Appendices – Round Table Summaries
253
Hartnell – EU Perspective – Viral Risk
DePaola – What We Can Learn from the Vibrio Risk Assessment and Apply to
Viruses
Discussion during the roundtable questioned whether it was appropriate to
conduct a risk assessment before methodologies for virus detection were validated
and standards were put in place. Although the EU and US are getting closer to that
point, one panelist felt that we could not wait another 3–5 years to perform a risk
assessment on norovirus and that it was critical to know what levels of norovirus are
going into the water. The US and Canada have a timeline for the risk assessment to
be completed in 1 year. In the EU, legislation based on virus criteria is expected to
be put in place shortly.
When asked how data generated by RT-PCR can be used in performing a risk
assessment when RT-PCR is known to detect both infectious and non-infectious
virus particles, it was pointed out that there was no other method available and
that any level of virus detection, whether infectious virus or not, indicates that the
shellfish were exposed to norovirus. It was generally recognized that inadequacies
of RT-PCR are the greatest contributors to uncertainty in a risk assessment. Systemic
problems can occur in risk models when inactive viruses are counted and can
throw the entire risk assessment into an unmanageable state. It is important to have
information on the impact of active versus non-infectious viruses as part of the risk
assessment. It was recognized that detection of non-infectious viruses could also
lead to over regulation of the industry.
Another question was on how the illegal practice of dumping boat wastes or
vomiting overboard in shellfish harvesting areas can be taken into account when
performing a risk assessment. It was acknowledged that dumping of boat wastes
is illegal, but that it does occur. There is a need to separate contamination by boat
waste (misbehavior) from contamination by other means (e.g., storm water runoff,
failure of sewage treatment plants and septic tanks, etc.). Different risk management
approaches are needed to handle illegal dumping, which is a random act, versus
contamination by other means (which are often more predictable). Separate analyses
need to be performed and some information on the incidence of illegal dumping
would be needed. One needs to know the magnitude of the problem as it can have a
high impact. These occasional misbehaviors are consequential and can have a high
impact compared to the failure of a waste water treatment plant, which has controls
in place for such occasions. It doesn’t matter how good waste water treatment is or
what the dilution is if someone is vomiting in that harvest area.
Factors required for a risk assessment include knowing the attack rate and the
number of infectious particles in the leftover food, so that one can determine a
distribution as part of a Monte Carlo simulation. In such a case, one would know
the number of exposed individuals (based on illnesses) and the number of viruses
consumed to provide some sense of the level of risk. A third component which has
been used in other contexts, but might be useful in a norovirus risk assessment deals
with sporadic illnesses not typically measured [or recorded]. These three pieces
of information need to be looked at together. This information must be combined
253
Hartnell – EU Perspective – Viral Risk
DePaola – What We Can Learn from the Vibrio Risk Assessment and Apply to
Viruses
Discussion during the roundtable questioned whether it was appropriate to
conduct a risk assessment before methodologies for virus detection were validated
and standards were put in place. Although the EU and US are getting closer to that
point, one panelist felt that we could not wait another 3–5 years to perform a risk
assessment on norovirus and that it was critical to know what levels of norovirus are
going into the water. The US and Canada have a timeline for the risk assessment to
be completed in 1 year. In the EU, legislation based on virus criteria is expected to
be put in place shortly.
When asked how data generated by RT-PCR can be used in performing a risk
assessment when RT-PCR is known to detect both infectious and non-infectious
virus particles, it was pointed out that there was no other method available and
that any level of virus detection, whether infectious virus or not, indicates that the
shellfish were exposed to norovirus. It was generally recognized that inadequacies
of RT-PCR are the greatest contributors to uncertainty in a risk assessment. Systemic
problems can occur in risk models when inactive viruses are counted and can
throw the entire risk assessment into an unmanageable state. It is important to have
information on the impact of active versus non-infectious viruses as part of the risk
assessment. It was recognized that detection of non-infectious viruses could also
lead to over regulation of the industry.
Another question was on how the illegal practice of dumping boat wastes or
vomiting overboard in shellfish harvesting areas can be taken into account when
performing a risk assessment. It was acknowledged that dumping of boat wastes
is illegal, but that it does occur. There is a need to separate contamination by boat
waste (misbehavior) from contamination by other means (e.g., storm water runoff,
failure of sewage treatment plants and septic tanks, etc.). Different risk management
approaches are needed to handle illegal dumping, which is a random act, versus
contamination by other means (which are often more predictable). Separate analyses
need to be performed and some information on the incidence of illegal dumping
would be needed. One needs to know the magnitude of the problem as it can have a
high impact. These occasional misbehaviors are consequential and can have a high
impact compared to the failure of a waste water treatment plant, which has controls
in place for such occasions. It doesn’t matter how good waste water treatment is or
what the dilution is if someone is vomiting in that harvest area.
Factors required for a risk assessment include knowing the attack rate and the
number of infectious particles in the leftover food, so that one can determine a
distribution as part of a Monte Carlo simulation. In such a case, one would know
the number of exposed individuals (based on illnesses) and the number of viruses
consumed to provide some sense of the level of risk. A third component which has
been used in other contexts, but might be useful in a norovirus risk assessment deals
with sporadic illnesses not typically measured [or recorded]. These three pieces
of information need to be looked at together. This information must be combined