The Minnesota DWI Case Of The Week is State v. Brazil, (Decided December 26, 2017, Minnesota Court of Appeals, Published) which stands for the proposition that the "uncertainty of measurement" of the Data Master Breath testing machine is not a sufficient basis to reverse a criminal conviction for Third Degree DWI even where the reported test result is a .16. This is a very bad case for the defense.
In Brazil, the Defendant consented to a breath test, and the DMT device measured and reported appellant's alcohol concentration as 0.16. The Defendant was charged with two counts of third-degree DWI with reference to the aggravating factor of an alcohol concentration of 0.16. The Defendant waived his right to a jury trial, and admitted under oath that he drank enough alcohol to affect his ability to safely drive a motor vehicle before he drove and crashed his car into a parked car. He also admitted that his alcohol concentration was 0.08 or more as measured within two hours of driving. The Defendant denied that his alcohol concentration was 0.16 or more, an element necessary to the gross-misdemeanor charges.
The state presented testimony from Karin Kierzek, a forensic scientist with the Minnesota Bureau of Criminal Apprehension (BCA). Kierzek testified that every DMT device in use in Minnesota comes to the BCA's lab annually for maintenance checks, calibration, and certification. All machines must provide results within the acceptable 0.003 or 3% margin of error in order to pass calibration. Kierzek also testified that DMT devices have a number of internal and external checks to ensure accuracy. These checks begin with having a trained operator administer the test. The operator observes the subject for at least 15 minutes to verify that the subject is not introducing mouth alcohol by burping, belching, or regurgitating. The DMT device tests itself by running a diagnostic test, which includes using air blanks to clear the sample chamber and ensure that there is no residual alcohol or measurable alcohol in the air surrounding the machine. The subject then provides two breath samples a minimum of three minutes apart to safeguard against measuring mouth alcohol, and a control sample runs between the two breath samples to determine if the instrument is working properly when it evaluates a known alcohol concentration. If the two breath samples from the subject are not comparable, the test results are deemed insufficiently reliable and retesting is suggested.
The DMT results showed that the DMT device used to test appellant's alcohol concentration went through the full sequence of checks and passed all of them. There is no indication of irregularity or malfunction. The air blanks produced readings of zero, meaning that the sample chamber was clear of alcohol. Appellant's first breath sample revealed an alcohol concentration of 0.164. The machine ran another air blank and a control sample with a target of 0.078, which produced a result of 0.077. Kierzek testified that the control sample reading was only 0.001 different than the known sample, which variance she testified was insignificant and meant that the machine was measuring alcohol accurately within tolerable limits. After the control-sample test, the machine ran another air blank, which again tested zero, and then appellant provided a second breath sample, which resulted in an alcohol concentration measurement of 0.175. One final air blank was run to clear the sample chamber and check the room air for any measurable alcohol. It also tested zero. Kierzek testified that appellant's final alcohol concentration was determined by taking the lower of the two reported sample results, 0.164, and dropping the third digit to reach a reported value of 0.16. This method of reporting "give[s] the most benefit to the subject," according to Kierzek's testimony. Based on her review, Kierzek opined that appellant's breath-test results were accurate.
Kierzek also testified that "[t]here is no perfect measurement" and no measurement can ever be absolutely accurate. She testified that there is an uncertainty-of-measurement range within which the tester could have confidence that a high percentage of results would fall. Factors that contribute to the uncertainty of measurement include the area in which the tests are performed, the instructions given by an operator, whether the subject is wearing cologne, and whether the subject has certain medical conditions. She testified that the uncertainty-of-measurement value "merely gives you a range of what you would expect to see given repeated samplings." For appellant's test in particular, Kierzek testified that, at the 99% confidence interval, the expected range of test results would be 0.1504 to 0.1886. The average from appellant's two breath-test results was 0.1695, and Kierzek testified that this is the "most likely result," and that repeated test results "would be symmetric around that point." She also agreed that, had appellant's breath been tested a third time, it could have fallen anywhere within the confidence interval that she identified, from 0.1504 to 0.1886, and agreed that a third test falling anywhere within that range is "a distinct possibility" that is not arbitrary or capricious. Appellant's counsel asked Kierzek whether she could "say that if [appellant's breath] was measured a third time . . . [the result] would be a .18 or if it would be a .15 ... [w]ithout speculating," to which Kierzek responded no.
The District Court found the Defendant guilty of Third Degree DWI i.e. having an alcohol concentration level of .16 or more and on appeal, the Appellant argued that the evidence was not sufficient, because the uncertainty-of-measurement range includes values below 0.16, meaning that some tests of appellant's breath—if enough were done—would be expected to fall below 0.16.
The problem with the Appellant's argument is that once a person has been convicted, the Appellate Court will invoke the rules designed to affirm the factual findings of the District Court. Or, as stated by the Minnesota Court of Appeals in this case:
"In considering the sufficiency of the evidence supporting a conviction, we thoroughly analyze the record "to determine whether the evidence, when viewed in a light most favorable to the conviction, was sufficient to permit the [factfinder] to reach the verdict which [it] did."
"We must assume that the factfinder 'believed the state's witnesses and disbelieved any evidence to the contrary.' State v. Moore, 438 N.W.2d 101, 108 (Minn. 1989). 'We will not disturb the verdict if the [factfinder], acting with due regard for the presumption of innocence and for the necessity of overcoming it by proof beyond a reasonable doubt, could reasonably conclude that" the defendant was guilty of the charged offense. Bernhardt v. State, 684 N.W.2d 465, 476-77 (Minn. 2004). We have referred to this as the "traditional standard of review.'
"The record contains no measurement of appellant's alcohol concentration lower than 0.16. While appellant argues that the state is required to prove his alcohol concentration within the uncertainty-of-measurement range, our case law has consistently rejected this argument when framed in terms of margin of error." ***
"While these earlier decisions were made in the context of implied-consent cases, they hold that the proponent of a breath test need not prove the measurement to have been absolutely and precisely correct. The proponent must show that 'the necessary steps have been taken to ensure reliability," and after that 'it is incumbent on the driver to suggest a reason why the [breath] test was untrustworthy'".
"The district court accepted the test result as adequate proof of appellant's alcohol concentration. It did so despite testimony that it is possible that a third test of appellant's breath might have revealed a reported result under 0.16. The record supports the district court's factual finding concerning appellant's alcohol concentration. The evidence is sufficient to support appellant's conviction of third-degree DWI."
The Court's opinion in this case is troublesome as it transfers the rejection of the "margin of error" argument in civil cases and applies it to a criminal case. In a civil case, the burden of proof is just the "preponderance of the evidence" or "more likely than not" standard. So the fact that a machine has a "margin of error" is not very important where you are just trying to decide if it is more likely than not that the subject was over the legal limit.
But in a criminal case, where the burden of proof is "beyond a reasonable doubt" then the uncertainty of measurement should apply to prohibit a conviction where it is known that the machine, testing the same sample repeatedly, is likely to give a result under the legal limit.
Moral Of The Story: Never waive your right to a jury trial!