Prima Facie Reliability 1
Some lawyers think that the reliability of an approved breath instrument analysis of blood alcohol concentration can be determined by a review of the printouts from the instrument alone. This hypothesis has no support whatsoever in international scientific literature(1). There is no empirical study that supports this hypothesis(2). No literature review has been published that supports this hypothesis.
The hypothesis is not scientific, because it is incapable of disproof; the general scientific community cannot acquire the Canadian approved instrument hardware or software, used by police services in Canada, to attempt to negative the hypothesis. The hypothesis is novel science. It is not good science.
The hypothesis is inconsistent with the scientific literature cited by the Supreme Court of Canada in R. v. St-Onge Lamoureux(3) and the scientific evidence considered by Parliament prior to Bill C-2.(4)
One would expect that anything published by the Canadian Society of Forensic Science would be based on science rather than faith. The Society's Alcohol Test Committee 2012 Position Paper(5), however, does not purport to be scientific literature; it is a rather a position paper that supports a law enforcement and government position that limits disclosure to persons accused of drinking and driving offences.
The Alcohol Test Committee of the Canadian Society of Forensic Science has stated the following in its 2012 Position Paper:
"The required quality control information which must be reviewed to assess the working order of an Al is typically produced during the subject breath testing procedure."
"In most cases the AI will automatically produce a printed record that supplies much of the infor- mation enumerated above."
"Review of these speciﬁed requirements is sufﬁcient to assess the accuracy and reliability of a subject's breath test results."
Respectfully, the position that the printed record, at time of use, is all that is needed to assess reliability, is inconsistent with evidentiary breath testing science as has been published for years.(6) The printed record is not sufficient to assess the accuracy and reliability of a subject's breath test results.
I note that the 2012 Position paper does not state that review of paper at time of use is sufficient to assess but rather that review of these specified requirements is sufficient.(7) Unfortunately many experts and Crowns seem to have misunderstood the distinction and are using the 2012 Position Paper as authority for the proposition that paper at time of use is sufficient for review as to reliability.(8)
Mouth alcohol, for example, has always been a problem in evidentiary breath testing. All evidentiary breath testing, if it is to have any probative value whatsoever, requires a an observation deprivation period of at least 15 minutes (in some jurisdictions 20 minutes). If a breath tech doesn't control for mouth alcohol, then inflated values are a very real possibility. Consider the following video of a subject with 116 to 120 mg/100 mls true underlying Blood Alcohol Concentration blowing into a 5000C after a sip of beer. Notice that the automatic system to detect mouth alcohol doesn't always work, especially if the subject has an underlying BAC below the inflated value. Notice how much difference it made. This huge error would not have been apparent in any paper produced by the approved instrument(9). No error tone and no "Invalid Sample" error message were generated. The error only becomes apparent from other information about the breath testing process. The error is only apparent if there is a breath room video showing the error in the breath tech's process(10) or there is evidence through examination-in-chief or cross-examination.
Contrary to what is stated in the CFS 8000C Training Aid (2013) at p. 80 of 238, a Diagnostics "pass" (on startup or Esc Esc D) and a self-breath test by the breath tech of zero DO NOT establish that the instrument is in proper working order and is capable of receiving a breath sample. The printouts from my 5000, in an experiment today, confirm that the instrument passed its cold startup Diagnostics (it got started), passed a stand-alone Diagnostics test (the test card below on the left) and passed a self-test by the operator at 000 mg/100mls (the test card on the right). This 5000, however, is the same instrument from my previous blogs about flooded sample chamber. Within a short time of startup, and these printed results, the DVM of the instrument, at rest, became 000. Every subsequent test today resulted in a "RANGE EXCEEDED" error. This instrument is not reliable and yet it produced passed diagnostics and a good self-test.(11)
Paper printouts at time of testing DO NOT establish proper working order or reliability of an approved instrument. Reliability can only be established if the reviewing scientist can see the results in context. Evidentiary breath testing is a process. Apparently normal test results may be prima facie evidence of reliability, but an expert must be able to see the full context of these printouts to give an opinion on reliability, or lack thereof.
(1) All References at the end of the 2012 ATC Position Paper are to its own journal, the Can. Soc. Forensic Sci. J..
(2) See Wigmore on Alcohol, Wigmore, James A., 2011, for abstracts of empirical studies on everything you wanted to know about alcohol. There are no empirical studies therein which support the larger ATC hypothesis in the Position Paper or the Crown perspective that paper printouts at time of use are sufficient to assess reliability.
(3) 2012 SCC 57, See references in "Authors Cited" and paragraph 26 of the decision to Hodgson, Brian T. “The Validity of Evidential Breath Alcohol Testing” (2008), 41 Can. Soc. Forensic Sci. J. 83.
(4) Evidence of Brian Hodgson, Chair, Alcohol Test Committee on June 12, 2007 to Standing Committee on Justice and Human Rights, Canada, Parliament, respecting Bill C-32 An act to amend the Criminal Code (impaired driving), later Bill-C2:
"These instruments are all automated instruments. They require operator involvement. But when strict protocols are followed and the instrument is working properly as per the recommended procedures, then the tests obtained, especially when they're 15 minutes apart, provide conclusive proof of the person's blood alcohol concentration at the time of testing."(italics added)
(6) Dubowski, Kurt M., Journal of Analytical Toxicology Vol. 18, October 1994
(7) I respectfully suggest that both review of paper at time of use AND review of the specified requirements are not sufficient. I will review the Position Paper "specified requirements" in a future blog entry.
(8) I note that Justice Watt's decision in R. v. Jackson, 2015 ONCA 832, at paragraph 136 does not state that paper at time of use is sufficient for review as to reliability.
(9) The printer was disabled during the time shown by this video.
(10) The video of the subject and the breath tech should cover the entire 15 minute observation period.
(11) There are a number of ways to produce a "Passed" Diagnostic on an unreliable instrument. For a humorous example see the video below:
Warning: These videos and images may not be used in Court. They are not evidence. The author is not an expert witness. These videos are for discussion among lawyers.