Archive for September, 2013

Does “Slurred Speech” Indicate Intoxication?

Monday, September 30th, 2013

As with the odor of alcohol on the breath, few police reports will fail to include an observation by the arresting officer that the arrestee exhibited "slurred speech". (See my earlier post, "Alcohol on the Breath: Evidence of DUI?"). The officer fully expects to hear slurred speech in a person he suspects is intoxicated, particularly after smelling alcohol on the breath, and we tend to "hear" what we expect to hear. And hearing it supplies the officer with corroboration of his suspicions.

Even assuming the honesty of the officer that the defendant’s speech was slurred, there is little evidence that this is symptomatic of intoxication. Impairment of speech is, for example, a common — and sober — reaction to the stress, fear and nervousness that a police investigation would be expected to engender; fatigue is another well-known cause. However, consider the following excerpt from Discover magazine:

"Bartenders, police officers and hospital workers routinely identify drunks by their slurred speech. Several investigative groups judged the captain of the Exxon Valdez oil tanker to be intoxicated based solely on the sound of his voice in his radio transmissions. But a team led by Harry Holien, a phonetician at the University of Florida, has found that even self-proclaimed experts are pretty bad at estimating people’s alcohol levels by the way they talk."

Hollien asked clinicians who treat chemical dependency, along with a group of everyday people, to listen to recordings made by volunteers when they were sober, then mildly intoxicated, legally impaired, and finally, completely smashed. Listeners consistently overestimated the drunkeness of mildly intoxicated subjects. Conversely, they underestimated the alcohol levels of those who were most inebriated. Professionals were little better at perceiving the truth than the ordinary Joes. "

He thinks his research could encourage police to be more wary of making snap judgments: Mild drinkers might come under needless suspicion…" Saunders, "News of Science, Medicine and Technology: Straight Talk", 21(1) Discover (Oct. 2000).


What Driving Symptoms Do DUI Cops Look For?

Wednesday, September 25th, 2013

Do you ever wonder just what the police are looking for when they’re out at night patrolling for drunk drivers?  And how you can avoid looking like one?

The following list of DUI driving symptoms, from a publication issued by the National Highway Traffic Safety Administration (DOT HS-805-711), is widely used in training officers to detect drunk drivers on the roads.  After each symptom is a percentage figure which, according to NHTSA, indicates the chances that a driver is under the influence of alcohol.  For example, NHTSA claims that its research indicates that “the chances are 65 out of 100″ that a driver who is straddling a center or lane marker has a blood-alcohol concentration of .10% or higher (the pre-.08% standard).

Turning with wide radius     65
Straddling center or lane marker     65
Appearing to be drunk     60
Almost striking object or vehicle     60
Weaving     60
Driving on other than designated roadway     55
Swerving     55
Slow speed (more than 10mph below limit)     50
Stopping (without cause) in traffic lane     50
Drifting     50
Following too closely     45
Tires on center or land marker     45
Braking erratically     45
Driving into opposing or crossing traffic     45
Signalling inconsistent with driving actions     40
Stopping inappropriately (other than in lane)     35
Turning abruptly or illegally     35
Accelerating or decelerating rapidly     30
Headlights off     30

Just to reenforce this as a mathematical science to impress to juries, NHTSA further claims — and officers are taught — that there is also a quick-and-easy formula for multiple symptoms:  “When two or more cues are seen, add 10 to the highest value among the cues observed”.

Of course, if these suspiciously precise figures are to be believed, then almost half of the folks who tailgate you every day are drunk – and almost half of the time you don’t brake smoothly you are, too.  Further, only 60% of drivers “appearing to be drunk” to the officer actually are. (Query:  How does a driver “appear to be drunk” to an officer following 100 feet behind? Slurred speech?  Alcohol on his breath?)  And speeding — one of the most common reasons for pulling DUI suspects over — is not even on the list. 


If Your Blood Sample Ferments Before It’s Analyzed, It Produces…Alcohol!

Friday, September 20th, 2013

In a post a few days ago, I discussed the necessity of adding an anticoagulant to blood samples taken in DUI cases to avoid an elevated blood alcohol result due to clotting. I also mentioned in that post that if a preservative were not also added to the sample, fermentation can take place. That prompted a number of queries about fermentation in blood alcohol analysis….

Blood is an organic substance and, like any organic substance, will decompose because of enzymes and bacterial action. One of the results of decomposition is that alcohol is created in the blood. In a blood sample originally containing no alcohol, decomposition and resulting fermentation can cause a reading far above the legal limit, depending upon the stage of decay. To stop or at least slow down this process, a preservative (commonly sodium fluoride) is added to the sample; it should also be refrigerated until analyzed by the crime lab.

The amount of preservative, of course, must be sufficient, and it must be actively mixed into the blood sample. While it is common among law enforcement agencies to use 20mg of sodium fluoride, many experts view this as insufficient. See, for example, Dick and Stone, “Alcohol Loss Arising from Microbial Contamination of Drivers’ Blood Specimens”, 34 Forensic Science International 17 (1987). Further, although labs often take a week or more before receiving and analyzing blood samples, normal levels of sodium fluoride will render the sample stable for only about two days. Kaye, “The Collection and Handling of the Blood Alcohol Specimen”, 74 American Journal of Clinical Pathology 743 (1980).

To complicate things further, not all microorganisms are affected by sodium fluoride. As researchers have discovered, one such microbe that is commonly found in the human body is Candida albicans:

It has been shown that several microorganisms occasionally found in blood specimens are capable of producing ethyl alcohol. Although (one study) found that sodium fluoride effectively inhibited alcohol production from a variety of microorganisms, one – Candida albicans — appeared to be unaffected by the addition of sodium fluoride. C. albicans is commonly found in man, usually in the oral cavity and digestive tract, and less commonly in the vaginal tract of women…. The legal ramifications of this are obvious. If an oganism common to man is capable of producing ethyl alcohol in stored blood, the question arises: Are the results of alcohol analysis reflective of an individual’s level of intoxication or of possible fermentation?    

Change and Kollman, “The Effect of Temperature on the Formation of Ethanol by Candida albicans”, 34(1) Journal of Forensic Sciences 105 (1989). 

So, for those wondering, “If I’m arrested for DUI, should I take a breath test or a blood test?”, the answer is….neither is reliable — although breath is clearly the least reliable. Coagulation and fermentation are just two of many problems inherent in blood analysis for alcohol. For an example of another, see “How do I know the Blood They Tested as Mine?”. But I’ve written many posts about the unreliability of breath testing as well, such as: “Breathalyzers — and Why They Don’t Work”; “Why Breathalyzers Don’t Measure Alcohol”; “The Mouth Alcohol Problem”; and “How to Fool the Breathalyzer”.

Take your choice…..The results will probably be “close enough for government work”.


What If the Blood Sample Clots Before It’s Analyzed?

Monday, September 16th, 2013

When a blood sample is taken from a DUI suspect for later analysis, it is usually done in one of two ways. The suspect may have the blood drawn at the police station by a technician, using a prepared kit containing a vial, or it may be taken by a nurse at a medical facility. In either event, it is critical that the vial in which the blood is contained is sterile and contains two things: a preservative and an anticoagulant.

The preservative, in conjunction with refrigeration, is to prevent the blood from fermenting — and thereby producing alcohol in the vial. The anticoagulant is to prevent the blood from coagulating, or clotting.

Why are we concerned about coagulation of the blood? Blood is made up of a mixture of solid particles supended in a liquid. The solid particles consist of red blood cells, white blood cells and clotting platelets; the liquid portion is called plasma. (The percentage by volume of the solid particles to the liquid is called the hematocrit of the blood: a hematocrit of .47, for example, would indicate that the individual’s blood consists of 47 percent solid particles (cells and platelets) and 53 percent plasma.) When blood clots, the liquid portion separates from the solid portion (blood cells and clotting platelets) This will be seen in the sample vial as a red clump at the bottom (cells) with a yellowish liquid on top (now called serum). When this sample is tested at the laboratory, usually days later, it is the serum that is tested for alcohol content; the clotted cells at the bottom are not included.

So what? Well, alcohol is attracted to water — that is, it is soluble in water. And since serum is a liquid and contains water, and alcohol is attracted to water, the serum in the blood sample will contain a higher percentage of alcohol than in the whole blood sample. The higher the percentage of serum in the sample being tested, the higher will be the blood alcohol concentration (BAC). Put another way, if two subjects have the same BAC in their bodies but the blood sample from one has clotted and so has a higher percentage of serum, that person’s “sample” will show a higher BAC.

Unfortunately, it is not uncommon for blood samples collected by police agencies or hospital personnel to contain no anticoagulant, or to contain insufficient amounts of the chemical. (And, of course, we have the emerging practice of just letting the police officer himself perform the blood draw and sample preservation out on the highway.) An additional problem is that the kits used by technicians usually contain a vial already containing a preservative (commonly sodium fluoride) and an anticoagulant (commonly potassium oxalate) in powder form at the bottom. However, when the blood is added to the vial, the technician does not shake it — and the chemicals are not mixed with the blood. Result: coagulation — and a falsely high blood alcohol result.


Government Crime Labs Paid for Evidence Resulting in Convictions?

Monday, September 9th, 2013

I've often pontificated in the past about the unreliability and inaccuracy of breath and blood alcohol analysis, and even questioned the competence and honesty of law enforcement crime laboratories.  Most people assume that evidence produced by these government labs are objective and trustworthy, reasoning that there is nothing to be gained by providing false results to facilitate convictions.

Now, one of my favorite political bloggers on the internet, Radley Bilko, has written an article on the Huffington Post pointing to an academic study that corroborates the questionable ethics and evidence coming from those labs:

New Study Finds That State Crime Labs Are Paid Per Conviction

I've previously written about the cognitive bias problem in state crime labs. This is the bias that can creep into the work of crime lab analysts when they report to, say, a state police agency, or the state attorney general. If they're considered part of the state's "team" — if performance reviews and job assessments are done by police or prosecutors — even the most honest and conscientious of analysts are at risk of cognitive bias. Hence, the countless and continuing crime lab scandals we've seen over the last couple decades. And this of course doesn't even touch on the more blatant examples of outright corruption.

In a new paper for the journal Criminal Justice Ethics, Roger Koppl and Meghan Sacks look at how the criminal justice system actually incentivizes wrongful convictions. In their section on state crime labs, they discover some astonishing new information about how many of these labs are funded.

"Funding crime labs through court-assessed fees creates another channel for bias to enter crime lab analyses. In jurisdictions with this practice the crime lab receives a sum of money for each conviction of a given type. Ray Wickenheiser says, ‘‘Collection of court costs is the only stable source of funding for the Acadiana Crime Lab. $10 is received for each guilty plea or verdict from each speeding ticket, and $50 from each DWI (Driving While Impaired) and drug offense.’’

"In Broward County, Florida, ‘‘Monies deposited in the Trust Fund are principally court costs assessed upon conviction of driving or boating under the influence ($50) or selling, manufacturing, delivery, or possession of a controlled substance ($100).’’

"Several state statutory schemes require defendants to pay crime laboratory fees upon conviction. North Carolina General Statutes require, ‘‘[f]or the services of’’ the state or local crime lab, that judges in criminal cases assess a $600 fee to be charged ‘‘upon conviction’’ and remitted to the law enforcement agency containing the lab whenever that lab ‘‘performed DNA analysis of the crime, tests of bodily fluids of the defendant for the presence of alcohol or controlled substances, or analysis of any controlled substance possessed by the defendant or the defendant’s agent.’’

"Illinois crime labs receive fees upon convictions for sex offenses, controlled substance offenses, and those involving driving under the influence. Mississippi crime labs require crime laboratory fees for various conviction types, including arson, aiding suicide, and driving while intoxicated.

"Similar provisions exist in Alabama, New Mexico, Kentucky, New Jersey, Virginia, and, until recently, Michigan. Other states have broadened the scope even further. Washington statutes require a $100 crime lab fee for any conviction that involves lab analysis. Kansas statutes require offenders ‘‘to pay a separate court cost of $400 for every individual offense if forensic science or laboratory services or forensic computer examination services are provided in connection with the investigation.’

"In addition to those already listed, the following states also require crime lab fees in connection with various conviction types: Arizona, California, Missouri, Tennessee, and Wisconsin."

Think about how these fee structures play out in the day-to-day work in these labs. Every analyst knows that a test result implicating a suspect will result in a fee paid to the lab. Every result that clears a suspect means no fee. They're literally being paid to provide the analysis to win convictions. Their findings are then presented to juries as the careful, meticulous work of an objective scientist.

No wonder there have been so many scandals. I'm sure we'll continue to see more.

If you were arrested for drunk driving, would you want your blood sample analyzed by a crime lab that is going to be paid if you are successfully convicted?  Or a lab technician from that same lab to testify in court that the breathalyzer used on you had been correctly maintained and calibrated?