After you consume alcohol, your body begins the process of absorption and then elimination. The alcohol is absorbed by the blood through the intestinal tract and is eliminated through excretion (sweating, exhaling) and metabolism.
Breath test machines work on the principle that the amount of alcohol you excrete through exhalation can be used to estimate the amount of alcohol in your blood. It is an indirect method of determining whether someone is over the legal limit of .08 grams per 100 milliliters of blood.
The more direct method, of course, is blood testing. Even blood testing is not a direct method, however, because what's actually tested in the chromatograph is the headspace gas in the blood tube.
As blood is fragile, care must be taken to preserve samples. Anticoagulants are added to the tubes to prevent clotting and an antiseptic cleaner is used around the blood draw site to prevent bacteria from infecting the person or the blood sample. The blood must also be refrigerated to prevent the growth of little yeasts called candida albicans which can cause the blood sample to ferment in the tube.
What can't happen in that blood tube, though, is the elimination of alcohol for there is nothing to metabolize it. If the blood is collected and stored properly, the sample should yield approximately the same result any time you choose to test it. It is in a state of vacuum.
Unless, of course, you're the Texas Department of Public Safety and you have to explain why some blood samples that tested over the legal limit later yielded results (when tested by a private facility) that were below the legal limit.
Keith Gibson, the Quality Manager for the DPS lab on Jones Road in Houston had a problem. It seems that when a private lab retested a sample of blood from a DWI case, the result came in quite a bit below what the DPS test said. Mr. Gibson wrote that he didn't want to issue a Quality Assurance Report because one of the samples came out below the legal limit on retest. He just wanted to draft a "deviation" so it could be brushed under the rug.
Forrest Davis, the Quality Assurance Coordinator for the DPS, agreed. He didn't like the idea of a Quality Assurance Report because the data from the retests would be included in it. Better just to pretend it never happened, right? Interestingly enough, Mr. Davis did note that the retest did not include any volatiles that would have suggested a degradation of the sample.
Junk science is still junk science, no matter what you call it.
Crime labs should not be under the control of the police or any agency involved in the investigation or prosecution of crime. These labs should be under independent control to remove the inherent bias found in crime labs run by the police.