Roadside Breath Test Operators Should be Held to Standard of Substantial Compliance

by Patrick T. Barone

Patrick T. Barone is an adjunct professor at Cooley Law School where he teaches "Drunk Driving Law and Practice."  Mr. Barone is also the co-author of two books on DUI-related issues, including Defending Drinking Drivers (James Publishing), a well-known and highly respected multi-volume national legal treatise.  He is a frequent lecturer on trial practice and drunk driving defense tactics. He can be contacted on the web at:

Michigan law provides that a roadside breath test or preliminary breath test (PBT), may be used as the sole evidence to establish probable cause for arrest.1  The PBT may also be used as the sole evidence establishing probable cause in support of a warrant to draw blood.2  However, because “reasonable cause” is needed in most circumstances to administer a PBT, the arresting officer usually requests the breath test at the conclusion of his/her roadside investigation.

Because of the evidentiary importance of the PBT in the arrest and prosecution of drinking drivers, it may be desirable to seek suppression of the test prior to trial.  The most often-cited reason for suppression is a failure on the part of the arresting officer to administer the test according to the law.  In pertinent part, Admin R. 325.2655 states:

(2) A procedure that is used in conjunction with preliminary breath alcohol analysis shall be approved by the department and shall be in compliance with all of the following provisions:

*   *   *

(b) A person may be administered a breath test on a preliminary breath alcohol test instrument only after it has been determined that the person has not smoked, regurgitated, or placed anything in his or her mouth for at least 15 minutes.3

Based on this rule, the relevant question is whether the administering officer determined that no contaminants were in the driver’s mouth for at least 15 minutes prior to the test.  This is a critical determination because preliminary breath testing instruments are incapable of screening for mouth alcohol or other contaminating substances.  Thus, if mouth alcohol or contamination is present, the PBT will report either as if it was bodily alcohol.  This can result in improper arrest and conviction, the improper issuance of warrants for blood testing, and possibly jail time imposed for apparent bond violations.

When arguing for the suppression of the preliminary breath test result, it is important for practitioners to understand the theory and science of preliminary breath testing, and to then be able to effectively educate the judiciary.  What follows is a brief primer on these issues.

The Underlying Theory of Breath Testing

Evidentiary breath testing instruments use the science of infrared spectroscopy to measure the amount of alcohol in a driver’s breath.  In such instruments, the alcohol in a breath sample is quantified by absorbing infrared energy filtered to specified wavelengths according to Beer’s Law.4

Portable breath testing instruments are electrochemical detectors known commonly as fuel cell devices.  In these, the detection of breath alcohol is based on the electrochemical properties of alcohol.  A signal is generated when the alcohol exchanges electrons with an electrode inside the device.5  As such, fuel cells are electrochemical energy conversion devices. They produce electricity from external supplies of fuel and oxidant.  In the context of breath testing, the alcohol is the fuel and atmospheric oxygen is the oxidant.  In the fuel cell, the alcohol is converted to acetic acid so that for each molecule of alcohol, two electrons are produced.  Collectively the electrons produce an electrical current which is measured.  This measurement is converted into a quantitative breath alcohol level.6

Mouth  Alcohol

Mouth alcohol is one of the principal causes of error in breath alcohol analysis. Mouth alcohol can contaminate expired breath that is captured by the machine thereby elevating the results dramatically and erroneously. Any breath testing machine’s “computer” will assume that the sample tested is 100% alveolar air from the lungs and, accordingly, will multiply any error by 2,100, the blood/breath partition ratio, to obtain a blood alcohol concentration. Therefore, even a minute amount of undigested alcohol in the mouth can have a significant impact on the breath test reading.7

Many commentators have advocated strict procedures to minimize sources of error in breath testing, including mouth alcohol.8  The only sure way to eliminate [residual mouth alcohol] is to make sure the suspect does not take any alcohol for at least 15 to 20 minutes before conducting a breath test.  Remember, too, that most mouth washes, breath sprays, cough syrups, etc., contain alcohol and will produce residual mouth alcohol.  Therefore, it is always best not to permit the suspect to put anything in his mouth for at least 15 to 20 minutes prior to testing.”9
To combat mouth alcohol problems, police officers generally are required by regulation, or instruction, to observe the suspect carefully for 15 to 20 minutes prior to testing to ensure that the suspect does not belch, burp, hiccup, or regurgitate.

How Slope Detectors Attempt to Detect Mouth Alcohol

Evidentiary breath test instruments, such as Michigan’s DataMaster, attempt to contend with the problem of mouth alcohol through the employment of slope detection.  Once the initial sample acceptance criteria are met, the DataMaster begins measuring the breath sample’s alcohol content by taking a reading every ¼ second.  Each two data points are averaged, and compared with the prior two data points.  If there is a downward trend in the measurements, it is assumed that mouth alcohol is responsible, and the test sequence is aborted before a final reading is taken.  In this instance, the DataMaster should stop the testing and an evidence ticket should be printed out reading “INVALID SAMPLE.”10

Slope detection is possible only if the device in question is capable of three things: (1) instantaneous measurement, (2) storage of the measurements in memory, and (3) computer processing capabilities sufficient to run the mathematic algorithm necessary to access the slope.  Because the DataMaster using infrared energy as a means of measuring the alcohol in the breath sample, and because it can store the readings take, it meets both of these criteria.  Whether the DataMaster can reliably measure mouth alcohol is a different question, not discussed here.  Our point is only that the DataMaster has these capabilities.

One potential problem with portable fuel cell devices is their inability to monitor the slope of the breath sample as it is being received. As discussed, slope detection requires very rapid analysis.  Fuel cell devices are considerably slower than necessary because they do not produce real time responses or measurements.  Response times for a new fuel cell may be between three and four seconds.  Response times begin to decrease with age, and the reaction time for an older fuel cell may be between eight and 10 seconds.  This is because aging causes the fuel cell to fatigue, and one manifestation of such aging is slower reaction times.11  The portable breath test devices also do not have computer processing capabilities or memory capabilities necessary to slope detection.

How Breath Test Instruments Attempt to Rule-Out Contamination

Evidentiary breath testing instruments  - as indicated above, alcohol is quantified by essentially measuring how much light is absorbed as it passes through a breath sample.  The more light or light energy that is absorbed, the more alcohol that is thought to be present.  But infrared light is not specific for alcohol.  Because of this only certain frequencies of the light are actually measured.  In the DataMaster, the target frequency is 3.37.  However, certain molecules, like Acetone, also absorb heavily at 3.37, but may absorb more heavily at other wavelengths such as 3.44 or 3.5.  By testing the breath with different filters in the optical path, and then comparing the measurements at these different wavelengths, the machine can try to determine if something other than alcohol is present.  If the readings taken by the machine at different wavelengths are outside of the acceptable range, then the DataMaster should stop the testing, and print out an evidence ticket with the “INTERFERENCE DETECTED”12 error code.

Portable breath testing instruments – Fuel cells are generally non-specific for ethanol, and can potentially respond to other alcohols that will also act as fuel, such as methy-, isopropyl- and to acetaldehyde.13

A case that is often cited by the state to show that the preliminary breath test should not be excluded by the trial court is People v. Mullen.14  There the court ruled against suppression because the “defendant presented insufficient evidence in the hearing below that the presence of paper in his mouth, three minutes before administration of the PBT, would significantly call into question the accuracy of the PBT result so as to preclude a finding of probable cause.”15  It is clear that the court in that case was not properly educated by the defense attorney, and did not actually know any of the information contained herein.  Do not let a similar injustice occur to your clients.  Be sure to always educate the judiciary about both the law as well as the science of breath testing.


1.  MCL 257.625a(2)(a).
2.  See id.(allowing arrest solely upon PBT result) and U.S. Const. Am. IV; Const. 1963 Art 1 §11 (requiring probable cause for both arrest and issuance of a warrant).
3.  Admin R. 325.2655(emphasis added).
4.  Encyclopedia of Forensic Sciences, Vol. 1, page 88 (Academic Press 2000).
5.  Id. Vol. 2, pg. 734.
6.  Barone, Defending Drinking Drivers, pg. 2-57, James Publishing (2011 ed.)
7.  Id. See Dubowski, “Quality Assurance in Breath-Alcohol Analysis,” 18 J. Anal. Tox. 306 (October 1994) (identifying, inter alia, various “necessary safeguards for breath testing including a pre-test deprivation-observation period of at least 15 minutes); see also Coldwell & Grant, “A Study of Some Factors Affecting the Accuracy of the Breathalyzer,” 8 J. Forensic Sci. 149 (1963); Caddy, Sobell & Sobell, “Alcohol Breath Tests: Criterion Times for Avoiding Contamination by ‘Mouth Alcohol’,” 10 Behav. Res. Meth. & Instr. 814 (1987) (rate of mouth alcohol dissipation approximates exponential decline);  Denny & Williams, “Mouth Alcohol: Some Theoretical and Practical Considerations,” Alcohol, Drugs & Traffic Safety - T86, P. Nordzij & R. Roszbeack, eds. at 355-358 (1987); Kempe, “Study of Dissipation Rate of Ethanol from the Oral Cavity,” 20 Law & Order 7 at 74 (September 1972); Harding, et al., “The Effect of Dentures and Denture Adhesives on Mouth Alcohol Retention,” 37 J. Forensic Sci. 999 (July 1992); American Medical Association, Committee on Medical Problems, “Manual for Chemical Tests for Intoxication” (1959) (“True reactions with alcohol and expired breath from sources other than the alveolar air [eructation, regurgitation, vomiting] will, of course, vitiate the breath alcohol results”).
8.  See Dubowski, Kurt M., Quality Assurance in Breath Alcohol Analysis, supra, and Garriott, James C., Garriott’s Medicolegal Aspects of Alcohol, 5th Ed., 232 (2008)(both detailing procedures to avoid mouth alcohol interference).
9.  H. Cohen & J. Green, Apprehending and Prosecuting the Drunk Driver; a Manual for Police and Prosecution, §4.07(1) at 4-38 (1993).
10.  See Michigan Breath Test Operator Training Manual, Michigan State Police Special Operations Division Alcohol Enforcement Unit, ed. 2003, pg. 8-9
11.  Barone, Defending Drinking Drivers, pg. 2-57, James Publishing (2011 ed.)
12.  See Michigan Breath Test Operator Training Manual, Michigan State Police Special Operations Division Alcohol Enforcement Unit, ed. 2003, pg. 8-9.
13.  Medical-Legal Aspects of Alcohol, 4th Edition, p. 197 (2003).
14.  People v. Mullen, 282 Mich. App. 14 (2008).
15.  Id. at 24.