On Tuesday the jury in the Jim Leyritz DUI manslaughter trial sat through hours of junk-science testimony from the head toxicologist of Broward County, Florida. The tale of how his testimony was admitted into evidence is as bizarre as what he told the jury.

Blood Alcohol Level Calculation

If anything should be well-established forensic science by now, a juror would expect blood-alcohol analysis to be it. So jurors in the Leyritz trial are likely to believe what the toxicologist told them yesterday.

Once again, we should all be thankful we aren’t on this jury. Because I’m not a juror in the Leyritz trial I heard the legal arguments about the evidence. I know the toxicologist’s testimony is evidence only of the shoddy nature of forensic toxicology.

It seems the cops drew two blood samples from the defendant, an hour apart (3 and 4 hours after the accident). Based on the differences between the two samples, the toxicologist determined the defendant’s metabolism rate and then extrapolated what his blood-alcohol level must have been at the time of the accident: about 0.182.

However, Judge Marc Gold ruled that the second blood draw was illegal; as a result the prosecution could not ask the toxicologist about the extrapolation he made based on the second sample.

Specifically, Judge Gold asked the prosecutor whether the toxicologist had “relied exclusively on” the second sample for his opinion or whether he could render an opinion without relying on it. The prosecutor (a young woman who frankly seems clueless) said he could; he had another “calculation” he could use.

Clearly the judge expected the toxicologist to testify that the blood-alcohol level was somewhere around 0.182 based on this new calculation; but, alas, he testified that the level was about 0.196, a difference of 0.014.

Judge Gold has previously ruled in this trial that plus or minus 25% is unacceptable in statistics, at least with a sample size of 3 blood-alcohol levels (0.04, 0.06, and 0.08) In this instance, he has only two numbers to evaluate (which makes it impossible to determine what the real margin of error is)—and I don’t mean 2 blood samples, I mean 2 blood-alcohol levels (0.182 and 0.196). There’s a 0.014 difference (14 points).

In other words, there’s no way of telling how accurate or inaccurate either of the numbers is.

Junk Science

What this proves is that either the method used by toxicologists to calculate the rate at which alcohol is metabolized based on two samples is wrong or the method used to extrapolate the level of blood alcohol at a given past time based on a sample is wrong.

If so, then there needs to be substantial reform in DUI forensic toxicology.

The Judge and the Jury

It will be interesting to see how Judge Gold handles this.

It seems to me (and I am not a lawyer) that he has several issues to resolve:

• The prosecutor did not understand what he meant by asking her if the toxicologist relied exclusively on the second sample for his opinion, and as a result she allowed the toxicologist to introduce into evidence an entirely new opinion, which the defense had never before heard.
• The defense did not know the calculation generally used to extrapolate a prior blood-alcohol level, or they would have anticipated that the toxicologist’s new estimate would be much higher than his first estimate.
• The toxicologist’s change in his testimony reveals the gross flaws in the whole “science” of DUI toxicology.
• The problem is that the defense can’t cross-examine this testimony and show the flaws, because the judge (at the defense’s request) has excluded the second blood sample.

Of course, it may simply be that in the ten minutes the toxicologist was given to calculate a new estimate, he made a math error.

Hoist on their own petards! All of them.