Risk Perceptions and Analyses
NIRT: Evaluating Oversight Models for Active
Nanostructures and Nanosystems:
Learning from Past Technologies in a Societal Context
SES-0608791
Susan M. Wolf, J.D. (PI)
Efrosini Kokkoli, Ph.D. (Co-PI)
Jennifer Kuzma, Ph.D. (Co-PI)
Jordan Paradise, J.D. (Co-PI)
Gurumurthy Ramachandran, Ph.D. (Co-PI)
Jee Ae Kim (Doctoral Student)
NIRT Research Goals & Methodology
1. Assessment of 6 oversight models utilizing criteria (literature
collection and analysis, expert elicitation, consensus):
–
–
–
–
–
–
drugs
medical devices
chemicals in the environment*
chemicals in the workplace*
gene transfer research (“gene therapy”)
genetically engineered organisms in the food supply
2. Application of oversight lessons to nano-bio-technology
(mapping, consensus)
3. Development of oversight models for nanobiotechnology
products and research (scenario analysis, consensus)
Risk Perceptions and Analyses-Outline
1.
Limitations of current risk assessment paradigm
2.
Lessons learnt from oversight of chemicals and chemical
RA
3.
Implications for nano-bio
4.
Risk perception criteria in evaluating oversight models
Data needed for classical RA
Production levels;
Emission factors
Environmental
media monitoring
Pollutant Emissions
Exposure
Quantification
Environmental
Concentrations
Environmental
Modeling
Human exposure
monitoring
Risk Assessment
Human Exposure
Exposure Modeling
Biologically relevant
dose
Pharmaco-kinetic
modeling
Effect Quantification
Interactions with
macromolecules
Adverse Health
Effect
Quantitative
Structure-Activity
Relationships
In vitro toxicity
studies
Animal toxicity
studies
Epidemiological
Studies
Limitations of current RA paradigm
zQuantitative risk assessment: basis of considerable amount
of policy making
z Supreme Court (1980) Benzene decision: OSHA must
find a significant risk to health (5th Circuit)
z EPA adopted risk assessment and risk management
as agency policy in 1984
z We have adequate information for only a handful of the
thousands of chemicals in use.
z Existing regulatory system is incapable of carrying out a
detailed RA for every chemical.
Pollutant Emissions
Particle Characteristics
Surface
characteristics
Particle shape
Particle
size
distribution
Agglomeration
Environmental
Concentrations
Uncertainties
about NP
Human Exposure
Concentration metric
Mass
Surface
area
Number
Biologically relevant
dose
Bulk and surface chemical
composition
Dermal
Ingestion
Ocular
Translocation
mechanisms
2. Measures of effect in
epidemiological studies
Adverse health
effects
Exposure routes
Inhalation
1. Mechanisms by which
cell signaling and gene
activation
lead
to
physiological effects
Mechanisms by which oxidative
stress leads to cell signaling and
gene activation
Dose metric
Mass
Surface area
Number
Mechanisms of oxidative stress
1. Surface area
2. Transition metals
3. Intracellular transport of
particles to mitochondria.
Consequences of Limitations
• Tremendous uncertainties precludes risk
assessments using conventional methods.
• Options to Conventional RA:
– Screening assays
– Performance-based standards (exposure limits) vs
process-based standards (UK)
– Tiered system of oversight (e.g., REACH in Europe)
– Expert judgment in RA
Expert Judgments and Risk
• Assessment of probabilities is necessarily
subjective (Herman Kahn, 1960)
• Reactor Safety Study, 1975
– Subjective probability assessments show broad
spread of values
– Experts are overconfident
– They tend to under-estimate risks
Risk Perception
• Risk = f (Probability, Consequences)
• Perceived Risk = f (Perceived probability,
Consequences, Outrage)
• Risk is an emotional concept
• Cannot be separated from psychology
Past Experience with Chemical Risk
Assessment
• When faced with great uncertainty in RA, both
experts (scientists, regulators, industry) and lay
people behave in “predictably irrational ways”.
(Kahneman and Tversky, 2000)
– Exaggeration of small/negligible risks
– Neglect of real risks
– Possibility of extreme, unanticipated events
1. How can scientists, regulators and industry
manage an unanticipated crisis,?
2. How does the oversight system engender public
confidence ?
Risk Perception Related Criteria for Evaluating
Oversight Models
A.
Development
i.
Impetus
ii.
Public Input
iii.
Empirical basis
B.
Attributes
i.
Treatment of uncertainty
ii.
Stakeholder inputs
iii.
Compliance & enforcement
iv.
Considerations of fairness
C.
Evolution
i.
Extent of change in Attributes
D.
Outcomes
i.
Public Confidence