Other online diaries:
Fri, 01 Oct 2004
Drug testing not open enough? - 18:40
When the news first hit Cyclingnews carried some of the details, in particular this paragraph.
Tyler Hamilton's case is the first ever positive for a blood transfusion, as up until very recently, doping via this method has been undetectable. A powerful blood test developed by Australian researchers was implemented at this year's Tour de France. The test didn't look for a particular banned substance, but instead examined whether there were any abnormalities in a person's blood as a result of artificial manipulation. At the Tour, it was announced that homologous blood transfusions could be detected, but autologous transfusions could not.
Of note is that it is a recently developed test and this is the first time anyone has ever been caught with this test. That would not be so bad, however it appears there may indeed be serious problems with this test. Cyclingnews spoke with Dr Michael Ashenden (this interview also contains a good glossary of terms on the issue). In this interview the Dr was very closed about some details of the test. Today some researchers and specialists in the field commented (search for the letters "The new blood test #1", "The new blood test #2", "The new blood test #4" and The new blood test #5) in the Cyclingnews letters columns.
First it is interesting to note "Clinical trials for a diagnostic product used in a hospital require 10-20,000 repetitions for approval", however if the papers on this method are to be believed this test has only been performed on a sample of 45 people known to have had transfusions and it was not tested on people known not to have had transfusions. Thus there is no knowledge of the possible false positive rate. (the letters I reference above discuss this much better than I). A noteworthy quote from one of the letters.
...in cyclingnews.com attributed to Dr Ashenden of "Science and Industry Against Blood Doping".
I comment on all of this to a large extent as it appears the information about the test is being kept secret and that is in itself damaging the entire process. This whole episode reeks of what computer people call Security by Obscurity, and as we all know that really never works in the long run.