ProMED posted a summary of the Inquiry from the Times and Martin Hugh-Jones (a moderator) followed with these thoughts.


ProMED-mail
<promed@promedmail.org>

[The summary of the report can be seen at
http://www.royalsoc.ac.uk/inquiry/idl_sum.pdf

The Inquiry's full report can be seen at
<http://reports.royalsoc.ac.uk/intro.htmweb>
and includes also the oral and written evidence. -Mod.AS]

[Reading the Summary & Main Recommendations there is very little one can
disagree with, and much to agree with. It is a sound document. But there
are two points that have been ignored and they are important:

[1] Every model must undergo validation and verification. While checking
the arguments and scales used is obvious, it is not always done. But I have not
seen one model tested retrospectively and critically against the reality in
the field. Predictions are not the same as reality. All that was done was to see if
the curves matched generally. This testing should have been done last year
during the epidemic and certainly against the modelers' demand for 3kms
culling. There was sufficient field experience and GIS topographic and detailed
agricultural data to put it in doubt and certainly to hard test. I have a phrase I use
on my students and those over-enamored of their computers and models, "Why
should I believe you when you have a computer pallor and no mud on your shoes?" The
truth is in the field, not in the computer. When models are checked and rechecked
against reality they can be fine-tuned and may eventually become useful.

Remember that they all start as merely the mathematical expression of the
model builder's presumptions and assumptions.

[2] During any emergency, big or small, appropriate data must be collected
and kept securely to allow a critical retrospective post-mortem. This may need
a special forensic team with funding to independently assess what had
happened and why, using ministry samples and data collected during the
epidemic and even additional data collected subsequently. When all the dust has settled and
honors awarded, one must be able to sit down and work out, without blame,
what went right, what could have been done better, and play "what-if" if one is to
ever get it right. Plus the next time will be different. If there are good
databases, they can be used to model the future and ask new questions. Also
these data should be readily shared.

Based on my experience, I fear both are hobby horses; I have built my share
of mathematical models. Some fall in love with their models but you should
never marry them. And the truth is in the field. - Mod.MHJ]

[see also:
Foot & mouth disease - UK: formal inquiries 20020713.4743
Foot & mouth disease - UK: vaccination 20020630.4642]...............................as/mpp/pg/dk