DUI Detection: The Unreliability of Field Sobriety Tests and Blood Alcohol Testing

DUI Detection: The Unreliability of Field Sobriety Tests and Blood Alcohol Testing

“Would you mind stepping out of the vehicle?”   

Standardized field sobriety tests (“SFSTs”) were developed in the 1970s through research sponsored by the National Highway Traffic Safety Administration (“NHTSA”).  NHTSA trains police officers nationwide in administering the SFSTs, which consist of several alleged-to-be accurate methods of determining whether a person is intoxicated. 

The field sobriety tests require a suspected drunk driver to exit the vehicle, stand in front of the officer, and

(1) gaze at an object that the police officer holds about 12 inches from the suspect’s face and continue to gaze at the object as the officer moves it back and forth across the suspect’s line of sight, all the while controlling any involuntary jerking of the eyes (“horizontal gaze nystagmus”);

(2) perform the same test, this time while the officer moves the object up and down before the suspect’s line of sight (“vertical gaze nystagmus”);

(3) walk a straight line for nine paces, touching heel-to-toe while counting out loud, then turn around without swiveling (take several small steps to turn around rather than swiveling) walk back nine paces in the same manner without stepping off the previously walked line while counting out loud; and

(4) stand on one leg for 30 seconds while counting out loud “one thousand one, one thousand two,” etc., without extending the arms outward to maintain balance.

These SFSTs have been termed (1) “horizontal gaze nystagmus” test (“HGN”); (2) “vertical gaze nystagmus” test (“VGN”); (3) “walk and turn” test (“WAT”); and (4) “one leg stand” test (“OLS”). 

We realize it might be difficult to perform the tests while sober.  So it should come as no surprise that, nearly twenty years ago, in the case of United States v. Horn, 185 F. Supp. 2d 530 (D. Md. 2002), a federal district court held that SFSTs were inadmissible as direct evidence of intoxication because, among other reasons, SFSTs are not generally accepted within an unbiased technical or scientific community.  The court emphasized that, despite the government’s expert witness testimony, NHTSA publications, and publications of communities of police officers and prosecutors, the SFSTs were not as reliable as asserted by those then-unchallenged sources.  The court did permit use of SFSTs to determine whether probable cause exists to arrest for DUI, however, reasoning that “[p]robable cause determinations turn on practical, nontechnical determinations.”  Id. at 533 n.1 (citing Beck v. Ohio, 379 U.S. 89, 91 (1964)). 

The court proceeded to downgrade the officer’s testimony about the defendant’s performance on the SFSTs to “nontechnical” lay witness opinion testimony under Rule 701 of the Federal Rules of Evidence.  NHTSA continues to tout the alleged scientific accuracy of SFSTs, however, pointing to “validation” studies that conclude that officers can even predict a person’s blood alcohol level (“BAC”) through their performance on SFSTs (!).  NHTSA’s claims are unscientific.  One of the defense expert witnesses in Horn, Dr. Harold Brull, a licensed psychologist, criticized the studies as

“merely summary reports, without foundation …” and suffering from “a serious methodological flaw,” in that the tests were done on actual motorists stopped by officers because their driving was unsafe, leading the officers automatically to suspect that they were intoxicated.  (citation to NHTSA validation studies omitted)  Use of this population likely will produce results that Brull characterized as “highly inflated.” 

Horn, 185 F. Supp. 2d at 544. 

Dr. Brull had “many years experience consulting in connection with the design and implementation of procedures to measure human attributes, especially in employment settings. He has designed and evaluated tests and procedures measuring human characteristics for over twenty years.”  Horn, 185 F. Supp. 2d at 543-44 (citation to trial court document omitted). 

NHTSA’s unverified assertions about the accuracy of SFSTs might be a relic of the superseded legal standard for admission of expert witness testimony in many jurisdictions—the Frye standard—which until 2016 was the legal standard in the District of Columbia, as well (the Frye/Dyas standard).  In 2016, however, in the case of Motorola Inc. v. Murray, 147 A.3d 751 (D.C. 2016) (en banc), the D.C. Court of Appeals adopted the federal legal standard—Rule 702 of the Federal Rules of Evidence, which requires the following: 

“A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if:

  • The expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue;
  • The testimony is based on sufficient facts or data;
  • The testimony is the product of reliable principles and methods; and
  • The expert has reliably applied the principles and methods to the facts of the case.”

The SFSTs haven’t been validated in the D.C. Courts under the federal standard.  See, e.g., Townsend v. District of Columbia, 183 A.3d 727, 733 n.10 (D.C. 2018) (emphasizing that D.C. Court of Appeals opinions validating the HGN test “generally predate this court’s decision in Motorola Inc. v. Murray, 147 A.3d 751 (D.C. 2016) (en banc), in which we abandoned Frye and adopted the test for admission of expert testimony set out in Rule 702 of the Federal Rules of Evidence and in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993): ”whether the reasoning or methodology underlying the testimony is scientifically valid” [emphasis added]  Motorola, 147 A.3d at 754 (quoting Daubert, 509 U.S. at 592-93)).

The superseded Frye/Dyas standard, in contrast, is much less rigorous: “general acceptance among practitioners in the relevant field of scientific inquiry.”  Townsend, 183 A.3d at 733 n.10.  One of the expert witnesses in the Horn case, Dr. Spurgeon Cole, discussed the problem with assuming the validity of the SFSTs based solely on the fact that law enforcement is trained by NHTSA to use them:

“Because of its widespread use, the [SFST] battery has been assumed to be a reliable and valid predictor of driving impairment.  NHTSA has done little to dispel that assumption [emphasis added].  Law enforcement cannot be blamed for its use of the [SFST] battery.  Training documents refer to NHTSA reports and provide what appears to be supporting evidence for the validity of the [SFST] battery.  In addition, there is little doubt that individuals who have high BAC levels will have difficulty in performing the [SFST] battery.  However, what the law enforcement community and the courts fail to realize is that the [SFST] battery may mislead the officer on the road to incorrectly judge individuals who are not impaired.  The [SFST] battery to be valid must discriminate accurately between the impaired and non-impaired driver.  NHTSA’s own research on that issue … has not been subjected to peer review by the scientific community….” 

Horn, 185 F. Supp. 2d at 541-42 (citations to trial court documents omitted). 

Meanwhile, after the arrest …

In the District of Columbia, a suspected drunk driver’s legal BAC is the BAC measured at the time of testing.  A chemical test administered on the driver may show a higher BAC than the BAC at the time of the alleged offense, however, because BAC reaches a peak sometime after the last drink (i.e., the scientifically-observed “blood alcohol curve”).  Elimination rates have been shown in laboratory testing to vary widely among individuals—from 30 minutes to as long as 4 hours.  Source: Paul C. Giannelli, et al., 2 Scientific Evidence § 22.04 (2020).

Nor does BAC rise and fall at a constant rate (i.e., in a perfect bell curve shape).  The blood alcohol curve varies because of different absorption rates among individuals.  There are also variances in absorption rates caused by the meal consumed with the alcohol, the type of alcohol, etc.  Source: Giannelli, et al., supra

A common chemical test used to produce a BAC score—the breath test (or “breathalyzer”)—is a flawed test for many reasons, although the D.C. Code deems BAC scores to be direct evidence of intoxication.  D.C. Code § 50-2206.01(9)(A)(i) even defines a BAC score of 0.08 or above as the legal definition of “intoxication.”  For brevity’s sake, let’s examine just one flaw in that assumption—the fact that breath testing equipment assumes a ratio of 2100:1 of alcohol in exhaled breath to alcohol in the blood.  The 2100:1 ratio is used in the United States, while in different nations, different ratios are used; the reason is that the ratio is an average for a national population (!).  Problems of scientific accuracy are inherent because the actual ratio of alcohol in exhaled breath to alcohol in the blood varies physiologically from person to person, and even in an individual during a one-hour time period.  Source:  Edward L. Fiandach, DWI: A Critical Examination of Breath Testing Assumptions and Techniques, 30 The Champion 47, 48 (Nov. 2006). 

Breath testing equipment uses many similar assumptions that render the results irrelevant for a given DUI/OWI defendant.  Despite the questionable assumptions used by breath testing equipment manufacturers to produce BAC scores, a BAC score of 0.08 is legally defined as intoxication in the District of Columbia, under D.C. Code § 50-2206.01(9)(A)(i).  For individuals under age 21, any measurable amount of alcohol is legally defined as intoxication, under D.C. Code § 50-2206.01(9)(A)(ii). Depending on your circumstances, you may have options.  Your attorney can discuss them with you.

DISCLAIMER:  The opinions expressed herein are general and might not be applicable to your case or circumstances.  If you need legal advice about a specific matter, consult an attorney in your jurisdiction.