1 of 17

1
Assessing and Managing the Risks Associated With
Eating Seafood

• Don Schaffner, Ph.D.
• Professor and Extension Specialist
• Rutgers - The State University of NJ

2
Overview

• Understanding risk analysis, risk assessment and
  risk management
• The steps in risk assessment
• Special attention focusing on exposure and
  dose-response
• Using Methyl Mercury (MeHg) as an example
• Risk management
• Continue with MeHg example
• Show how FDA Foods Advisory Committee feedback
  was used to improve the risk message

3
Risk Analysis Components

• (Quantitative) Risk Assessment
• How big is the risk, what factors control the
  risk?
• Scientific process
• Risk Management
• What can we do about the risk?
• Political process
• Risk Communication
• How can we talk about the risk with affected
  individuals?
• Social and psychological process

4
Risk Assessment

• Hazard Identification
• What agents, food(s) and people are involved?
• Exposure Analysis
• What is the chance of exposure?
• What is the level of exposure
• Dose-Response Analysis
• What is the human health effect of the exposure?
• Risk Characterization
• Complete picture of the assessed risk

5
Hazard Identification

• Agent Developing neurological system is very
  sensitive to toxicity of methyl mercury
• Food Mercury is a naturally occurring and also
  released into the air through industrial
  pollution. Mercury can accumulate in streams and
  oceans. Bacteria transform mercury into methyl
  mercury. Fish absorb methyl mercury as they feed.
  Methyl mercury builds up more in some fish than
  others depending on what they eat.
• People Women (of child-bearing age, who are
  pregnant, who could become pregnant, who are
  nursing mothers) and young children

6
Dose-Response Analysis

• Reference dose (RfD) is an estimate (with
  uncertainty spanning perhaps an order of
  magnitude) of a daily oral exposure to the human
  population (including sensitive subgroups) that
  is likely to be without appreciable risk of
  deleterious effects during a lifetime.

7
Background on MeHg RfD

• EPA derived a MeHg RfD in 1995 based on a MeHg
  poisoning incident in Iraq.
• Congress mandated (1997) that EPA fund a National
  Research Council study to determine if the RfD
  was scientifically justifiable
• NRC panel found RfD to be scientifically
  justifiable (2000) but suggested EPA revisit the
  issue using new 3 new studies
• The three new studies are the Seychelles Islands
  study, Faroe Islands study and the New Zealand
  study.

8
Data used for RfD

• Seychelles Islands (-)
• 779 mother-infant pairs, infants followed from
  birth to 5.5 years, standardized
  neuropsychological endpoints, maternal hair
  mercury concentrations.
• Faroe Islands ()
• About 900 mother-infant pairs, Children tested w/
  variety of tasks at 7 years of age, cord blood
  mercury and maternal hair mercury measured.
• New Zealand ()
• 38 children of mothers with hair mercury levels
  during pregnancy greater than 6 ppm matched with
  children whose mothers had lower hair mercury
  concentrations, 237 children were assessed on a
  number of neuropsychological endpoints (similar
  to Seychelles study) at 6 years of age

9
RfD determination

• There are several ways to determine a RfD, and
  one is to used a Benchmark Dose (BMD)
• BMD calculations use a model to relate exposure
  to effect, but to use it we need to know
• how low is abnormal
• If exposed, chance of becoming abnormal
• Add uncertainty factor
• Intra-human variability
• RfD was determined to be 0.1 ug/kg/day, which
  corresponds to 5.8 mg MeHg/L blood

10
Exposure assessment

• What foods are responsible for the exposure?
• What is the chance of a particular individual
  being exposed?
• If exposed, what is the level of exposure?
• If enough data are available, a computer model
  can be used to predict exposure, and to
  investigate interventions.

11
Exposure assessment model
Seafood consumption
Diet-blood ratio
Blood-hair ratio
MeHg by Species
MeHg Exposure
MeHg Blood Levels
MeHg Hair Levels
Species market share
12
Exposure scenarios

• Divide fish into high, medium and low MeHg
  species
• Run different computer simulations including
• No dietary exclusions at all
• Consumption from the Medium and Low groups, but
  not High
• Consumption from the Medium and Low groups,
  limiting the amount from the Medium group
• Consumption Low group, no Medium or High

13
Exposure model summary

• Model appears to predict the National Health and
  Nutrition Examination Survey (NHANES) data from
  the JAMA article
• Model will be peer reviewed, and can be used to
  inform risk management decisions

14
Risk Management

• The results of the exposure and dose-response
  assessments have given us a picture of the
  situation
• Now we must weight competing issues
• Fish is
• a good source of protein, provides important
  nutrients and is generally affordable
• Fish may also contain substances that are harmful
  to health

15
Risk Management

• About 8 of at-risk population above the RfD
• Objective reduce this , and still keep fish in
  the diet

16
Improved risk message

• As a results of feedback FDA received from its
  Foods Advisory Committee (FAC)
• FDA and EPA should combine their two independent
  advisories
• Tell people what eat a variety of fish mean?
• People want to know about canned tuna so tell
  them!
• Unify commercial and recreational fish message
• Create specific advice for children

17
Summary

• The risk analysis process and its components
• The step in risk assessment, included examples
  from exposure assessment and dose-response for
  methyl mercury
• The nature of risk management and the approach
  used by FDA and EPA in creating their new (March
  2004) advisory for methyl mercury in fish for
  women and children