four-step  risk assessment process.  Note that the health
assessment  protocol provides  input to each step  of EPA's risk
assessment  process.

CDC and EPA currently have different "philosophies" regarding
use of risk assessment: While EPA is taking  steps to ensure
consistent  use of  risk  assessment across the  Agency, CDC has
stayed away from performing  quantitative risk assessments.
CDC's  reluctance to use quantitative risk  is  understandable,
given  the  history of  controversy which  has surrounded  its
use.   The  differences  between the  two  agencies  seems  in part
due to the different  weights scientists  assign  to various
types  of  scientific evidence (e.g.,  epidemiological studies,
animal data).  Also,  while  both CDC  and EPA  have responsi-
bilities  to protect public  health,  the  agencies  have  very
different agendas.  CDC is  a health  research agency;  its

-------
                               -35-
HQ Media Office
or ORD

States

Shared

CDC

Not Applicable
                             TABLE 13

                   Dose-Response Responsibility
NO. Of
Sections
(Absolute
Freq. )
16
1
1
3
5
Pet. of
Sections
(Relative
Freq. )
61.5
3.8
3.8
11.5
19.4
Pet. of Risk
Assessment
(Adjusted
Freq.)
76.1
4.8
4.8
14.3
-
Users







               TOTAL
26
100.0
100.0

-------
                                  -36-

                                Table 14

        A General Description of CDCs Health Assessment Protocol

"Hazard Identification"

1) consult desk references, including:
   Sittig - Handbook of Hazardous and Toxic Chemicals  contains information
   on 600 most commonly encountered chemicals;  extremely valuable in
   hazard identification.  Format includes information on routes of
   exposure, recommended clean-up actions;  lists EPA levels of exposure
   recommended to protect health and the environment; cites reference
   documents for more information.

2) Dreisbach - Handbook of Poisoning is also a useful hazard ID tool;
   lists acute and chronic health effects data.  Computer search e.g.,
   OHMTADS (Suggestion: OHMTADS needs updating; also have to weed
   through a lot of irrelevant materials in other data bases and printed
   resources to find data that may or may not be useful)

3) consult CDC staff (toxicologists and epidemiologists)

4) determine potential presence of receptor population if further
   investigation warranted at this point: 1) do site inspection to find
   bridge between substances of concern and receptor population

"Dose/Response"

1) EPA library and CDC staff gather available data on dose/response
   and CDC determines hazard

2) if worker safety and exposure problem; consult NIOSH; if possible
   food chain problem, consult FDA.

"Exposure Evaluation"

1) Review results of Environmental Media sampling

2) take body burden measurements, including blood and urine samples.
   Team of physicians and nurses obtain samples and administer questionnaire
   to population suspected of being at the highest risk.  If they find
   positive results, expand the scope of their sampling to determine
   how widespread exposure is.  This process can take as little as 24
   hours, depending on the nature of  the crisis.  Otherwise, process
   can take up to 2 months.

"Risk Characterization"

   Do not perform this step definitively.  Qualitative  characterization
   of risk  after implicit conparison  to other  risks people exposed to
   on a daily basis.   Do not quantify estimates of risk.

   Note:  Quotes are used because underlined terminology was imposed
   upon participant's  comments.

-------
                            -37-
involvement is greatest in the area of epidemiological re-
search.  EPA is an executive agency charged with regulating
human exposures to relatively ambiguous chemical agents in
the environment; its mandate is to protect the public from
both existing and potential health impacts.  Faced with these
regulatory responsibilities, EPA has adopted risk assessment
as a tool for targeting resources where health impacts are
greatest.  CDC and other health research agencies have
expressed concern that there are too many uncertainties in
the data to quantify risk levels from toxic chemical
exposures.

As in the hazard identification step of risk assessment,
external agencies and parties (CDC, NIOSH, OSHA) most fre-
quently (48%) provide staff with estimates of how toxic various
contaminants are to human health (See Table 12).  Another 28%
rely upon ORD Research Offices.  Although it appears Regional
staff access a number of different expert sources for dose-
response data, the sources often give conflicting opinions on
how toxic various chemical agents really are.  There is no
mechanism for achieving consensus among the experts; it is
up to the individual program as to what opinion carries more
weight.  In some CERCLA programs, given the interagency
agreement,  CDC gathers the available data on dose-response
and determines the "hazard" of a particular chemical or
chemical mixture.  In Water, a staff toxicologist makes
interpretations on dose-response data.  In TSCA, there was a
Regional expert to provide assistance in this area.  The
remaining programs consult various sources outside the
Regional office.  The output from this step of the site-
specific risk assessment process reflects this lack of
consensus among experts -  the results most likely confound
rather than simplify the data for decision makers.

  Exposure Assessment

50% of the Sections reported the Region had primary responsi-
bility for asessing exposure to environmental contaminants
(See Table 15).  In most cases, this involved estimating ambient
concentrations, with little or no orientation toward estimating
concentrations that the public might actually be exposed to.
This was not an omission on the part of staff; rather it was
the case that regulations and standards directed that measure-
ments be taken at the source of contamination, not the endpoint.
Many regulations in Air and Water programs simply require
detection of certain levels of contaminants in ambient air or
drinking water to establish a "risk" is present.  They do not
require the characterization of receptor populations.  This
is one area several staff felt needed improvement for conducting
site-specific risk assessments.

-------
                -38-
              TABLE 15



 Exposure Assessment Responsibility




Region
States
Shared
External Party
Not Applicable
No. of
Sections
(Absolute
Freq.)
13
4
2
1
6
Pet. of
Sections
(Relative
Freq.)
50.0
15.4
7.7
3.8
23.1
Pet. of Risk
Assessment Users
(Adjusted
Freq.)
65.0
20.0
10.0
5.0
-
TOTAL      26           100.0        100.0

-------
                            -39-
23% responded the Regions had secondary responsibility for
exposure assessment, with primary responsibility delegated
to States (15%) or shared with Headquarters (8%).  The 15%
included delegated programs in Air and construction grants
administered under the Water program.  In these cases, the
Regional office was given limited responsiblity to oversee
the performance of exposure assessments carried out by the
eight states in Region IV.

Tables 16 and 17 are crosstabulation tables that depict
how institutionalized the exposure assessment process is in
the Regional office.  The Regional office's statutory basis
for exposure assessment is evenly distributed among strong,
medium and weak categories.  Nonetheless, 85% of this sub-
group have written protocols for performing field studies and
estimating exposure, often in the form of standard operating
procedures.   This high percentage is due to the fact that many
of those who indicated responsibility for exposure assessment
provide the  field or model data on ambient or source concen-
trations that is the basis of the exposure estimate.  How to
obtain field data was highly institutionalized among the
technical support staff; most had standard operating procedures
they followed.  These procedures were often 5-10 years old,
and had not  been designed to meet the requirements of newer
regulatory programs that require establishing an endangerment
to human health, rather than simply a source of contamination.
To sum, it appears that although protocols were standardized,
the effectiveness of the protocols in measurement of impacts
to public health was low [Note: CDC is currently assisting BSD
in revamping some of their sampling strategies for public
health impacts].

Tables 16 and 17 also depict how institutionalized risk
assessment is in situations where States have primary respon-
sibility.  75% of this subgroup reported the statuatory basis
for States to perform exposure assessments came from regulations
and standards.  50% said States had a written set of procedures
from the Agency to help them do so.

A wide variety of sources of information and assistance for
performing exposure assessments were reported (See Table 12).
27% indicated external agencies and parties provided the most
assistance in exposure assessment.  23% said they obtained
exposure information through field studies, while 15% relied
on models.  Only 12% relied on BSD for this type of assistance;
this was often due to the Agency's policy of contracting out
for studies  of this nature.  It also could be explained by
the type of  assistance ESD provided - i.e., more geared toward
ambient environmental than public health impacts.

-------
                                -40-



                              TABLE 16

                         Crosstabulation of
            Exposure Evaluation Responsibility by Basis



                  Exposure Evaluation Responsibility
Exposure
Evaluation
Basis*


Stronger

Medium

Weaker

Not App.
COLUMN
TOTAL


Not App Region States

5
38.5
343
50.0 30.8 75.0
4 1
30.8 25.0
3
50.0
6 13 4
23.1 50.0 15.4


Shared External ROW
Agency TOTAL
1 6
100.0 23.1
10
38.5
2 7
100.0 - 26.9
3
11.5
2 1 26
7.7 3.8 100.0
* Statutory bases for risk assessment were categorized as:
  Stronger - required to perform risk assessment by law, regulation, and
             guidance
  Medium   - required to perform risk assessment by regulation
  Weaker   - directed to perform risk assessment by guidance

-------
                             -41-



                           TABLE 11

                      Crosstabulation of
        Exposure Evaluation Responsibility by Approach



               Exposure Evaluation Responsibility
Exposure
Evaluation
Approach*
Written
Protocol
Some
Protocol
No
Protocol

Not App.
COLUMN
TOTAL

Not App

—

—
1
16.7
5
83.3
6
23.1

Region
11
84.6
1
7.7
1
7.7

•"•
13
50.0

States Shared
2 2
50.0 100.0

— —
1
25.0
1
25.0
4 2
15.4 7.7

External ROW
Agency TOTAL
1 16
100.0 61.5
1
3.8
3
11.5
6
23.1
1 26
3.8 100.0
* Standardized approaches for risk assessment were categorized as:
  Written protocol  - documented standard operating procedures
  Some protocol     - received training or followed standard procedures
                      that weren't documented
  No protocol       - directed to perform risk assessment by guidance

-------
                            -42-
  Risk Characterization

This step of risk assessment is critical to the utility of
the entire process - risk characterization involves processing
and assembling the data collected on toxicity and exposure
(i.e., the magnitude of the impact and the probability of
occurrence) into one output, the final estimate of risk to
human health.  In addition, to be most useful to the decision
maker, the technical input to this step should be translated
into terms that laymen can understand.

Table 18 shows how responsibility is distributed among
the key actors in risk characterization.  39% of staff said
their Section had primary responsiblity for characterizing
the risks to human health in site-specific risk assessments.
This step clearly was less institutionalized than hazard
identification or exposure assessment (See Tables 19 and 20).
Within the responsible subgroup, 30% had a strong statutory
basis for performing risk characterization.  Another 50% said
they simply consulted their programs' standards, where levels
considered to constitute a risk to human health were already
prescribed.  With regards to standardized approaches, 50% had
written protocols that told them what a risk characterization
should include; 30% had no protocol for performing this step.

34% of Sections oversaw or interpreted risk characterization
performed by States, other federal agencies, Headquarters
Program offices, or ORD.  The remaining 27% indicated risk
characterization was not done definitively in site-specific
assessments.  In these cases, field data on observed concen-
tration levels or estimates of ambient exposure were presented
alongside toxicity data based on epidemiological studies or
animal tests.  There was generally no effort made to combine
these data to estimate the risk to nearby human populations.
Staff were reluctant to perform this step due to previously
mentioned limitations in the exposure data, a lack of consen-
sus on the human toxicity of complex mixtures, and the percep-
tion that this step was not institutionalized as part of the
analyst's responsibilities.

Where  risk characterization relates  to Regional operations,
the estimates of risk performed, managed, or overseen are
expressed  in both quantitative and qualitative terms.  Ad-
justing for  those who weren't  involved  in  risk characteri-
zation, 33%  of the Sections answered  that  human health risks
are estimated,both qualitatively and  quantitatively  (See
Table  21).   When quantified, estimates  are most often
expressed  as excess cancer deaths per million population;
receptor  populations  at  the site are  often not characterized.

-------
                               -43-
                             TABLE 18

               Risk Characterization Responsibility
HQ Media Office
or ORD

Region

States

Shared

External Party

Not Applicable
NO. Of
Sections
(Absolute
Freq.)
5
10
2
1
1
7
Pet. of
Sections
(Relative
Freq. )
19.2
38.5
7.8
3.8
3.8
26.9
Pet. of Risk
Assessment Users
(Adjusted
Freq. )
26.3
52.6
10.5
5.3
5.3
-
               TOTAL      26           100.0        100.0

-------
                                  -44-
                                TABLE 19

                           Crosstabulation of
             Risk Characterization Responsibility by Basis
                   Risk Characterization Responsibility
Risk
Characterizat ion
Basis
              Not App     HQ Media    Region   States    Shared    External    ROW
                           or ORD                                   Agency    TOTAL

Stronger

Medium

Weaker

Not App.
COLUMN
TOTAL

™
1
14.3
1
14.3
5
71.4
7
26.9

™"
1
20.0
1
20.0
3
60.0
5
19.2
3
30.0
5
50.0
2
20.0

—
10
38.5

•"*
1
50.0

"•
1
50.0
2
7.7
3
11.5
1 1 10
100.0 100.0 38.5
4
15.4
9
34.6
1 1 26
3.8 3.8 100.0
  * Statutory bases for risk assessment were categorized as:
    Stronger - required to perform risk assessment by law, regulation, and
               guidance
    Medium   - required to perform risk assessment by regulation
    Weaker   - directed to perform risk assessment by guidance

-------
                                  -45-



                                TABLE 20

                           Crosstabulation of
            Risk Characterization Responsibility by Approach
                    Risk Characterization Responsibility

Risk
Characterization
Approach*
            Not App    HQ Media     Region   States    Shared    External     ROW
                         or ORD                                    Agency     TOTAL
Written 1
Protocol 14.3
Some
Protocol -
No
Protocol
6
Not App. 85.7
COLUMN 7
TOTAL 26.9
5
50.0
1
10.0
3
30.0
5 1
100.0 10.0
5 10
19.2 38.5
1 1
50.0 100.0
1
50.0
_ _
- -
2 1
7.7 3.8
8
30.8
2
7.7
3
11.5
1 13
100.0 50.0
1 26
3.8 100.0
  * Standardized approaches for risk assessment were categorized as:
    Written protocol  - documented standard operating procedures
    Some protocol     - received training or followed standard procedures
                        that weren't documented
    No protocol       - directed to perform risk assessment by guidance

-------
                               -46-
                             TABLE 21

                 Nature of Risk Characterization:
Quantitative vs
No. of
Sections

2
3
5
5
11
Qualitative
Pet. of
Sections
(Relative
Freq. )
7.8
11.5
19.2
19.2
42.3
Pet. of Risk
Assessment Users
(Adjusted
Freq. )
13.4
20.0
33.3
33.3
-
Qualitative

Quantitative

Both Quant. &
 Qual.

Input Quant.->
 Output Qual.

Not Applicable


                  TOTAL    26             100.0            100.0

-------
                            -47-
Another 33% indicated they received quantitative data on
toxicity and exposure and translated these into qualitative
expressions of risk for use by decision makers and the public.
This was done because quantified estimates of cancer deaths
were viewed as conservative; they over-estimate the actual risk
present at specific sites.  Regional staff have learned these
numbers often put them in a bind with both the affected public
and industry.  These "worst case" numbers often trigger
over reactions by the affected parties, who are basically
averse to accepting any additional risk.  On the other hand,
the numbers are challenged by industry as not being "realistic"
estimates of risk.  The production of upper bound estimates
of risk only partially serves the needs of decision makers,
who would like information on the lower to upper bound range
of risks a particular community is facing in an exposure
situation.

Compared to the other steps of the risk assessment process,
sources of information and assistance for risk characterization
were scarce (See- Table 12).  Where assistance is available,
it is usually from medical experts at CDC.  Although CDC has
scientific expertise in public health disciplines, again it
should be noted that they do not use quantitative risk as-
sessment, and are at odds with methods employed as a matter
of policy at EPA.  In addition, CDC is contracted only to
provide assistance in CERCLA actions.

The paucity of sources and assistance available for performing
risk characterization is due in part to the high number of
staff who responded that no one was responsible for performing
this step of risk assessment in their program.  If no one was
doing it, there was no reason to ask for their sources.  It
also may be indicative of a need for more assistance in risk
characterization.
The Site-Specific Risk Assessment Process; A Summary Model

Based on the previous steps, it is now possible to describe
the process as a whole.  Risk assessment as currently practiced
in the context of Region IV operations is explained through
the process model in Figure 4.  The solid lines represent the
actual flow of information during the process.  Dotted lines
are used to represent how the process could proceed to more
effectively utilize available data and expert opinion.

Risk assessment is typically performed at specific sites
in the Region, rather than on a specific chemical as in
regulatory development.  The site-specific risk assessment
process is initiated in response to public concern, or, less
frequently, as a result of state and local monitoring activities

-------
                                       -48-
                       REGION IV RISK ASSESSMENT OPERATIONS
                 HAZARD
              IDENTIFICATION
               TOXICS
              EVALUATION
                 EXPOSURE
                 EVALUATION
    RISK
CHARACTER! ZATION
MONITORING
PUBLIC
CONCERN
                   HQ
REGION

PROGRAM

OFFICES
                  States
 ORD

  HQ

 CDC

NIOSH

 OSHA
                                REGION

                                PROGRAM

                                OFFICES
                                                  GW AIR WATER
                                                                  ->

                                                                  ->
  REGION

  PROGRAM

  OFFICES
                                MANAGEMENT
                                                                           ECONOMIC

                                                                           POLITICAL
                                                      -- CULTURAL

                                                         SOCIAL
                                     FIGURE 4

-------
                            -49-
Th e problem is often a mixture of potentially hazardous
chemicals discovered in a particular site or in effluent from
that site.  The first step is hazard identification.  It
involves weighing the available evidence to determine whether
substances at a site present a hazard to the environment or
human health.  The general nature of the hazard is investigated
in terms of possible adverse environmental and health impacts,
e.g., damage to ecosystems, impacts on water quality, cancer
in humans.  Hazard identification is a preliminary assessment
that helps one decide whether further investigation is needed,
and if so, the extent of that investigation at a specific
site.  The output is not a list of potential human health
effects as was found in regulatory development settings, but
rather a scoping of potential adverse environmental and human
health impacts for further investigation.  The key actor in
hazard identification at specific sites is the Regional
office; Regions are primarily responsible for managing or
overviewing assessments performed by contractors, permittees,
and other external agencies.  Regions often share this
responsibility for hazard identification with States, and use
regulations and guidance from Headquarters to manage this step.

After hazard identification is performed, and the judgment
made that further study is warranted, Regional staff begin to
seek the answers to two questions:  1) how toxic are the chemical
actors at a specific site (i.e., toxics evaluation) and 2)
how widespread is the contamination (i.e., exposure evaluation).
These steps are often performed simultaneously rather than
sequentially.  Dose-response information already in the form
of acceptable daily intakes and unit cancer risks are gathered
from EPA Headquarters and other federal agencies and listed
in the assessment.  Regions don't themselves perform the dose
response calculations; rather, they rely on opinions and
calculations of scientists from ORD and CDC.  This is the
"toxics evaluation" step in Figure 4.

Exposure evaluation is done by contractors under the super-
vision of the Program office, and sometimes with assistance
from BSD.  Data from models or field studies are used to
estimate concentrations of substances in air, water, or soil.
Data collected often document the concentration of toxics at
the source rather than the concentration receptor populations
might receive.  Media are often assessed independently due to
organizational structure, and the output of exposure assessment
is often a table of concentrations found in the different media
at various points of the site.

The risk characterization step is not done definitively
in many site-specific risk assessments.  This is due to the
large uncertainties in defining what the risk to human health

-------
                            -50-


might be, based on the output produced in the toxicity and
exposure evaluations described above.  When a risk charac-
terization is made, it is usually a qualitative judgment
based on either the assessor's or a recognized expert's
opinion.  The resulting assessment contains many independent
outputs and significant, though freqently implicit, uncer-
tainty.

The dotted lines in Figure 4 show how the current site-
specific risk assessment process could proceed to increase
the use of scientific judgment and reduce the barrage of
inputs a decision maker is currently faced with.  The goal of
this approach would be to utilize the best scientific and
technical expertise available to the Region to produce a
consensus on what the risk to human health is at a specific
site.

There are several techniques used among professionals
to reach consensus on specific issues pertaining to the inter-
pretation of scientific data.  A panel of senior staff with
programmatic perspective and technical expertise could serve
this function.  This has a precedent - Headquarters program
offices initiated such a group, The Risk Assessment Forum,
to discuss risk assessment issues in regulatory development.
Delphi or other analytic techniques could be used in cases
where certain programs or personalities dominate the consen-
sus-building process (Porter et al., 1980)

Staff Recommendations for Improving the Process

Staff offered numerous suggestions for improving the
scientific quality of site-specific risk assessments (see
Table 22).  These suggestions centered around five major
points.  58% of the Sections reported a need for improved
coordination and information exchange between key actors in
the risk assessment process.  They felt the quality of a risk
assessment would increase with the availability of a wide range
of multi-media specialists (e.g., toxicologists, statisticians,
exposure modelers).  As these types of staff are organization-
ally scattered across Regional offices, at Headquarters, or at
CDC, risk assessments require greater intra- and interagency
coordination to be done effectively.

27% of Sections noted that consistent methods for performing
site-specific risk assessments should be implemented, especially
in the area of exposure assessment.  There have been occasions
where repetitive studies of the same site were undertaken by
different program offices to arrive at the same endpoints,
i.e., impacts to human health and the environment.  Application
of consistent methods and elimination of conflicting guidance
could help alleviate the need for repeat studies at the same
site.

-------
                                    -51-
                                  TABLE 22

                           Staff Reconmendations
Category Label

Consistency
Techniques
'Strong" Dose Response
Information
Health/Envi ronmental
Balance
Coordination and
No. of
Sections
(Absolute
Frequency)
7
7
5
3
15
Percent of
Sections
(Relative
Frequency)
18.9
18.9
13.5
8.2
40.5
Percent of Risk
Assessment Users
(Percent of
Population)
26.9
26.9
19.2
11.5
57.7
  Information
Exchange
                                                                	*
TOTAL
37
100.0
*Total exceeds 100% since sections may have given more than 1 recommendation.

-------
                            -52-
Another 27% recognized the need for more information on
techniques useful to perform exposure assessment and risk
characterization.  In the area of exposure assessment,  help is
needed to set up monitoring programs and sampling strategies
to more accurately estimate impacts to human receptor popu-
lations.  In risk characterization, help is needed to improve
our skills in translating technical information into laymen's
terms which can be made explicit to the public.

19% of Sections emphasized a need for stronger and readily
available dose response information in the form of unit
cancer risks and acceptable daily intakes.  A centralized
data base within the Agency should be established to provide
this sort of up-to-date information on health effects.

12% indicated a need for risk assessment strategies that put
equal weight on assessment of both health and environmental
impacts.  For example, current hazard ranking systems for
evaluation of toxic waste dumps put greater weight on human
health than environmental impacts.  These staff warned of the
trade-offs we are making when we try to separate health from
environmental impacts for purposes of prioritization.  Worse
problems arise over time as populations shift or substances
migrate, and impacts once considered as "environmental" can
become serious and costly health problems.

-------
                          CHAPTER III

               CONCLUSIONS AND RECOMMENDATIONS

A. Conclusions

This study examined the nature and use of risk assessment in
one of EPA's ten Regional offices.  Although we focused
primarily on the practice of human health risk assessment
from toxic chemical exposures, we found risk assessments in
Regional operations have a more general focus on both en-
vironmental and human health impacts than do risk assessments
performed in regulatory development.

Regional staff involvement in risk assessment can be in any
one of three roles: assessment, management or oversight of
the risk assessment process.  The majority of Regional program
activities have a direct impact on human health.  There is a
strong association between activities directly impacting
human health and risk assessment use.

Risk assessments in Regional operations are done at specific
sites and often involve complex mixtures of toxic chemicals.
This setting is very different from the setting in which
regulatory development occurs, where risk assessment is most
often performed on specific chemicals to set nationwide
standards.  It necessitates the site-specific risk assessment
be more focused on exposure rather than dose/response assess-
ment, which is typically performed at the national level.
Site specific assessments also require greater sensitivity to
local community concerns.

The site-specific risk assessment process involves some or
all of the following steps: (1) hazard identification,
(2) toxics evaluation (based on previously done dose-response
assessments), (3) exposure evaluation, and (4) risk charac-
terization.  This process is wider in scope and less targeted
towards specific health endpoints than risk assessments
performed in regulatory development.  Regional risk assessment
responsibilities are frequently shared with Headquarters
offices, States, and external agencies.  The Regional office
itself clearly has more responsibility for two steps of the
risk assessment process - hazard identification and exposure
assessment.  Regions rely upon a variety of sources for dose/
response data, and risk characterization is not performed
"definitively" in many assessments.

-------
                             -54-
Across the Regional programs, standardized protocols
are more abundant for collecting the data necessary to
perform exposure assessment than for any other step of risk
assessment.  Many of these data collection protocols are
inadequately designed to assess impacts on the receptor
populations exposed to toxic chemicals.  This in turn affects
the quality of risk characterization, and as a result, the
risk characterization step is often not done definitively.

The level of risk assessment activity varies across Regional
programs; currently CERCLA programs, and technical support
staff to CERCLA, show the highest levels of risk assessment
activity.  These assessments are typically done to estimate the
potential endangerment to human health and the environment at
Superfund toxic waste sites.  CERCLA programs have received
endangerment assessment guidance from the Office of Solid
Waste and Emergency Response directing them to incorporate
risk assessment into their remedial investigation/feasibility
studies.

Risk assessments have been performed in Air programs upon the
request of States and Regional programs to estimate population
exposures to ambient levels of toxic chemicals.  The recent
RCRA reauthorization also calls for Regions/States to manage
exposure assessments performed by permittees at hazardous
waste facilities.  Furthermore, there is potential for Regional
programs to become more active in risk assessment in the
future, due to (1) EPA's adoption of an explicit risk assessment
policy, and (2) trends in legislation that have moved from
regulating risk to zero towards achievement of a balance
between the risks and benefits of chemicals to society.

The Department of Health and Human Services (Centers for
Disease Control) plays a major role in risk assessments
performed in CERCLA programs, due to the interagency agreement
between CDC and EPA for health assessments at CERCLA sites.
EPA and CDC currently have different policies on the use of
risk assessment; while EPA is taking steps to ensure consistent
use of  risk assessment to the extent possible across its
programs, CDC has stayed out of standardizing risk assessment
practice.  These policy and cultural differences between the
two agencies sometimes result in conflicting opinions between
the two  agencies regarding the health risk posed by toxic
chemical exposures at certain sites.

The risk assessment process  requires a high degree of coordi-
nation  among Regional program offices, and between the Region,
States,  and other federal agencies.  This  is due to the fact

-------
                             -55-
that while media are confined within regulatory programs,
site-specific occurrences of chemicals are not.  Furthermore,
individual organizational units seldom perform the entire
proecess.  The quality of risk assessment depends on the
collective effort of several parts of the Agency.

Chemicals often impact more than one medium or transfer
across several media at a particular site, and thus require a
risk assessment involving a wide range of expertise.  Regional
staff with expertise or experience in assessing health and
environmental risks are scattered across Regional programs.
However, EPA's organizational structure and the "turf con-
siderations that naturally develop in decentralized organi-
zations do present barriers to the Regional office's effec-
tiveness in assessing health risks associated with intermedia
toxic chemical exposures.  A large number of staff reported a
need for improved coordination and information exchange
between key actors in risk asessment, including Regional
programs, the Region and Headquarters, States, and other
federal agencies.

B. Recommendations

0    The Regional office currently has no mechanism to cut
     across organizational boundaries and integrate the
     expertise needed to perform risk assessments involving
     several media.  A multidisciplinary, interprogram team
     of staff with expertise in specific areas of risk assess-
     ment and programmatic perspective could improve the
     consistency and scientific quality of site-specific risk
     assessment practices.  This Team would act as an informa-
     tion network and "Regional Clearinghouse" for assessments
     evaluating exposure from several different routes.  The
     team could also perform an external liaison function to
     assist in and coordinate the risk assessment process with
     Headquarters, States, and other external agencies.

0    The role of the Centers for Disease Control (The Agency
     for Toxic Substances and Disease Registry) in the site-
     specific risk assessment is in need of further clarifi-
     cation.  CDC can provide valuable expertise to the
     process, but it is currently unclear to many Senior
     managers how CDC's opinion should be incorporated into
     the risk assessment.  An interagency policy group should
     be formed to take a look at the philosophical differences
     and working relations between the two Agencies.

-------
                          -56-
Exposure assessment protocols and guidance are often in-
adequate to assess impacts on receptor populations in
cases where health risk is an issue.  A flexible set of
"how to" guidelines on exposure assessment could help to
improve the quality and consistency of this process.  A
Regional workgroup with experience in this area, in con-
junction with ORD and CDC, would be one way to pool these
Agencies' experience and expertise to develop the guidelines
while minimizing the chance of "re-inventing the wheel."

A centralized data base should be developed and maintained
that contains up-to-date toxicity information and names of
contact persons in the Agency who can assist in using
these numbers.  The Regions depend on Headquarters and
other federal agencies for calculations of acceptable
daily intakes and unit cancer risks from toxic chemicals.
This information is acquired from a variety of sources and
is often confusing to staff with little background in
toxicology.  Conflicting information and the lag time to
obtain this information present barriers, especially since
sites often contain complex mixtures of chemicals to be
researched.  Currently available data bases should also be
made "user-friendly" and updated.

Training programs tailored to site-specific risk assessment
and Regional needs should be developed; to date, most
training and information Regions receive is geared toward
scientific audiences who perform risk assessment to develop
regulations and set standards.  Areas where training could
be beneficial include the utilization of computerized systems
and available data bases  (some currently available systems
seem to be "missing the mark"), interpretation of toxico-
logical data, how to perform exposure assessment, and
Agency risk assessment policy.

Staff made several suggestions to improve risk assessment
in Regional operations.  We recommend all their suggestions
be considered - the failures of organizations to achieve
their objectives are too often due to lack of communication
between the staff who develop policy and those who must
implement  it.

These suggestions  included:

-  improve  coordination and information exchange between
   Headquarters, Regions,  States, and the external agencies
   which  have key roles in  the risk assessment process;

-------
                       -57-


facilitate the use of consistent methods for performing
site-specific risk assessment, especially in the area of
exposure assessment;

provide training for Regional staff on how to set up
monitoring programs and sampling strategies for
estimating impacts to receptor populations; on risk
characterization; and on the communication of risk to
the public;

make current Agency knowledge on dose-response relationships
and toxicity available and accessible to Regional staff

give serious consideration to the trade-offs we implicitly
make when greater weight is put on human health than environ-
mental impacts from hazardous waste contamination.  The dif-
ference between "environmental" and "health" impacts are often
only a matter of time; although this type of prioritization
made seem prudent in the short-term, worse problems will
inevitably arise over time as populations shift and
substances migrate.

-------
APPENDICES

-------
                                         APPENDIX A-l

                                       REGION IV ORGANIZATION
                                       REGIONAL ADMINISTRATOR
                                       Charles R. Jeter - 4727
                                    DEPUTY REGIONAL ADMINISTRATOR
                                        John A. Little - 4727
   OFFICE OF REGIONAL COUNSEL
    James H. Sargent - 2335
  OFFICE OF ASSISTANT REGIONAL
ADMINISTRATION FOR POLICY & MGHT.
    Howard D. Zeller - 3476
    ENVIRONMENTAL SERVICES
           DIVISION
    James H. Finger - 3136
                                                            Match 1984

                                       Atlanta Ccnraercial Prefix - 881
                                       FTS  Prefix                - 257
                                       Athens Confrereial Prefix  - 546
                                       FTS  Prefix                - 250
                               OFFICE OF CONGRESSIONAL fc
                                            AFFAIRS
                                        Vacant -  2013
                  EXTERNAL
                                                                                                                           I
                                                                                                                          VJl
                                                                                                                          VD
                                                                                                                           I
  frRTER MANAGEMENT
      DIVISION
Paul J. Traina - 4450
AIR fc WASTE MANAGEMENT
       DIVISION
 Ton W. Devine - 3454

-------
                                 OFFICE OF ARA FOR POLICY AND MANAGEMENT
                                          ASSISTANT REGIONAL
                                  ADMINISTRATOR PQR POLICY fc MANAGEMENT
                                        toward D. Zeller - 3476
  ENVIRONMENTAL  *
ASSESSMENT BRANCH
Edward T.  Heinen
      3776
                                                            DEPUTY EEO OFFICER
                                                        Otis Johnson, Jr. - 79OO
  PERSONNEL AND
ORGANIZATION BRANCH
William A.  Waldrop
       3485
       FEDERAL FACILITIES
       COORDINATOR
       Arthur G.  Linton
             3776
POLICY AND PROGRAM
 EVALUATION BRANCH
J. R. Franzmathes
       2140
                              BUDGET OFFICE
                              John B. Gardner
                                   2374
       ENVIRONMENTAL
       REVIEW SECTION
       Sheppard N. Moore
            7901
MANAGEMENT BRANCH
   Robert Roth
     3274
                                                                                                                 o
                                                                                                                 I
                              Clifford R.  Hoqatt-354'
                                                                                     ADP  MANAGEMENT SECTION
                                                                                     Clifford  R.  Davis -  2316
                                                           FINANCIAL MGMT.  SECTION
                                                           William  H. McBride  - 3402
       NEPA COMPLIANCE
       SECTION
       Robert B.  Howard
           3776
                                                             FIN.MGMT.UNIT
                                                             Vacant - 3277
                                                                                       GRANTS AIM. UNIT
                                                                                       Donald H. Taylor -2152

-------
                                           AIR AND WASTE  MANAGEMENT DIVISION
                                                        DIRECTOR
                                                    Thomas W.  Devlne
                                                    DEPUTY DIRECTOR
                                                    George L.  liar low
                                                      FTS  257-3454
EMERGENCY i REMEDIAL
  RESPONSE BRANCH
      Al Smith
 *  FTS 257-3931
RESIDUALS MANAGEMENT
       BRANCH
   Jin Scarbrough
*   FTS 257-3016
      SITE SCREENING
      & ENCR.  SECT.
      Wayne Hathis
      FTS 257-2234
      EMMGENCY  RESPONSL
      & CONTROL  SECT.
      George  Moein
      FTS  257-3931
      REMEDIAL  ACTION
      SECTION
      Allen Bartlett
      FTS 257-2643
     WASTE  PLANNING
     SECTION
     John Dickinson
     FTS 257-3016
     WASTE  ENGR.
     SECTION
     Doug McCurry
     FTS 257-3433
     WASTE  COMPLIANCE
     SECTION
     Jim  Holdaway
     FTS  257-4552
PESTICIDES & TOXIC
      BRANCH
    Roy Clark
*  FTS 257-3222
    PESTICIDES SECT.
    Kent Williams
    FTS 257-3222
    TOXICS SECTION
    Ralph Jennings
    FTS 257-3864
                                                                                     1
 AIR MANAGEMENT
     BRANCH
James T. Uilburn
* FTS 257-4298

AIR PLANNING
SECTION
Richard Schu'.t
FTS-257-3286


                            AIR ENGINEERING
                            SECTION
                            Richard S. DLbose
                            FTS 257-7654
                                                                                   AIR COMPLIANCE
                                                                                   SECTION
                                                                                   Tommie Glbbs
                                                                                   FTS 257-4298
                           STATIONARY SOUR!
                           PLANNING UNIT
                           Archie  Lee
                           FTS  257-3286
                                                                         h-'
                                                                         I
                           MOBILE SOURCE
                           PLANNING UNIT
                           Ron Mcllenry
                           FTS 257-2864
                                                                                NORTHERN
                                                                                COMPLIANCE UNIT
                                                                                Floyd Ledbetter
                                                                                FTS 257-4298
                                                                                SOUTHERN
                                                                                COMPLIANCE UNIT
                                                                                Jesse  Baskervill
                                                                                FTS  257-4253

-------
                                             WATER MANAGEMENT DIVISION
              I"
FACILITIES PERFORMANCE BRANCH
     John Marlar - 4793  *
     WATER QUALITY SECT.
     James Kutzman - 3116I
     PERMITS SECTION
     Jim Patrick - 3012
           NORTH AREA UNIT
           Robert Wboten - 2156
           SOUTH AREA UNIT
           Robert McGhee - 2913
     MUNICIPAL OPERATIONS SECT.
     David Olson - 2211
     INDUSTRIAL OPERATIONS SECT.
     Gilbert Wallace - 3973
                                                      DIRECTOR
                                                Paul J. Traina -.4450
                                                   DEPUTY DIRECTOR
                                                Frank  J.  Silva -  4450
FACILITIES CONSTRUCTION BRANCH
     Harold Hopkins - 2207  *
         | AL/GA UNIT - Curt Pehn - 2017
   :i
         JJFL/MS UNIT - Peter McGarry - 7428     |
         j KY/NC UNIT - William Taylor - 2328   j

         'j SC/TN UNIT - Douglas Lanktord ~ 3973 j
                                                      1
                                        WATER SUPPLY BRANCH *
                                       Donald Guinyard - 3781
OFFICE OF APPLIED
TECHNOLOGY
Robert Freeman, Jr. - 4180
       SOUTH AREA GRANTS
       MANAGEMENT SECTION
       John Hagan - 3633
                                                DRINKING WATER SECTION
                                                Robert Jourdan - 3781
                                        GROUNDWATER SECTION
                                        David  Peacock - 3866
ro
I
       NORTH AREA GRANTS
       MANAGEMENT SECTION
       Dan Ahem - 2005

-------
                                           ENVIRONMENTAL SERVICES DIVISION
                                                  (Athens, Georgia)
                                                      DIRECTOR
                                               Jones H. Finger - 9136
                                                   DEPUTY DIRECTOR
                                                Billy H. Adams - 3421
      ANALYTICAL SUPPORT BRANCH
         Bobby Carroll - 3112  *
               I
  CHEMISTRY SECT.
Ton Bennett - 3111
LABORATORY EVALUATION
AND QUALITY ASSURANCE
       SECTION
  W&de Knight - 3165
                                                          ECOLOGICAL SUPPORT BRANCH
                                                               Lee Tebo - 2294 *
      AIR fc WATER
    MONITORING SECT.
Doyle T. Brittian - 3197
                                                                                                         1
  ENVIRONMENTAL
  BIOLOGY SECT.
Russ Todd - 2294
                                                                                                                       LO
                                                                                                                        I
                                             ENGINNERING SUPPORT BRANCH
                                               Michael Carter - 3117 *
                                                         I
HAZARDOUS HASTE SECT.
  Doug Lair - 3117
                                                                              1
                                                                    MUNICIPAL fc  INDUSTRIAL
                                                                          SECTION
                                                                     Charles  Sweatt  -  3351
                                                                 r
                                                                              1
                                                           MUNICIPAL UNIT
                                                         Ron Barrow -  3588
                                                                   1
                                                           INDUSTRIAL UNIT
                                                          Steve Hall - 3588

-------
                        Appendix A-2


In this survey, standardized measurement and procedures
were followed to (1) enhance the reliability of observation;
(2) facilitate replication studies in other Regional offices
and (3) permit statistical analysis of data collected.
Qualitative methods were employed to (1) capture the frame
of reference and definition of the situation of a given
participant to avoid instrumentation artifacts of standardized
measurement procedures, and to (2) permit a more detailed
examination of the risk assessment process (McClintock,
1979).

Our survey protocol employed an interviewer to obtain
answers for the same set of questions across each regulatory
program under review.  In addition, the interviewer acted
as participant-observer; i.e., felt free to exercise judgment
to encourage ad hoc comments from program staff on any
aspects of the risk assessment process in Regional operations.
This gave our measurements greater sensitivity, in that we
were able to record strong and weak responses, qualify re-
sponses as needed, and be more responsive to difficulties the
participants had in interpreting the questions.

Several organizational units (i.e., Sections) were inter-
viewed from each Program Office.  These sections served later
as units of analysis.  The questions asked were close-ended,
so that responses could be aggregated for further analysis.
The questions covered those characteristics we felt would
define the nature and function of risk assessments performed
in Regional operations.  These questions were of four major
types:

(1) relevant background on Regional programs - for example,
    the major decisions made at the Regional level that affect
    human health;

(2) the incidence and frequency of use of risk assessment in
    Regional programs;

(3) the characteristics of risk assessments used or performed
    by Regional Program Offices - for example, their components
    and established procedures employed; and,

(4) the value of risk assessment as an input to Regional
    decision making and suggestions for improvement.

-------
                            -65-
Our survey strategy was designed to facilitate our getting
the greatest amount of information possible on risk
assessment activities by sidestepping those in Regional
office unlikely to use it.  As a result, the protocol was
administered across six Program offices: Air (CAA), Toxics
(TSCA and FIFRA), Water, CERCLA, RCRA, and the Office of
Policy and Management (0PM).  [Note: 0PM administers several
of the smaller Regional programs.]

Twenty-six sections from these offices and the Environ-
mental Services Division, which provides technical support
to these offices, were surveyed (see Table 1).  These
twenty-six sections composed the entire population of staff
believed to be possible users of risk assessment techniques
at the Regional office.  Questions were directed at overall
Section activities, not an individual's activities.  In
this way, we were able to census all the activities of the
population subset rather than those of random individuals.
The population subset of possible risk assessment users was
extracted from the total population at the Regional office
on the basis of one criterion:  Sections were excluded from
the survey whose operations clearly had no reason to assess
human health impacts.  The Regional office at the time of the
study was organized into 34 Sections with a total of 567
full-time employees.  Application of the exclusion criterion
resulted in elimination of eight Sections from the survey.
The Sections eliminated were involved in finance, budget,
personnel, and program evaluation activities.

There are many factors that can threaten the validity of
survey results.  The interviewer and the population surveyed
may possess certain biases that can strongly affect the
information exchanged between them.  In addition, we recog-
nize the interviewer's proclivity for encouraging spontaneous
discussion on risk assessment during the interview was
likely to introduce some bias.

To improve the validity and reliability of the protocol,  it
was pre-tested on several Agency staff prior to general
administration. Still we must acknowledge that the mere
fact a survey was being conducted on the use of risk assess-
ment may have affected the responses we obtained.  The
existence of a survey composed  entirely of questions on  the
use of risk assessment could easily challenge an  answer  of
"no use" in some situations.  Also, subtle aspects of  the
way in which questions are phrased  and  responses  elicited
have impacts on  the  judgments people make in  answering  the

-------
                              -66-
questions.  In this case, a survey on risk assessment could
permanently alter the opinions of the respondents about the
use of this approach.  This result, not totally unintended,
is significant since the population we surveyed was the
professional staff who are instrumental in implementing
EPA's new risk assessment policy.  So, although some may
interpret the survey as "corrupting the sample", we felt
that any feedback effects were of a positive nature and
served as an opportunity to discuss with staff Agency risk
assessment policy.

Each interview involved three components: an introduction,
definition of relevant terms, and a set of twelve questions
(see Questionnaire) which explored the four points listed
on page 59.   The construction of each of these is discussed
in the following paragraphs.  The introduction to the
survey was designed to address several points.  Above all,
we wanted to put participants at ease, given that institu-
tional forces might be at work.  This was accomplished by
attempting to anticipate the major concerns participants
may of had over providing information.  These concerns were
identified through brainstorming sessions with veteran
Agency staff members and the pre-test.  Out of these dis-
cussions it was decided the introduction should (1)
establish that the interviewer was from the Regional office,
but not pinpoint a specific regulatory program, (2) let
participants know information was desired on activities,
not individuals, and confidential; and (3) assure partici-
pants we had no preformed notions on whether they should or
shouldn't be doing risk assessments.

Clear definition of relevant terms, i.e., what a risk
assessment is, was important for several reasons.  First,
our survey results could have measured a different activity
every time a Program offices' terminology changed.  Second,
it insured comparability with Agency risk assessment studies
done to date, because definitions were the same.  Third, an
underlying value was the education of staff as to what
EPA's new risk assessment policy was all about.

Questions 1 through 5 provided us with general back-
ground information on each Section's activities.  In addi-
tion, the questions were designed to gradually narrow the
scope of the interview to discussion of those activities
which might involve use of risk assessment techniques.  In
order to do this we again employed an exclusion criterion:
programs which were perceived by staff as having no direct
impact on human health were eliminated from further discus-
sion.

-------
                             -67-
Questions 6 through 9 were designed to obtain more detailed
information on the incidence and frequency of risk assessment
activities, and define why and how risk assessments were
done. We constructed a summary definition of risk assessment
based on the NAS and EPA reports (U.S. EPA, 1984b; NAS, 1983)
(See Table 2).  Participants were then asked one step at a
time which parts of this definition matched what their
Section did.  Again, an exclusion criterion was used to
obtain detailed information on only those parts of the risk
assessment process a section was responsible for.

Questions 10 and 11 were designed to capture the Regional
perspective on how useful the risk assessment process
was to decision-makers, and what could be done to improve
risk assessment in the context of Regional operations.  We
felt the insights of those staff who perform risk assessment
could provide valuable feedback to those developing risk
assessment policy.  These questions were framed to encourage
both positive and negative feedback from staff, and to
avoid an appearance of interviewer bias.

Finally, Question 12 was asked to check the reliability of
our earlier exclusion criterion of no direct health impacts.
The typical answer to this question was "no", qualified
with a statement such as "our activities haven't warranted
use of risk assessment so we're unfamiliar with its appli-
cation."

Data Collection

Data was obtained primarily from semi-structured interviews
with EPA supervisory and project staff.  Responses were
coded immediately where possible, and the same structure
for data collection was employed across all programs to
facilitate comparison, contrast, and analysis.  The inter-
views were supplemented by document collection as secondary
sources of information.  Documents included key project
reports, risk assessments, and written procedures used by
Program offices to assess human health risks.

Several steps were taken in constructing the protocol
to assure that measurements were obtainable across all
programs.  We were aware that individual attitudes and
institutional forces could either facilitate or inhibit the
flow of information from participants to the interviewer.
As cooperation and consent of the Program offices was not

-------
mandatory, it was necessary to demonstrate that the survey
was intended to serve a beneficial purpose to Regional
administration and staff.  This purpose of the study was
explained to participants as: to learn how risk assessment
concepts fit Regional operations, and to provide staff with
information on how risk assessments can be done in site-
specific situations.

First, we obtained permission from all supervisors in the
chain of command to have staff participate in the survey.
This involved a briefing to Division Directors to propose
the idea and obtain any suggestions or comments on its
appropriateness.  Second, we met with Branch Chiefs to
discuss the survey, the risk assessment process, obtain
permission to interview staff, and to have them suggest
appropriate participants.  Invariably Branch Chiefs recom-
mended Section Chiefs; some also named other key personnel
who had more involvement with assessing human health impacts
from pollutants.  Third, Section Chiefs were contacted over
the phone to set up appointments for the interviews at a
time that was convenient to them.  They were told the
interviews were approximately one hour in length, and they
were encouraged to invite any of their staff to participate.
In some cases, after the initial interview with Section
Chiefs, they suggested we talk to other staff involved with
risk assessment activities.  As a result, where levels of
activity were greater, the amount of information obtained
and the number of staff interviewed usually increased.  This
was desirable, given our previously stated goal of getting as
much detail as possible when risk assessment activity was
relevant to Program operations.

-------
                             -69-

                RISK ASSESSMENT SURVEY QUESTIONS*


1.  What are the major programs of your Section?  (LIST ALL IMMEDIATELY)

2.  Which of these programs do you percieve as directly
    affecting human health?

•*«  (IF XJBS TO 2) How many of these program actions do you make
    a year?     (LIST NUMBERS ON ALL MAJOR PROGRAMS IMMEDIATELY)

    NOW, ONE PROGRAM AT A TIME, ASK:

4.  What are your major activities (decisions/review points)
    for each of these programs?   (ALSO CODE ASSESSOR/MANAGER/REVIEWER)

5.  Which of these activities do you percieve as directly affecting
    human health?

6.  (FOR THOSE ACTIVITIES THAT RECEIVE A YES ON 5, REFER TO
    HANDOUT) Do you perform or review the performance of these
    steps in any given capacity?

7.  (IF YES TO 6) How are health risks typically characterized
    in these assessments? (CODE QUAL. VS QUANT.? UNITS OF MEASUREMENT!)

8.  (IF YES TO 6) Approximately what percent of your program
    time and resources are spent assessing (or reviewing assess-
    ments of) health risks?

9.  (IF YES TO 6, REFER AGAIN TO HANDOUT STEPS)
    A) Who has responsibility for each of these steps?
       (CODE HQ,REGION,STATE,OTHER FED AGCY,OTHER)

   IF RESPONSE IS REGION, ASK B, C, and D by step:

    B) What is the basis for your responsibility?
       (CODE LAW/REG/POLICY/GUIDELINE/OTHER/NONE)
    C) Do you have a standardized approach for these steps?
       (CODE Y/N)                 (IF YES GET COPIES OF DOCUMENTS)
    D) What is your key source of information for these steps?
       (CODE ESD/CONTRACTOR/LITERATURE/OTHER and FIELD STUDIES/
        MODELING/OTHER)

10. (IF YES TO f6) I'd like to get an idea of the appropriateness
    of risk assessment to regional operations.
    A) Can you cite any examples where risk assessment
       was useful to your program decisions?
    B) In these examples, were health risks significantly
       reduced?   (Code Y/N & record explanation)    cost-effective?
    C) Can you cite any examples where risk assessment was
       not useful to your program decisions?
    D) Why was risk assessment not useful?
  MALL CAPS ARE PROMPTS FOR INTERVIEWER)

-------
                            -TO-
IL (IF YES TO 6)  Can you think of any improvements to the
    risk assessment process that would make it more useful
    in your program activities?

12. (IF NO TO 6,  GO OVER HANDOUT) I'd like you to brainstorm a
    minute - can  you imagine any uses for risk assessment
    in your program activities?

-------
                        Appendix A-3
This set of forms was designed to be used for recording
data at the interviews in a systematic fashion.  The forms
were adjustable in length to the amount of information
collected, and included a matrix table for detailed infor-
mation on each risk assessment step.  Coding schemes were
developed for each variable in the matrix to codify the
data as a preliminary step to analysis.

-------
Sect ion:
                                             APPENDIX A-3

                                         RISK ASSBSSNBNT SURVEY FORM 1
Program
 2
Y/N
 3
l/yr
                                              4A
                                          Activities
  B     5    6    7A     B
A/M/R  Y/N  Y/N  QT/QL  UNIT

-------
                                           RISK ASSESSMENT SURVEY FORM 2
Sect ion %
   8
 Effort

 % TINE AND RESOURCES
        EPA
       Contractor
 Features
Hazard
 ID
Dose/
Response
                                                                      Exposure
    Risk
Characterisation
 i
--j
uo
A) Responsibility
   (HO/Region/State)
B) Basis
   (Law/Reg/policy/
    guideline/
    other/none)
C) Approach
   (Established
    Procedure Y/N &
    Nane of DocuMt)
D) Source
   {BSD/Contractor/
    Other t Field/Model)

-------
10. Examples





A)	
B)outcome
C)
11. Recommendations
12. Activity

-------
                        APPENDIX A-4
Analytic Strategy

After initial results were produced, the factual portions
of the study were photocopied and reviewed by the key parti-
cipants.  This review served as a simple procedure for
validating the data collection process.  The participants
were asked to review and critique the presentation of facts
and interpretations for balance.  This effort also helped
minimize any bias due to the preparer's assumptions or
perspectives.  The data were subjected to both qualitative
and quantitative analysis.  Statistical techniques, compli-
mented with narrative, were employed to analyze and reduce
the survey data.  A hand calculator was used to aggregate
responses on programs to the section level.

Each section where risk assessment entered into program
activities was treated as a unit of analysis.  The focus
within a unit of analysis was on those activities that
participants consider to be hazard identification, dose
response evaluation, exposure evaluation, and risk
characterization.

The Cyber 170 Model 765 mainframe computer at Georgia Tech
processed the data on each section using the statistical
Package for the Social Sciences version 8.0.  A sample
program is provided in Table A-4a.

The analysis was completed in three steps.  First, the
background characteristics and risk assessment activities
were aggregated across programs to summarize the incidence
and frequency of these variables at the Regional office.
Next the relationships between background variables and risk
assessment activities were examined to determine factors
related to use in each program.  Finally, this information
was used as the conceptual basis to develop a process model
of risk assessment in Region IV operations. The data were
converged around alternative models, and the model most
congruent with the actual data was selected to explain the
relationships between key actors in the risk assessment
process.

-------
                               -76-


                              Table A-4a

                      SAMPLE SPSS COMPUTER PROGRAM
RUN NAME
FILE NAME
VARIABLE LIST
N OF CASES
INPUT FORMAT

INPUT MEDIUM
MISSING VALUES
VAR LABELS
VALUE LABELS

PEARSON CORR
OPTIONS
STATISTICS
FACTOR
STATISTICS
FINISH
SCAN   26 EOR
O ? EXIT
LINES   26
TED,FACTOR2.
/BYE
 FACTOR ANALYSIS
 FACTOR
 SECTION,RESP,BASIS,APP,HIS, DRS,
 ASSESS,MANAGE, OVERVW,RISKT,QLQT,REGION,
 REC,STEPS,REGS, USE
 UNKNOWN
 FIXED (F2.0,3F4.0,2F6.0,4F3.0,2F1.0,F3.0,F2.1,
F1.0.F3.0)
 DISK
 RISKT (999)/
 OVERVW,OVERVIEW/RISKT,SECTION TIME FOR
 RISK ASSESSMENT/STEPS,NUMBER OF STEPS
 PERFORMED/
 REGS,REGULATORY PROGRAM/
 STEPS PERFORM NO STEPS (.5)PORTION OF
 ONE (1)1-1.5 STEPS (2)2-2.5 STEPS (3)3 STEPS
 STEPS,RISKT, USE
 1
 1
 VARIABLES«RISKT, STEPS, USE/
 TYPE-PA2/
 MINEIGEN-1.0/
 ROTATE-VARIMAX/

 FACSCORE/
 4,5,6,7,8

   26

-------
                             -77-
Variables were summarized with appropriate descriptive
statistics that measured central tendency (mean, mode),
dispersion (range, standard deviation), and frequency of
responses.  These measures were used to describe the over-
all characteristics and risk assessment activities of the
Regional Office.

To examine the relationship between background variables
and risk assessment, crosstabulations were performed on
nominal and correlation analysis on interval level data.
Factor analysis techniques were used to explore interdepen-
dencies among the variables and develop a composite score
by which sections were ranked according to the magnitude of
their risk assessment activity (Rummel/ 1968).  [Note:
Similar scales have been constructed to rank individuals
according to several characteristics indicating intelligence
- e.g. , IQ scores.]

-------
                                                   APPENDIX A-5


                                           MAJOR ACTIVITIES AT REGION IV
Regu-
latory
Prog ran


1. CAA
§309
NEPA


2. NEPA
Compli-
ance
for EPA
Actions

Organizational
Unit
OFFICE OF POLICY
& MANAGEMENT

Env. Review
Section


NEPA Compliance




Major Activities


Review substance of environmental
documents to determine which
portions to route to other
regulatory programs for more
specialized review (e.g. air,
water, noise impacts)
Evaluate quality of documents
according to OFA Policy and
Procedures, 1-2-3.
Evaluate impact of selected
alternative according to
Agency-wide rating system,
LO-EC-EO-EU
Prepares EISs



Environmental assessments and
other NEPA related studies
Perception of
Public Health
Impact


None.
Indirect. Most pro-
minent in our review
of nuclear power
facilities.
Direct.
Direct.



Direct.
Risk Assessment
Steps Performed, Mana-
ged or Overseen*


None. If a risk assmt
present, send it to
appropriate program
for more specialized
review.
HI, DR, EX
None.
HI, EX and RC



HI, EX and RC
Role **


O
0
O
M



M
                                                                                                                     CO
                                                                                                                     I
 * HI=Hazard ID; DR=Dose Response; EX-Exposure;  RCH=Risk Characterization

** A=Assessor; M=Manag«nent; Q=Oversee;  B=Both(Assessor and Management)

-------
                                                        APPENDIX A-5
                                             MAJOR ACTIVITIES AT REGION IV
     Regu-
     latory
     Prog ran
Organizat ional
Unit
 Major
 Activities
Perception of
Public Health
Impact	
Risk Assessment
Steps Performed, Mana-
ged or Overseen*	
                                                                                                                  Role
                                                                                                                       **
3. Building
      Mgmt.
Mgmt. Services
Safety  (10% of Management
Services program)
                              Facilities*
                              Government
                              Property and Equipment
                              Security
                              Public Notices/Hearings

                              Ccranunications
                              Credentials
                              AV/Printing Duplication
                              Mail Room
                              Identification
Direct.  Esp« for
EPA personnel in
categories 1, 2, 3

Direct.
HI, EX
                                                                            HI, EX
M
                                                                                    M
                                                      None.
                                                           None.
                                               M
                                                                                                                          vo
                                                                                                                          I
             Water Division
4. NPDES
   Permits
   Issuance
Permits Section
 Decide what conditions  (i.e.
 limitations) should be  placed
 in permits to meet technology
 or water quality requirements.
 This may involve setting numerical
 limits or bicmonitoring when an
 unusual  substance is found.

 At minimum, determine what's
 expected of permittee in line
 with the effluent guidelines.
 Effluent guidelines are technology-
 based limits, which serve as a
 baseline of what's expected.

 If effluent discharges  to public
(water supply, assess quality of
(water with respect to human health
Direct.
HI and EX
M
                                                                   Direct.
                                                                            HI and EX
                                               M
                                                                  (Direct.
                                                                            (HI and EX
                                               M

-------
        APPENDIX A-5




MAJOR ACTIVITIES AT REGION IV
Regu-
latory
Program



5. CWA
NPDES
Permits
Compli-
ance





6. Construc-
tion
Grants
201









Organizational
Unit

Water Division
(cont'd)
Industrial
Operations
Section







Applied
Technology
Section










Major Decisions
Activities



Overview Inspections on treatment
facilities.



Overview discharge monitoring
reports (DMRs) done by permittees.
Direct implementation of CWA (ie.,
inspection and DMRs) in Florida
(only state not delegated).
Technical assistance to states
upon request.

(a) innovative alternative (IA)
technology program.
(b) Municipal sludge management.
(c) Municipal land treatment of
wastewater.
(d) Value engineering - maximizing
value for investment made in
treatment plant.
Policy development of Construction
| (Grants program at Regional level.

Perception of
Public Health
Impact


Direct. Only if
permit is in vio-
lation situation.


Direct.

Direct.





None.

Direct.
Direct.

None.


Indirect.


Risk Assessment
Steps Performed, Mana-
ged or Overseen*


DR, EX, RC. Have done
risk assessment in
serious violation
problems, (eg. Kentucky
raw waste problem)
DR, EX, RC.

DR, EX, RC.





None.

EX
EX

None.


None.




Role **


O




O
i
o
M Y





A

A
A

A


M


-------
       APPENDIX A-5




MAJOR ACTIVITIES AT REGION IV
Regu-
latory
Program



7. Con-
struction
Grants
201





8. Public
Water
System
(SDWA)





Private
Well
Contami-
nation
(SDWA)

Organizational
Unit

Water Division
(cont'd)
Grants
Management
(South Area)






Drinking Water
Section













Major Decisions
Activities



Approve facilities plan.




Award Construction Grants
Review and approve priority systems
and priority lists

Make sure that PWSs comply with
requirements of SDWA regarding
maximum permissable level
contaminants.
Provide technical assistance
to states upon request.
Overview State primacy
activities in regards to
SDWA.
Review data and judge significance
of health risks.

Notification of well owners.


Perception of
Public Health
Impact


None.




Direct.
None . Administ rat ive
precursors to grants

Direct.



Direct.

Indirect.


Direct.


Direct.


Risk Assessment
Steps Performed, Mana-
ged or Overseen*


None; States do this in
process of 305B report
and in developing Con-
struction Grants
program list.
None.
EX


HI, DR, EX, RC



HI, DR, EX, RC

None, but potential for
use.

HI, DR, EX, RC


HI, DR, EX, RC




Role **


M




M
0 i
o
t-
l
0



A

0


A


M


-------
                                                       APPENDIX A-5
                                             MAJOR DECISIONS/ACTIVITIES AT REGION  IV
     Regu-
     latory
     Program
Organizational
Unit
Major Decisions
Activities
Perception of
Public Health
Impact *	
Risk Assessment
Steps Performed
or Reviewed *
                                                                                                                 Role
9. Under-
   ground
   Injection
   Control
Groundwater
Section
Assist States in developing pri-
macy, ie. permitting programs for
underground injection control.
(During interim, Region inplements
permitting program.)
                              Oversee States permit  issuance
                              process once they've assumed
                              primacy.

                              Oversee States enforcement
                              program once they've assumed
                              primacy.
Direct.  They all
have potential to
directly affect hu-
man health.  However,
permit granted on
grounds that in-
jected material will
never migrate into
a drinking water
source.  No fluid
movement Ts allowed.

Direct.
                                                     Direct.
None, but potential for
use if leak did occur
or when Agency moves to
regulate Class 5 wells.
M
                                                                                                                          i
                                                                                                                          oo
                                                                                                                          ru
                                                                                                                          I
                                                                           None.
                                                          None, but potential  for
                                                          use if  leak did occur
                                                          or when agency moves to
                                                          regulate Class 5 wells.
                                               O

-------
                                                      APPENDIX A-5
                                          MAJOR DECISIONS/ACTIVITIES AT REGION IV
  Regu-
  latory
  Program
              Organizational
              Unit
              Major Decisions
              Activities
                                     Perception of
                                     Public Health
                                     Impact *
                      Risk Assessment
                      Steps Performed
                      or Reviewed *
                                                                                                              Role
                                                                                                                    **
          Water Division
            (cont'd)
Sole
Source
Aquifer
              Groundwater
              Section (cont'd)
              Designation of sole  source aquifers-
              1 to date in Region  (Biscayne)

              Review individual  projects to
              determine whether  a  significant
              impact on aquifer.
Direct.
                                                                Direct.
Prog ram
Support
                               Peer review of study plans for
                               hazardous waste investigations
                               and Remedial Action Master Plans
                               (RAMPs).
                                                   Direct,
                                                           HI.
                                                           HI.  Determine if
                                                           contaminants likely
                                                           to enter an aquifer.
                                                           If it is likely, they
                                                           are  considered hazards
                                                           and  under the SDWA are
                                                           not  eligible for
                                                           financial assistance.
                                                           Potential to perform
                                                           risk assessment on all
                                                           sole source aquifer
                                                           projects.

                                                           None.  BSD or
                                                           Superfund contractor
                                                           does assessment.
                                                                                                                       i
                                                                                                                      oo
                                                                                                                      U)
                                                                                                                       I
          Environmental
          Services Div.
10
Provide
all
chemistry
support
req. by
EPA prog.
Support
monit1 ing
efforts
in all
media.
Chemistry
 Section
Analyze samples collected in-house
by Hazardous Waste,  Air and Water
Monitoring, and Municipal and
Industrial Sections (majority under
Superfund program).
Direct.
EX.  As it relates to
samples collection and
analyses.  Reg ion
estimates quantity
of material that exists
in environment.

-------
                                                       APPENDIX A-5
                                           MAJOR DECISIONS/ACTIVITIES AT REGION IV
   Regu-
   latory
   Program
Organizational
Unit
Major Decisions
Activities
Perception of
Public Health
Impact *
Risk Assessment
Steps Performed
or Reviewed *
                                                                                                                Role
           Chemistry Sect.
             (cont'd)
                 Coordinate nationally contracted
                 labs' input collected under
                 Superfund.

                 Regional and national technical
                 assistance for analysis of env.
                 samples; quality control associated
                 with it.
                                     Direct.
                                     Direct.
                      None.
                                                                                       EX.   Region'estimates
                                                                                       quantity of material
                                                                                       that exists in env.
                         O
.  RCRA
Support
  Haz. Waste
   Section
 CERCLA
 and RCRA
 Support
Overview states compliance
activities.

Performs inspections for interim
status and permitted RCRA
facilities.

Enforcement - Sample waste from
facilities to determine whether
they should be in RCFA permitting
system.

Enforcement - Perform waste oil
investigations to detect sham
recycling operations (i.e., re-
cycling operations which don't
truly breakdown waste).  Facilities
exempt from RCRA if producing
legitimate recycled products.

Provide technical support for
criminal and civil cases (more
required for criminal investigat-
ions).
Direct.
                                                                 Direct.
                                                                 Direct,
                                                      Direct.
                                                     Direct.
HI, DR, EX

HI, DR, EX
O

A
                                                                                                               i
                                                                                                              CO
                                                                            HI,  EX,  limited DR
                                                                                      HI
                      HI, EX
                         A

-------
                           APPENDIX A-5
               MAJOR DECISIONS/ACTIVITIES AT REGION IV
Regu-
latory
Progran









CERCLA-
Superfund
Support















Organizational
Unit
Haz. Waste
(cont'd)

























Major Decisions
Activities
Enforcement - Perform closure/post-
closure studies. Do sampling, etc
to determine whether a facility was
closed up properly (similar to an
endangerment assessment) .
Provide training and technical
assistance to Regional program
offices, States for both RCRA
and CERCLA programs.
Do some of Superfund site investi-
gations (majority are done by
contractor) .
Provide field support (along with
contractor) for immediate
response efforts.
Overview and evaluate contractor's
field investigations on-site.
Coordinate field investigations -
provide logistical contract lab
support needed for sampling events
(Regional Sanpling Contractor Center
- RSCC).
Provides information needed to per-
form MITRE & endangerment assess-
ments (eg. info needed to determine
whether a site is a source of hazar-
dous; material; whether there's a
Perception of
Public Health
Impact *
Direct.




Direct.



Direct.


Direct.


Direct.

Direct.




Direct.

Direct.


Risk Assessment
Steps Performed
or Reviewed *
HI, EX, limited DR




HI, EX



HI, DR, EX


HI, EX


HI, EX

HI, EX overview




HI, EX (Note: NA to
RCRA as regulations
define what is and isn't
hazardous substantial) .

Role **
A




A



o
CO
VJ1
1
A


0

M




A




(vector for  it to spread).

-------
                                                         APPENDIX A-5




                                              MAJOR DECISIONS/ACTIVITIES AT REGION  IV
Regu-
latory
Program

TSCA &
FIFRA
Support
NPDES
Compli-
ance
Mon'ing
(Water
Div.
Enforce-
ment &
Compli-
ance
Support)
Water
Quality
Studies
(support
permits,
drinking
water, &
enforce-
ment
sections)
Ambient
Air Moni-
toring
(Support
Air
Manage-
ment
Branch)

Organizational
Unit

Haz. Waste
Section
(cont'd)
Municipal &
Industrial
Section


















Air & Water
Monitoring







Major Decisions
Activities

Officially provide support also for
TSCA and FIFRA; in practice this
hasn't occurred.
Do field sampling activities at
municipal and industrial facilities
(1 state).

Overview states activities in
seven delegated states (7 states) .

Technical assistance and training
to state and municipal personnel.


Waste assimilation (i.e., to
determine how much waste is a safe
level in water) and drinking
water quality studies.






Overview air monitoring programs
in support of Air Management Branch,
State and local agencies.

In case of toxics, perform
assessment for Air Management
Branch, State, and local agencies.


Perception of
Public Health
Impact *
	


Direct. Do sample
for toxic chemicals.


	







Direct.









Direct.



Direct.




Risk Assessment
Steps Performed
or Reviewed *
	


HI, EX in water quality
studies, RC in water
quality studies.

	







HI, EX, RC









HI



HI






Role **
	 ,


A



0


A
i
co
I
A









0



A



12.
13.

-------
                                                          APPENDIX A-5
                                              MAJOR DECISIONS/ACTIVITIES AT REGION IV
Regu-
latory
Program

RCRA &
CERCLA
Water
Moni-
toring
Support










Wetlands
Evalu-
ation ( in
support
of 404
wetlands
prugra i)





Organizational
Unit

Air & Water
Monitoring
(cont'd)

A



B



C



Environmental
Biology










Major Decisions
Activities

Provide support to Emergency
Response and Residuals Management
branches; State and local agencies.

Overview water monitoring programs
of states and local agencies in
support of Emergency Response,
Residuals Management Branch.
Assist when toxics problems in
water; (Hazardous Waste Section
performs assessments for water
toxic problems ) .
Operate STORET (Water) and SARODE
(Air) data bases; these are
national data bases - we provide
services for Regional system.
Identification of wetlands.

Field investigations for water
quality studies; evaluate biological
(e.g. indicator species) and
physical parameters.

Evaluate impacts of dredge
and fill activities
Collect data to support court
litigations and public hearings.

Perception of
Public Health
Impact *
Direct, and from
a multi-media stand-
point.

n



»



None.



None.

None.




Nona.

None.


Risk Assessment
Steps Performed
or Reviewed *
HI, EX



II II



II II



None.



None.

None.







Role **
None.



O



A ,
OO
-o
1

A



A

A




None.

None.


A

14.

-------
                                                      APPENDIX A-5

                                          MAJOR DECISIONS/ACTIVITIES AT REGION IV
  Regu-
  latory
  Program
Organizational
Unit
Major Decisions
Activities
Perception of
Public Health
Impact *
Risk Assessment
Steps Performed
or Reviewed
                      HI, DR, for aquatic
                      organisms only.
                                                                                                               Role
Water
Program
Support
(some
support
to other
programs)
 Envi ronmental
  Biology
  (cont'd)
Aquatic toxicity biononitoring; do
static toxicity testing for samples
collected by ESD's municipal and
industrial section.  This is
all in support of NPDES permit
program; States where delegated.

Establish whether an effluent is
toxic or not through acute toxicity
testing (e.g
                                                                Indirect.
                           Do water quality monitoring using
                           algal assay (use algae with quick
                           response to changes in nutrient
                           levels) .

                           Do acute and chronic toxicity
                           testing assessing impacts of test
                           ing for power plants.

                           Do toxic substances monitoring
                           in support of RCRA enforcement.
                           Using soil samples, perform
                           toxicity testing to assess
                           health and environmental effects.

                           Do sediment oxygen demand and nu-
                           trient exchange measurements for
                           Facilities Performance Branch of
                           Water (02 demand; this input to
                           to model to determine waste load
                           allocations) .
                                                      Indirect, due to
                                                      dilution to larger
                                                      water bodies.

                                                      Indirect.
                                                                                      HI, DR for an environ-
                                                                                      mental assessment.
                                                           HI, DR
                                                                                                              i
                                                                                                              co
                                                                                                              co
                                                                                                              I
                                     Indirect.
                                                      Direct.
                                                                            DR
                                                           HI, DR
                                                      Direct.  Related to
                                                      human health as
                                                      determine whether a
                                                      sanitation problem
                                                      exists.
                                                           None.

-------
             APPENDIX A-5
MAJOR DECISIONS/ACTIVITIES AT REGION IV
Regu-
latory
Program


15. CERCLA
Site
Screening






CERCLA
Compli-
ance















Organizational
Unit
Air & Waste
Management
Division
Site Screening
& Engineerirg
Section







Investigation &
Compliance
Unit















Major Decisions
Activities

Site identification.

Site investigation.

Site priori tizat ion, e.g., determine
which sites should be investigated
first, using MITRE model.

Put sites on NPL.

Determine responsible parties;
handle technical support contracts
for information gathering and
analysis needed to determine res-
ponsible parties (note: have access
to NEIC-Colorado and other govern-
ment enforcement agencies) .
Coordinate in-house between Remedial
Action, Emergency Response and
Office of the Regional Counsel on
CERCLA enforcement aspects. Issue
administrative orders, issue
notification letters negotiate
with responsible parties.
Provide peer review of technical
documents produced through course
of remedial action or emergency
response.
Perception of
Public Health
Impact *

Direct.

Direct.

Direct.



Direct.

None.






Direct.






Direct.



Risk Assessment
Steps Performed
or Reviewed *

HI, EX, RC

HI, EX, RC

HI, EX, R.



HI, EX, RC

None.






HI, EX, RC. Ideally
endangerment assessment
done before remedial
action or emergency
response undertaken.


None.



Role **

A

A

A


I
M °
I
M






M






O




-------
            APPENDIX A-5




MAJOR DECISIONS/ACTIVITIES AT REGION IV
Regu-
latory
Program

CERCLA
Compli-
ance
(cont'd)














16. CERCLA
Classic
Emergency
Response








Organizational
Unit

Site Screening
& Engineering
Section (cont'd)















Emergency
Response and
Control Section









Major Decisions
Activities

Coordinate efforts for cost recovery
with ORC; write technical portions
of cost recovery documents.

Research NPL sites; rely on infor-
mation from state, county, and
local sources to do this. (Note:
States can request that a site not
be evaluated for the NPL. Have had
this happen before.)
Superfund Comprehensive Accomplish-
ment Plan (SCAP) ensure enforcement
funds are used for what they're
intended for. Add and subtract
projects on SCAP to acccmodate
changing enforcement needs. Do
research for future SCAP candidate
tasks.
Notification of the National Res-
ponse Center (NRC) .

Evaluation by staff receiving call.
Containment and control by On-Scene
Coordinator (OSC). In Region IV
the person making the lead agency
decisions and determinations is the
OSC; we have 8.
Removal and disposal.
Documentation and cost recovery.

Perception of
Public Health
Impact *
None.



Direct.





None.







Direct.


Direct.
Direct.




Direct.
None.

Risk Assessment
Steps Performed
or Reviewed *
None.



HI support material.





None.







HI.


HI.
HI, EX




HI.
None.


Role **
A



A





M

!
V£>
O
1



M


M
A




M
M

-------
                                                          APPENDIX A-5
                                              MAJOR DECISIONS/ACTIVITIES AT REGION IV
Regu-
latory
Program

CERCLA
limed i ate
Removal
of Haz.
wastes
f ran
uncon-
trolled
waste
sites






CNV311
Spilx
Pre-
vention
Program

Su >er£ ind
Re led i d
Action
PK'jgran
(clean up
hssajpcv.ius
waste
sites on
NPL.)

Organizational
Unit

Emergency
Response and
Control Section
(cont'd)


















Remedial
Action Section








Major Decisions
Activities

Notification of the NRC.

Preliminary assessment which
involves identification and
evaluation of the hazard;
sampling (includes a health
assessment) .

Determine the removal action to
be taken based on results of
preliminary assessment.
Prepare action memo if iimnediate
removal requested.
Oversee removal action if done by
responsible party.
Cost recovery.
Engineering field inspection of
facilities that store petroleum
products both above and under-
ground.


Remedial investigation/feasibility
studies; i.e. studies to get more
information on NPL sites.



Write record of decision docuusents.


Perception of
Public Health
Impact *
None.

Direct.





Direct.


None.

Direct.

None.
Direct.





Direct.





Direct®

Construction of the remedy. j Direct,

Risk Assessment
Steps Performed
or Reviewed *
None.

HI.





HI.


None.

HI.

None.
HI after a spill occurs;
not during inspection.
However, risk assessment
could be done in the
future as a preventa-
tive measure.
HI, DR, EX, RC -
accumulate research
done elsewhere as part
of endangerment assess-
ment.

EX, RC




Role **
H

•A





M


n
i
-C
0

n
o





n





r

EX, RC JM
17.

-------
                                                   APPENDIX A-5

                                       MAJOR DECISIONS/ACTIVITIES AT REGION IV
Regu-
latory
Program
18. CDC In-
teragency
Agreement




19. RCRA
Waste
Planning



Organ i zat ional
Unit
Remedial Action
Section (cont'd)




Waste Planning
Section
9



Major Decisions
Activities
Scientific support consultation,
guidance manuals, training of HHS
personnel, review of environmental
test results for indication of
health hazard, listing of areas
closed or restricted because of
toxic substances contamination.
Emergency Response support sampling,
testing, health-related field
guidance and lab support.
Ongoing health studies and assess-
ments - provide for development of
standard guidelines and protocols
for lab and epidemological studies.
Make regulation interpretations.
Analyze state programs to determine
whether they meet Federal standards.
Administer RCRA grants.
Provide technical assistance
to states upon request.
Perception of
Public Health
Impact *
Direct.


Direct.
Direct.
Direct, but in a
very limited sense.
n n n
n n n
n n ii

Risk Assessment
Steps Performed
or Reviewed *
HI, DR, EX, RC


HI, DR, EX, RC
HI, DR, EX, RC
HI.
HI.
None.
HI, DR, EX

Role **
A


A
i
A ru
I
O
O
O
A

1  Waste Planning is an oversight function.  Authorized states were given quality assurance check, but we
   do not oversee how states assess health risks.   RCRA sets up Agency in informational role? what states
   do with information is up to them.

-------
                                                          APPENDIX A-5

                                              MAJOR DECISIONS/ACTIVITIES AT REGION IV
      Regu-
      latory
      Program
Organizational
Unit
Major Decisions
Activities
Perception of
Public Health
Impact *	
                      Risk Assessment
                      Steps Performed
                      or Reviewed *
                                                                                                                   Role
20. RCRA/Haz.
     Waste
    Permt'ing
Waste Engin'ing
   Section
Issuance of permits in states that
don't have authorization yet.  This
involves review of the Part B
application, the draft permit,
public notification, etc.  Two of
eight states are in this category.

Oversight of permitting process.
This involves reviewing draft
permits and applicatons for permits
to determine whether RCRA require-
ments are met.  6 of eight states
are in this category.	
Direct.
                                                                    Direct.
                      HI; potential to perform
                      entire assessment with
                      10/84 RCRA reauthori-
                      zation.
                                                                            HI.
                         M
                                                                                                                           LO
                                                                                                                           i
21. RCRA
    Compli-
    ance/
    Enforce-
    ment
    Waste
  Compliance
Oversee compliance/enforcement of
7 states authorized to implement
RCRA. (Directly manage Alabama's
RCRA c/e activities, as their
authorization was recently revoked.)

Perform hazardous waste site
inspections; review records kept
by states.

When violations discovered, issue
administrative orders under
Sections 3013, 3008 and 7003 of
RCRA.  This is a stepwise process,
involving a warning letter, next
a letter of violation, finally
an administrative order (AO).
Direct.
None.
                                               O
                                                                    Direct.
                                                                    Direct.
                                                                            None.
                                                                            HI, DR, EX, RC if 7003
                                                                            action necessary
                                                                            (endangerment assess-
                                                                            ment).
                                               M

-------
                                                          APPENDIX A-5
                                              MAJOR DECISIONS/ACTIVITIES AT REGION IV
      Regu-
      latory
      Program
Organizational
Unit
Major Decisions
Activities
Perception of
Public Health
Impact *	
                      Risk Assessment
                      Steps Performed
                      or Reviewed *
                                                                                                                   Role
                                                                                                                        **
22. FIFRA
 Pesticides
 Section
Enforcement of FIFRA regulations
as products move through ccranerce.
These efforts directed toward
those products, identified as the
" bad actors":  (a) identification
and (b) removal of products not
registered, suspended, or cancelled
from registration.  Consider types
of population, exposure potential,
etc.)
                               Provide technical assistance to
                               states; disseminate toxicity in-
                               formation (e.g. route of entry,
                               general health effects)  to states
                               and citizens upon request.

                               Form cooperative agreements with
                               and administer grants to states to
                               perform activities (e.g. stop
                               sales)  exactly as Region would in
                               two areas:  (a)  enforcement
                               activities, (b) certification and
                               training activities.
Direct.
                                                      Direct.
                                                      Direct,
                      HI, EX.  But not a       M
                      formal assessment.  Do
                      an informal, quick
                      assessment in order to
                      determine civil penalty
                      to violator.  Assess
                      factors that affect
                      both safety and
                      efficacy of pesticide.
                                                           HI, EX.  Occasionally
                                                           need to interpret risk
                                                           assessment documents,
                                                           but go outside section
                                                           on these occasions.

                                                           None.                    M
                                                                                                                           -Cr
                                                                                                                           I

-------
                                                           APPENDIX A-5
                                             MAJOR DECISIONS/ACTIVITIES AT REGION IV
      Regu-
      latory
      Program
Organizational
Unit
Major Decisions
Activities
Perception of
Public Health
Impact *	
Risk Assessment
Steps Performed
or Reviewed *
                                                                                                                   Role
                                                                                                                         **
23.  TSCA
Toxics Section
Development of toxic substance
regulations - develop information
on specific chemicals as they are
proposed for manufacture in our
Region.  (Not directly involved in
premanufacture notification
process.)

Enforcement of TSCA regulations on
specific chemicals (7), from manu-
facture through disposal, at any
point during this cycle.

A) Asbestos inspection program -
walk through to determine whether
                               asbestos present
                               condition.
                                  in "friable"
                               B) Inspection of manufacturing
                               facilities - walk through, take
                               photographs, examine records and
                               research information.
None.
                                                                    Direct.
                                     Direct.
None.
                                                                                                                            i
                                                                                                                           vo
                                                                                                                           Ul
                                                                                                                            I
                                                                            HI, and potential for
                                                                            EX, RC
                                                                            HI in all; EX in some
                                                                            cases.

-------
                                                      APPENDIX A-5
                                        MAJOR DECISIONS/ACTIVITIES AT REGION IV
 Regu-
 latory
 Program
Organizat ional
Unit
Major Decisions
Activities
Perception of
Public Health
Impact *
Risk Assessment
Steps Performed
or Reviewed *
                                                                                                              Role
                                                                                                                   **
TSCA
Toxics Section
   cont'd
C) Conduct audits of research
facilities in collaboration w/FDA,
NIOSH, CDC.  Research facilities
include both government and private
independent labs.  We examine re-
search data and the statistical
analysis of such data to determine
whether facility is following the
protocol that has been developed
internationally for onoogenicity
teratogenicity testing by the
Organization for Economic
Cooperation and Development
(OECD).

D) Provide technical assistance for
abatement to parties responsible
(done in cooperation w/GA Tech )
upon their request.
Direct.
HI, DR, EX, RC
M
                                                                                                                       en
                                                                                                                       I
                                                      Direct.
                                                                                     None, but have po-
                                                                                     tential to do risk
                                                                                     assessment here.

-------
            APPENDIX A-5


MAJOR DECISIONS/ACTIVITIES AT REGION IV
Regu-
latory
Program
24. CAA
Section
105
Grants



Process
SIP re-
visions
related
to air
planning






State
Toxic
Sub-
stances
Programs





Organizational
Unit
Air Planning
Section

























Major Decisions
Activities
Administer grants for air pollution
planning and control programs.

Overview States development of
such programs.
Audit States files in regards to
Section 105 grants.
Rate and approve SIP revisions ,
primarily to State regulations.
Review supporting information to
these regulations.

Handle all aspects of review pro-
cess (e.g. circulate for comments,
Federal Register publicatons),
technical and administrative.
Provide technical assistance to
States and Region IV programs upon
request.
Do air toxic substances dispersion
modeling for States; have capacity
to do multimedia modeling.

Develop compliance strategies for
particular sources.
Do risk assessment (e.g. provide
national experts for sane steps;
| perform exposure assessment) so

Perception of
Public Health
Impact *
None.


None.

None.

Direct.




None.



Direct.


Direct.



Direct.

Direct.


Risk Assessment
Steps Performed
or Reviewed *
None.


None.

None.

None.




None.



HI, EX, RC


HI, EX, RC



None.

HI, EX, RC


Role **
M


O

O

M




M



A


A



M

A


States maV dev*»lnr> Heir o/-var»a»"i/~>o 1
                                                                               I
                                                                              VD

-------
                                                         APPENDIX A-5
                                              MAJOR DECISIONS/ACTIVITIES AT REGION IV
Regu-
latory
Program

States
Toxics
Programs
(cont'd)
CAA
Preven-
tion of
Signi-
ficant
Deterio-
ration
(PSD)
Permits
and non-
attain-
ment
Sta-
tionary
Source
Control
Strategy
Develop-
ment and
Compli-
ance
support
for State
Implemen-
tation
Plans
(SIPS)

Organizational
Unit

Air Planning
Section (cont'd)



Air Engin'ing
Section

























Major Decisions
Activities

Provide technical assistance at
States request.



Oversee States' activities in
issuance of PSD and non-attainment
permits.








Oversee States' classification of
facilities according to title.

Review SIP sources for consistency
with CAA Section 110.

Provide technical assistance and
support to States (e.g. apply
models to determine whether states
achieving anbient pollutant cri-
teria) .





Perception of
Public Health
Impact *
Direct.




Direct.










Direct.


Direct.


Direct.









Risk Assessment
Steps Performed
or Reviewed *
HI t EX, RC.




HI, EX, RC










HI, EX, RC


HI, EX, RC


HI, EX, RC











Role **
A




O




I
^O
oo
1




O


0


A








25,

-------
                                                    APPENDIX A-5
                                       MAJOR DECISIONS/ACTIVITIES AT REGION IV
Regu-
latory
Program

New
Source
Perfor-
mance
Standards
(NSPS)
National
Emission
Standards
for Haz.
Air Pol-
lutants
(NESHAPS)
Acid Rain
Control
Program

26. CAA
Compli-
ance












Organizational
Unit

Air Engin'ing
Section (cont'd)















Air Compliance
Section













Major Decisions
Activities

Oversee States activities and SIPs
for consistency with Section 111
CAA; largely a delegated program.



Oversee States activities and SIPs
for consistency with Section 112
CAA; largely a delegated program.




Assisting in policy and program
development; oversee ambient
monitoring being performed by
states for EPA.
Oversee delegated State and local
compliance programs to ensure they
meet NAAQS.* States develop SIPs
to control major emission sources;
states develop emission limits to
meet NAAQSs. Once new regulations
are developed (e.g. NSPSs), they are
quickly delegated to states.
National Air Audit System (NAAS):
A) Do annual audits; send out
questionnaire.
B) Do file reviews; randomly select
files and evaluate their practices
against NAAS criteria to determine
1 whether source is in compliance.

Perception of
Public Health
Impact *
Direct.





Direct.






Direct.



Direct.







Direct
,,

n




Risk Assessment.
Steps Performed
or Reviewed *
HI, EX, RC





HI, EX, RC






HI, EX, RC



EX, RC







None.
..

••






Role **
O





O






O



O







0
n

n



                                                                                                                      '.Q
                                                                                                                       I
*NAAOS:  National Anbient Air Quality Standards

-------
                         Bibliography

Anderson, Elizabeth L. and the Carcinogen Assessment Group,
  U.S. EPA, "Quantitative Approaches in Use to Assess
  Cancer Risk", Risk Analysis, Vol. 3, No. 4, 1983.

Chestnutt, Thomas W., "A Risk Assessment of The Earthquake
  Hazards in the Southeastern United States", Georgia
  Institute of Technology, Atlanta, Georgia, 1983.

Committee on Environmental Decision Making, National
  Research Council, Decision Making in the Environmental
  Protection Agency, National Academy of Sciences,
  Washington, D.C., 1977.

Committee on the Institutional Means for Assessment of
  Risks to Public Health, National Research Council, Risk
  Assessment in the Federal Government;  Managing the
  Process, National Academy Press, Washington, D.C., 1983.

Committee on Risk and Decision Making, Risk and Decision
  Making; Perspectives and Research, National Research
  Council, National Academy of Sciences, Washington, D.C.,
  1982.

Conrad, J. "Society and Risk Assessment: An Attempt at
  Interpretation", Technology and Risk Assessment, London,
  Academic Press, 1980.

Crouch, Edmund and Wilson, Richard, "Estimates of Risk",
  Journal of Business Administration, Vancouver, Canada,
  1980.

Geertz, Clifford, The Interpretations of Cultures, Basic
  Books, New York, 1973.

"How to Assess Cancer Risks", Science, Vol. 204, May 25,
  1979.

Kates, R.W., "Managing Technological Hazard;  Research
  Needs and Opportunities", University of Colorado, Boulder,
  1977.

-------
                            -101-
McClintock, Charles C., Brannon,  Diane and Maynard-Moody,
  Steven, "Applying the Logic of  Sample Surveys to
  Qualitative Case Studies:  The Case Cluster Method",
  Administrative Science Quarterly, Vol. 24, Cornell
  University, Ithaca, New York, 1979.

Mogee, Mary Ellen, "Risk Assessment in the Regulatory
  Process: Rule Making in the Environmental Protection
  Agency", The George Washington  University, Washington,
  D.C., 1982.

Porter, Alan L., Rossini, Frederick A., Carpenter, Stanley
  R. , A Guidebook for Technology  Assessment and Impact
  Analysis, Vol. 4, New York, 1980.
Risk Management Subcommittee, Toxics Integration Task
  Force, Final Report (Task A.2), U.S. Environmental
  Protection Agency, Washington, D.C., 1983.

Ruckelshaus, William D., "Risk in a Free Society", EPA
  Journal, April, 1984.

Ruckelshaus, William D., "Science, Risk, and Public Policy",
  Vital Speeches of the Day, August 1, 1983.

Sieber, Sam, "The Integration of Field Work and Survey"
  Methods, American Journal of Sociology, 1978.

U.S. Environmental Protection Agency. (1984a) Your Guide
to the United States Environmental Protection Agency.

U.S. Environmental Protection Agency.  (1984b, Dec.)  Risk
  assessment and management:  framework for decision making.
  EPA-600/9-85-002.  Washington, D.C.  Available from Office
  of Policy, Planning, and Evaluation.

U.S. Environmental Protection Agency.  (1984c, Nov. 23)
  Proposed guidelines for carcinogen risk assessment; request
  for comments.  Federal Register 49:46294-46301.

U.S. Environmental Protection Agency.  (1984d, Nov. 23)
  Proposed guidelines for exposure assessment; request for
  comments.  Federal Register 49:46304-46312.

U.S. Environmental Protection Agency.  (1984ef Nov. 23)
  Proposed guidelines for mutagenicity risk assessment;
  request for comments.  Federal Register 49:46314-46321.

-------
                            -102-
U.S. Environmental Protection Agency.  (1984f, Nov. 23)
  Proposed guidelines for health assessment of suspect
  developmental toxicants and request for comments.  Federal
  Register 49:46234-46331.

U.S.  Environmental Protection Agency.  (1984g)  Draft Agency
  Report: Risk Assessment/Risk Management

U.S. Environmental Protection Agency.  (1985)   Proposed
  guidelines for the health risk assessment of chemical
  mixtures and request for comments; notice.  Federal
  Register 50:1170-1176.

Yin, Robert K.  "The Case Study as a Serious Research
  Strategy, " Knowledge;   Creation, Diffusion, Utilization,
  Vol. 3, No. 1, 1981.

Yin, Robert K. and Heald, Karen A., "Using the Case Survey
  Method to Analyze Policy Studies", Administrative Science
  Quarterly, Vol. 20, 1975.

-------