The Blood Alcohol Test

      Under California's criminal jury instruction scheme, CALJIC 16.831 is the standard jury instruction on the definition of "under the influence."  It does not state that alcohol impairment must be to an "appreciable degree," but that is a correct statement of the law. People v. Schoonover (1970) 5 Cal.App.3d 101, 105; People v. Weathington  (1991) 231 Cal.App.3d 69, 78.    

      Several Superior Court Appellate Department opinions disagree on the effect of a prosecution expert's admission of a possibility that the defendant's Blood Alcohol Concentration ("BAC") was below the statutory legal limit. In People v. Campos (1982) 138 Cal.App.3d Supp. 1,  the court found that direct blood analysis with a 0.10% result, accurate to ±.005% according to the prosecution expert, was not legally sufficient evidence to prove a BAC of 0.10% or more. Id. at 4-5.  In People v. Rangel (1988) 204 Cal.App.3d Supp. 10, the court held that a breath analysis with 0.10% and 0.11% results, accurate to ±0.02% according to the prosecution expert, is legally sufficient evidence of a BAC of 0.10% or more. Id. at 13-15.  However, another California disagrees fundamentally with the idea that prosecution admissions of possibilities of BAC's below 0.10% have anything to do with sufficiency of the evidence. In People v. Randolph (1989) 213 Cal.App.3d Supp. 1, the Ventura County Appellate Department held that it is not a reviewing court's proper role to say that a 0.10% breath test result, with a prosecution admitted error of ±0.01%, must lead to a reasonable conclusion that the defendant's blood alcohol level was below 0.10% at the time of driving. Id. at 10-11.  Thus it was for a jury to decide whether or not such an interpretation of the evidence was reasonable for purposes of the circumstantial evidence instruction.  Id. at 6. Another case held that, where the defendant's alcohol concentration was measured at 0.089% and 0.090% at the time of the test, it was error to instruct the jury that the defendant is presumed to be under the influence if he has 0.10% or more at the time of the test.   People v. Wood (1989) 207 Cal.App.3d.Supp. 11, 15. 

      The Rangel decision has its own problems, primarily with its use of defective mathematics. As pointed out by Dr. Gerald Simpson⁶, the Rangel Court took judicial notice of elementary probability theory and attempted to apply it to breath test results, and thus the application of an average value to an individual is an elementary probability determination with serious limitations. If just the average value is assumed for an individual, the probability that it is correct for some individual in a normally-distributed sample is about 8%....According to Dr. Simpson, the appropriate approach is to set the required confidence level, e.g., 95% for a reasonable doubt standard, and treat all values within this range as having an equal probability of being correct when applied to an individual. This is because it is simply not known which value within the selected range is correct; all values in this range are uncertain. Further, Dr. Simpson, opines that when this elementary probability theory applied to the situation in hand, one cannot assume that any particular value in the full range of values is more probable because even the most probable value, the average, has a likelihood of only 8%.  

      Contrary to Dr. Simpson's views, Judge Woods of the Rangel Court looked at each value within the range and asked what the significance is of the likelihood that the value is correct.  With the most likely value has a likelihood of only 8%, a problem arises with Judge Woods' method of giving significance to the likelihood of individual values. Judicial notice should not be used where it allows for the trusting one opinion of the meaning of the evidence over another opinion because things judicially noticed are not supposed to be disputable. Witkin, Chapter II in California Evidence, 3d ed. (West Group, 1986), Vol. 1, §80.  The error in Rangel was in taking judicial notice of something that is so difficult for laymen to understand and properly apply, and the mere fact of having an expert explain elementary probability theory to the court makes the theory and its application nor properly judicially noticeable.  

      In deciding Rangel, the court reaffirmed its decision in Campos, but distinguished Campos on the fact that the likelihood of Rangel having an alcohol concentration of 0.099% was

only 4.35%, whereas the likelihood was 50% that Campos was under 0.10%. Properly, the 4.35% probability should not be compared with Campos' 50% probability, but instead should be compared with the maximum 8% probability that any other particular BAC in the range is correct. Rangel was more than half as likely to have an alcohol level of 0.099% as he was likely to have 0.105% (the average of 0.10% and 0.11%, 8% likely). He was more than half as likely to be innocent as he was to be guilty. 

      Blood tests for alcohol are generally perceived to be more accurate than breath or urine tests, primarily because of the belief that breath tests and urine tests are indirect methods of determining alcohol levels in blood requiring conversion.  Because of the perceived reliability of blood tests, trials challenging the validity of blood tests occur less frequently than other methods of alcohol testing. The best defense to blood cases frequently rests inside blood vials because alcohol levels in blood vials can actually increase on their own via the process of fermentation.

      The blood sample is usually obtained from the antecubital vein, which is on the inside of the elbow.  Prior to drawing any blood the puncture site needs to be sterilized. Alcohol should not be used to sterilize the area.  Obviously, the area should not be cleaned with rubbing alcohol, but it occurs because hospital personnel are not always as attuned to the needs of law enforcement as are the on-call phlebotomists who work regularly with the police. If it does occur, it leaves open a big challenge to the test results as alcohol was present upon the needle being injected into the vein.  It is also a possible error for the prosecution if the blood is drawn from an artery instead of a vein, such as in California. [See, e.g., 17 California Code of Admin. Regs. §1219.1 (a), "Blood samples shall be collected by venipuncture from living individuals as soon as feasible after an alleged offense."  Thus, the client should be interviewed and their arm should be inspected at the draw site as soon as possible to determine if the blood was drawn properly from a vein.  

      After the sample is obtained, it must be mixed with an anticoagulant and a preservative. [See, e.g., 17 California Code of Admin. Regs. §1219.1(e)(2) ("The blood shall be mixed with an anticoagulant and a preservative"; also see Florida Admin. Code 11D-8.012(3) ("Immediately after collection, the tube must be inverted several times to mix the blood with the preservative and anticoagulant"). The purpose of mixing the sample is to adequately disburse the anticoagulant and preservative chemicals into the blood sample in order to ensure proper disbursal in the sample as the chemicals cannot work if they are not mixed into the blood they are meant to preserve.  This should be testimony elicited from the DUI practitioner's expert witness. Testimony should be elicited from the person drawing the blood on whether they used a mixed the sample adequately as a gentle rocking of the vial does not suffice to adequately disburse the chemicals into the blood. Further, the evidence envelope, where the blood vial is ultimately placed, usually has instructions on the envelope on how to draw and handle the sample, including the steps invariably discusses mixing the sample.

      The Phlebotomist should be examined lightly because they are witnesses are often sympathetic witnesses to jurors, and it is unlikely that the witness will have a specific recollection of the actual blood draw of the defendant.becuase the the encounter is brief and  performed countless blood draws since the taking of the blood of your client. Even if the person was qualified to draw blood and performed the task in the proper method, there is still some information that can be obtained in the cross-examination. This information may later proof fruitful when cross-examining the state's chemist.

      Once the amount of blood drawn has been determined, an accounting of the blood can then take place. The crime laboratory usually measures the sample size, and the same tracking takes places with respect to the amount of blood sent to the defense's independent lab. Sometimes those two samples added togethre do not add up, and missing blood is an issue the can be attacked on what is not there.  The DUI defense practioner should keep in mind that  the prosecution maintains the burden of roof beyond a reaosnable dount that the sample was not compromised. In re Winship (1970) 397 U.S. 358, 364. 

      Usually the prosecution can readily establish the chain of custody of the blood sample by only establishing that it is unlikely that the blood sample was tampered with or contaminated. See People v. Lach (Ill. App. 1998) 707 N.E.2d 144, 149. This then permits introduction of the subsequent blood alcohol test results. Gaps in the chain of custody typically go the weight of the evidence and not its admissibility. See Nimmons v. State (Fla. App. 2002) 814 So.2d 1153, 1155.

There are limits as to what omissions in the chain of custody the courts will forgive: See State v. Williams (South Carolina 1990) 392 S.E.2d 181, 182, [inadequate chain of custody held where no hospital personnel could identify who labeled, sealed and transported defendant's blood, especially when the defendant's E/R record was initially mislabeled]; Culver v. State (Ind. 2000) 727 N.E.2d 1062, 1068, [no chain of custody established where the record was silent as to the presence of any official during the blood draw].  

      Because of these types of holdings, the DUI defense practitioner should never stipulate to the chain of custody under the guise of "saving time," and should put an objection on the record to the chain of custody of the blood test evidence. The main reasons for not stipulating to the chain of custody of the blood sample are:  The prosecution might not be able to establish the chain of custody. This is the home run. The case may end right here; and often there are multiple witnesses involved in the chain of custody: the person who drew the blood, the person who picked up the blood from the evidence locker, the person who logged the evidence in at the laboratory, the person who opened the evidence envelope, the analyst, etc. 

      Contesting the chain of custody takes no effort on the part of the defense. Proving it does take energy on the part of the prosecutor who has the duty to lay the foundation.  Once the chain of custody has been established, any peripheral issues such as any discrepancy as to how much of the sample was the tested go to the weight of the evidence and not its admissibility. See Self v. State (Georgia App. 1998) 503 S.E.2d 625, 628, [discrepancies in the amount of blood drawn or tested goes to weight of evidence.)  


Fermentation in the Blood Sample makes the test nonsense

      The focus of the attack on blood tests is fermentation because alcohol can be produced on its own in a blood vial when glucose in the blood combines with microorganisms can cause the endogenous production of alcohol, and advanced blood alcohol testing procedures like gas chromatography are incapable of determining where the alcohol being analyzed originated. Thus, fermentation is an attack that is available in almost every case involving a forensic blood alcohol analysis. One way to prevent fermentation is to add a preservative such as the common sodium fluoride, usually 20 mg in a 10 ml blood tube. Most forensic laboratories rely on the manufacturers to insure that the preservative is in the blood vials, while at the same time most state laws require that a preservative be used without specifying a required amount. the failure of the prosecution to prove the existence of any preservative may mandate exclusion of the test results.  See State v. Bosio (Wash. App. 2001) 107 Wn. App. 462, 466-468, [evidence excluded as state failed to prove blood sample had a preservative.]  Thus, the prudent DUI defense practitioner should have an independent laboratory test not only for BAC but also for preservative content;   If the results from the independent laboratory show less than 1% sodium fluoride in the sample, the defense can attack the test result as being unreliable. A fermentation defense can still be presented even if the results from the independent laboratory revealed a proper amount of preservative in the blood sample. This is because fermentation can take place even with a proper amount of sodium fluoride if the sample is not promptly refrigerated. [Chang and Kollman, The Effect of Temperature on the Formation of Ethanol by Candida Albicans in Blood (1989).34 Journal of Forensic Sciences 105.  See also Rafferty v. State (Florida App. 2001) 799 So.2d 243, 248, [no refrigeration for eight days].
 
      If challenging the admission of a blood sample taken in a medical facility on privacy grounds is possible, that is the first attack on the blood sample that should be made.  Federal HIPAA laws prevent the disclosure of personal medical records and information without the person's written consent. See, e.g., 42 U.S.C. 201 et seq.  More specific to law enforcement, medical records may not be used in a criminal proceeding absent a court order or good cause. The United States Supreme Court held that a public hospital violated the Fourth Amendment when it turned over to the police urine test drug results from pregnant patients. [Ferguson v. City of Charleston (2001) 532 U.S. 67, 85-86, [the Fourth Amendment's general prohibition against consensual, warrantless, and suspicionless searches necessarily applies to such a policy.] The Georgia Supreme Court reached a similar result regarding the privacy rights in a patient's medical records based on state law. See King v. State, (Georgia 2000) 535 S.E.2d 492, 497.  Thus under HIPAA and the Court's common law, it could be argued that the intoxication treatment is substance abuse treatment.

      Without any arguments precluding the admission of blood alcohol evidence, analyzing the blood properly should be the area focused on.  There are two threshold determinations are necessary in the process of reviewing the analysis of a blood alcohol test result: Was the sample analyzed by a forensic laboratory or by a hospital or clinical laboratory?; and was whole blood was analyzed, or was plasma or serum analyzed?  Generally, a forensic laboratory will perform a gas chromatography test on whole blood, and a hospital or clinical laboratory will frequently perform immunoassay testing on plasma or serum.

      A forensic laboratory rarely performs analysis on anything other than whole blood. However, occasionally a hospital or clinical laboratory will test the alcohol content of plasma or the serum portion of the blood sample instead of whole blood.

      The primary reason forensic laboratories test whole blood is that the law prohibits driving with a certain amount of alcohol in blood. Blood is uniformly interpreted to mean whole blood. If a plasma or serum sample is analyzed it must be converted to a whole blood value.  See Newcomb v. State (Ind. App. 2001) 758 N.E.2d 69, 72, [no evidence was introduced of the conversion ratio of the reported serum alcohol value to a whole blood alcohol level—conviction reversed because the statute requires a whole blood alcohol value.]  Hospitals perform alcohol analysis on plasma and serum samples because it allows for a testing methodology that is quicker and easier to perform than the gas chromatography method used by forensic laboratories. Plasma and serum alcohol values are almost always higher than the alcohol values in the same sample of whole blood.  Whole blood is blood without any separation of blood cells or without any clotting done to the blood. The concentration of alcohol in a person is highest wherever the water content is highest. Since both serum and plasma have essentially the same water content, the alcohol content in a serum sample is virtually identical to the alcohol content in a plasma sample. whole blood tends to have a lesser concentration of alcohol in it than plasma or serum in the same blood sample. A serum or plasma blood value is converted to whole blood value by divided the serum or plasma value by a whole blood value, usually by an average ratio number, such as 1.16, or a range of possible values.  The ultimate issue to put to the prosecution's expert is that you really do not know what the true value of the defendant's blood alcohol level is. At lower alcohol levels this may be all that you need to get the defendant under a per se limit.

      The current trend in technology for blood sample analysis is gas chromatography. It can be performed either by direct injection or by headspace analysis. Direct injection inserts part of the blood sample into the chromatograph. Headspace gas chromatography measures a portion of the vapor above the blood sample.  It uses Henry's Law providing for a proportional relationship between liquid and the air above the fluid. Attacks on the blood test performed by a gas chromatograph need to establish at least two points: since the sample we are dealing with is microscopic, it does not take much of a problem with the testing process sample to greatly skew the result; and that blood sample was not given individualized attention as it was just one of numerous samples that were tested that day and lacking individualized attention, the possibility of error looms.

 One universal fact of all BAC testing is that once absorbed in the blood stream, alcohol distributes in the body to the various organs and tissues in proportion to their water content and blood supply.  About 95 to 98% of the alcohol consumed is metabolized via enzymatic oxidation, and the remainder is excreted unchanged in other body matrices. Breath, blood and urine are the predominate specimens in DUI/DWI investigations, with recent interest building in using saliva. However, the scientific literature regarding the determination of impairment or influence almost universally correlates performance with a BAC. Since whole blood is generally cited or assumed, analysis of alternative specimens must include the appropriate conversion to the corresponding BAC.  All specimens vary from blood in composition and water content. Since alcohol does not bind with plasma proteins, the concentration in body fluids such as urine and saliva is higher than that in whole blood, which contains cellular materials. Saliva and urine are roughly 99% w/w water, whereas whole blood is about 80% w/w water. Thus, a measured alcohol concentration in urine or saliva, must be converted to the whole blood legal standard. 

      Whole blood is the only specimen that can give a reliable BAC It is the whole blood that travels through the body to the brain, where alcohol concentration in the blood and the brain are in equilibrium. Since alcohol effects are manifested in the brain, whole blood is used to correlate a particular alcohol level with projected alcohol effects. The goal of every scientist is to collect and analyze a specimen best suited for the reliable interpretation of the case, but this does not always happen because whole blood is not often collected. Other specimens must be converted to an equivalent whole blood alcohol. As such, these specimens are indirect alcohol measurements. Whenever an indirect method is used, the chance of error in the calculations increases dramatically. When an alternative specimen is analyzed, special care must be taken to understand the issues and to properly interpret the result.  Many States laws circumvent the conversion process by having a set conversion ratio. Thus, a the forensic chemist must consider the full range of conversion ratios for a particular sample matrix when performing a case analysis.  Whenever conversion ratios are used, there is a certain amount of analytical error as the conversion factor built-in to the statute may not be accurate when applied to all individuals.

            Whole blood is a complex mixture of red blood cells (erythrocytes), platelets (thrombocytes), and white blood cells (lymphocytes and phagocytes). Each type has a specific function: Red cells take up oxygen from the lungs and deliver it to the tissues; Platelets participate in forming blood clots; White cells (lymphocytes) are involved with immunity; and

Phagocyte cells ingest and break down microorganisms and foreign particles. Together with lymphocytes they make up the white blood cells.

      Whole blood, plasma, and serum are often used in the analysis of ethanol. Each substance is not only slightly different in function, but also different in water content. Plasma is the resulting liquid when all the red blood cells are removed from whole blood.

Serum is what is left when the proteinaceous clotting factors, such as albumin, globulin, fibrinogen and prothrombin are removed. Serum alcohol concentration (SAC), or plasma alcohol concentration (PAC), may be converted to a whole blood BAC by using a conversion ratio. the SAC/BAC (or PAC/BAC) ratio range is quite wide from 1.03 to 1.26. As a practical matter, blood cells are rarely used for the determination of alcohol content. When looking at a hospital laboratory report, where blood was the specimen tested, it is highly likely that the test was performed on serum. In this situation, the full range of conversion ratios should be considered. However, in a forensic laboratory, the specimen tested is whole blood, which is subject to clots and lower red cell count, both of which can result in a higher BAC.