US: Padieu case gets the 20/20 treatment; phylogenetic analysis totally misrepresented

The case of Philippe Padieu, the French-born Texan found guilty in May 2009 on six counts of aggravated assault with a deadly weapon and sentenced to 45 years for five counts and 25 years for the remaining count – all of which run concurrently - was featured last Friday night on US ABC TV's tabloid-style news magazine, 20/20.

Five and half million viewers watched as Mr Padieu faced trial by media yet again. All six parts of the one hour show (actually 39 minutes minus commercials) are available to watch online.

Part 1: Women recall HIV criminal's allure
Part 2: HIV diagnosis rocks women's lives
Part 3: Women take matters into own hands
Part 4: HIV serial dater faces victims in court
Part 5: Man convicted of HIV crime speaks
Part 6: Women want case known to protect others

It's basically sold as the story of a group of scorned women uniting to put Mr Padieu behind bars, summarised beautifully by the accompanying story on the ABC news website headlined, 'How Women United to Stop HIV-Positive Man, Women's Horror at Diagnosis Replaced With Mission: Stop Man From Infecting Others.'

There's so much I could say about the show, which is something of a milestone in criminal HIV transmission reporting in the mainstream media, but I'm going to limit my comments about the very worrying misrepresentation of phylogenetic analysis as 'proof' that Mr Padieu was the source of all the women's HIV infection. Perhaps blog readers could fill in the comments sections with insights and criticisms of their own about this programme.

[Update: Catherine Hanssens of The Center for HIV Law and Policy has some terrific comments and insights in her Sept 29th blog post.]

In Part 4 of the show, presenter/journalist Elizbeth Vargas says that it was Mr Padieu's "own DNA" that proved he was guilty. But phylogenetic analysis is all about testing the genetics of HIV, not the individual. They then showed one of the US's foremost experts in HIV forensics, Dr Michael L Metzker, of Baylor College of Medicine, Houston, Texas, who testified for the prosecution that Mr Padieu's virus was extremely similar to that of the six women. Except here he says definitively that Mr Padieu was "the source" of the six women's HIV. I've written about the limitations of phylogenetic analysis many times: the issues are summarised here.



In the final part of the show, we are introduced to 'Lisa' who dated Mr Padieu in 1997, and was diagnosed HIV-positive that same year. The show gives Dr Metzker a sample of Lisa's blood and he says that "preliminary analysis" suggests that Mr Padieu was the source of all seven women's HIV infection. The show concludes that Mr Padieu "gave Lisa HIV in 1997" and goes on to suggest, without a shred of evidence, that he had been diagnosed earlier than 2005 and knowingly infected Lisa and possibly hundreds of other women.



I'm extremely disappointed in Dr Metzker for totally misrepresenting what phlyogenetic analysis can prove. It is impossible to conclude, given the many limitations of phylogenetic analysis, that Mr Padieu infected Lisa in 1997. It is, in fact, just as possible that Lisa infected Mr Padieu.

I don't expect 20/20 to explain the science (in fact, I expect them to get it wrong), but I do expect Dr Metzker, who is (was?) considered to be a respected scientist, to be less definitive about his conclusions. Maybe Dr Metzker would like to explain how he could be so sure - it would be very helpful to know if he has developed new, as yet unknown, techniques in phylogenetic analysis that can definitively pinpoint timing and direction of transmission.