close

Вход

Log in using OpenID

Comments on the Meaning of “…as appropriate…”

embedDownload
Comments on the Meaning of
“…as appropriate…”
Bernard D. Goldstein, MD
University of Pittsburgh
Graduate School of Public Health
bdgold@pitt.edu
“EPA shall incorporate, as appropriate, based on
chemical- specific datasets and biological
effects, the recommendations…of the National
Research Council’s Review of the Environmental
Protection Agency’s Draft IRIS Assessment of
Formaldehyde into the IRIS process”
The House Report (112-151) accompanying the
Consolidated Appropriations Act of 2012 (Public Law
112-74)
NRC, Review of EPA’s Integrated Risk Information System (IRIS)
Process, Page 3: Emphasis added
The burden of proof is on the advocates of
these new approaches, even though they
are based on exciting and intellectually
stimulating methodological advances, to
show they will be helpful in the
management of risk.
(Goldstein, BD. Risk assessment of environmental chemicals: If it
ain’t broke… Risk Analysis 31:1356-62, 2011)
Integrating evidence rationally requires an
implicit or explicit set of guidelines…Such
frameworks range from ones that involve a
rigid, algorithmic integration process to
ones that provide loose guidelines and
allow experts substantial freedom in
applying them.
NRC, Review of EPA’s Integrated Risk Information
System (IRIS) Process, Page 91
All steps of the IRIS process, especially the
evidence integration and conclusions
reached, are necessarily laden with human
judgment, as are most scientific endeavors.
NRC, Review of EPA’s Integrated Risk Information System (IRIS)
Process, Page 25 Emphasis added
Range of Expert Judgment
X XX XX X
Range of Expert Judgment
X
X X
X
X X
Communication
of Science
IARC Overall Evaluation of Carcinogenicity to
Humans
1
Carcinogenic to Humans
2A
Probably Carcinogenic
2B
Possibly Carcinogenic
3
Not Classifiable
4
Probably Not Carcinogenic
Weight of Evidence
Steps in Developing a Data Quality
Objective
1.
2.
3.
4.
5.
6.
7.
State the problem
Identify the decision
Identify inputs to the decision
Define the study boundaries
Develop a decision rule
Specify limits on decision errors
Optimize the design for obtaining data
(From EPA QA/G-4HW: EPA Hazardous Waste Office)
What are the Problems with IRIS?
(NRC Review of IRIS: Summary)
• Rigorous editing needed to reduce redundancies,
inconsistencies and text volume
• Assessment methods should be described more
fully
• QC processes should be enhanced
• Review and evaluation processes should be
standardized
• Appropriate expertise on evaluation teams
should be assured
What are the Problems with IRIS?
(NRC Review of IRIS: Chapter 8)
• Assessment methods need to be updated in a
continuing strategic fashion
• Inefficiencies in the IRIS program need to be
systematically identified and addressed
• Evolving competencies that reflect new
scientific directions are needed
What are the Problems with IRIS?
(NRC Review of IRIS: Repetitive throughout document)
The IRIS process takes too long!!!
Missing from the problem statement:
A proposal to change risk assessment
methodology should be accompanied by some
analysis of what adverse environmental or
human health impact risk managers could have
avoided if only the risk assessment had been
more accurate or appropriate.
(Goldstein, BD. Risk assessment of environmental
chemicals: If it ain’t broke… Risk Analysis 31:1356-62, 2011)
Problem Formulation
 The problem formulation step requires
stakeholder involvement, including elicitation
and careful attention to their perception of a
problem and their needs for a solution.
 For a decision tool, stakeholders include both
those affected by the decision tool and those
who use the tool in their decision process.
The NAS/NRC contribution to the problem
- We fail to fully recognize that EPA and state/local
agencies need to be responsive in a timely fashion to
decisions that are best made sooner rather than later
- We fail to fully recognize that EPA and state/local
agencies have now and for the foreseeable future
limitations in resources that force them to focus on
highest risk issues
- We seem to believe that because something is
possible to accomplish scientifically, it is appropriate
to add unneeded decimal places to risk estimates.
The NRC Formaldehyde Committee did
not make Recommendations about the
IRIS Process
• Preface: “The committee closes with
recommendations for improving the IRIS
assessment of formaldehyde and provides some
general comments on the IRIS development
process”
• Chapter 7. “The committee offers some
suggestions for improvement in the IRIS
development process that might help the … EPA if
it decides to modify the process”
Review of the Environmental protection
Agency’s Draft IRIS Assessment of
Formaldehyde Committee
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Chair, University of Southern California, Los Angeles
Vice-Chair, University of North Carolina at Chapel Hill
Miami University, Oxford, OH
Massachusetts Department of Environmental Protection, Boston
University of Washington School of Public Health and Community Medicine,
Seattle
Pacific Northwest National Laboratory, Richland, WA
North Carolina State University, Raleigh
Lovelace Respiratory Research Institute, Albuquerque, NM
University of Wisconsin, Milwaukee
University of North Carolina at Chapel Hill
University of Georgia, Athens
University of Illinois, Chicago
National Opinion Research Center, University of Chicago, IL
University of South Florida, Tampa
The University of Texas M. D. Anderson Cancer Center, Houston
Review of EPA’s Integrated Risk Information
System (IRIS) Process Committee
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Chair, University of Southern California
University of California, Irvine
University of California San Francisco
University of Washington
Johns Hopkins School of Public Health, MD
North Carolina State University
University of Washington
University of Arizona
Harvard School of Public Health, MA
University of Michigan
Argonne National Laboratory, IL
Carnegie Mellon University, PA
Center for Public Environmental Oversight, CA
University of Iowa College of Public Health
University of South Florida
IARC Overall Evaluation of Carcinogenicity to
Humans
1
Carcinogenic to Humans
2A
Probably Carcinogenic
2B
Possibly Carcinogenic
3
Not Classifiable
4
Probably Not Carcinogenic
Weight of Evidence
KNOWN
PROBABLE
POSSIBLE
KNOWN
PROBABLE
POSSIBLE
What else should be done?
• What percentage of chemicals are “on the line”, or how
broad is the expert opinion on most agents; i.e., are
there many problem cases indicating that every
chemical should be extensively reviewed, or are there
so few that they could be singled out for more
extensive review? (Bad cases make bad process)
• What can be learned from review of those agents for
which “weight of evidence” classification by IRIS, EPA,
NTP, FDA, IARC, OSHA etc has been upgraded or
downgraded
• What can be learned from estimating the risk averted if
decisions had been made earlier?
Examples of Lack of Specificity. Are
these an opportunity for EPA?
 Risk-of-bias assessments on all studies used by
EPA as primary data sources: “Whatever
approach is adopted”
 Quantitative approaches to evidence integration
include meta-analysis; probabilistic bias analysis
and Bayesian analysis: “The committee is not
recommending a particular approach”
 Uncertainty analysis: multiple options throughout
“If it ain’t broke don’t fix it”
- Bert Lance
“If it ain't broke, don't fix it' is the slogan of the
complacent, the arrogant or the scared. It's an
excuse for inaction, a call to non-arms.”
-Colin Powell
"If it ain't broke... the government will fix it until
it is."
-Dell Hunt, 2001
1/--pages
Пожаловаться на содержимое документа