Single-Column Gas Chromatography and a Bungled HGN Test: How State v. Rudloff Got It Wrong, and What Michigan OWI Defenders Should Do About It
A Nebraska appellate court recently handed down a decision that should alarm anyone who values forensic accuracy in drunk driving prosecutions. In State v. Rudloff, No. A-25-313 (Neb. Ct. App. Apr. 28, 2026), the court upheld a DUI conviction resting on two deeply compromised pillars: a blood alcohol test performed using a single-column gas chromatography method that the testing laboratory itself had already abandoned as inferior, and a Horizontal Gaze Nystagmus test administered so improperly that it would not pass scrutiny in any serious forensic setting. The court called both admissible. Neither deserved that label.
The Blood Test: Methodology the Lab Already Threw in the Trash
To understand why the blood test in Rudloff is scientifically indefensible, you need to understand what single-column gas chromatography actually is and what it is not.
Gas chromatography flame ionization detection, known as GC-FID, is a legitimate analytical chemistry technique with a long history in forensic toxicology. The operative word is history. The single-column variant of that technique, the version used to analyze Rudloff's blood, separates compounds using a single stationary phase. The fundamental problem is that a single column cannot definitively distinguish between ethanol and other volatile compounds that may be present in a blood sample and that elute at or near the same retention time. Isopropanol, acetone, and other endogenous or exogenous volatiles can co-elute with ethanol on a single column and artificially inflate the reported alcohol concentration. The analyst has no way to know whether the peak she is measuring represents pure ethanol or a mixture of compounds masquerading as ethanol.
This is precisely why the analytical chemistry and forensic toxicology communities moved toward dual-column gas chromatography. Two columns with chemically distinct stationary phases provide independent confirmation that the compound being measured is actually ethanol and not a chemical impostor. A result that appears on both columns at the expected retention time, in consistent concentration, can be trusted. A result from a single column standing alone cannot be verified in the same way.
To put this in plain terms: single-column GC-FID for blood alcohol analysis is roughly analogous to using an environmental testing protocol designed for groundwater screening to make a forensic determination of human blood alcohol content. The technique may detect something, but without the confirmatory second column, the analyst cannot be certain what she has detected or whether the reported number accurately represents the ethanol concentration in the sample. This is not a minor procedural quibble. It strikes at the heart of what forensic science is supposed to provide: a reliable, independently verifiable result.
The Nebraska Public Health Environmental Laboratory apparently understood this limitation clearly enough that it replaced single-column analysis with dual-column analysis in January 2022, just five months after Rudloff's blood was tested. That institutional decision is a tacit acknowledgment that the single-column method was inadequate. A laboratory does not upgrade its methodology at significant cost and operational disruption unless it has concluded that the old method was producing results it could no longer stand behind with confidence.
The defense expert, Kevin Schug, made exactly this point. He testified that the test results were not scientifically reliable based on how the analysis was conducted and how the instrument was calibrated and validated prior to testing. The court dismissed his opinion as going only to the application of the methodology rather than to the validity of the methodology itself. That distinction, while doctrinally tidy, is scientifically hollow in this context. When a single-column instrument is improperly calibrated or validated, the analyst has no independent check to catch the error. The absence of a second column is not an application problem. It is a design flaw in the methodology that makes application-level errors undetectable.
The court relied heavily on the State's expert, Dr. Henry Nipper, who testified that single-column chromatography is still a "highly reliable technique." That testimony deserves scrutiny. Dr. Nipper simultaneously acknowledged that dual-column chromatography "removed some doubt" as to the accuracy of single-column results. If there was doubt to be removed, there was doubt present. Telling a jury that a method is highly reliable while conceding in the same breath that a superior method was developed specifically to address that method's reliability concerns is not a ringing endorsement. It is a measured retreat dressed up as confidence.
The Daubert Framework and Why the Court Missed the Point
The Nebraska court applied the Daubert/Schafersman gatekeeping framework, which requires a trial court to act as a gatekeeper to ensure that expert testimony rests on a valid methodology that can reliably be applied to the facts at hand. The court found that the gatekeeping function had been performed and that no abuse of discretion occurred. That conclusion is defensible as a matter of appellate deference. It is not defensible as a matter of forensic science.
The Daubert inquiry is supposed to consider factors including whether the methodology has a known or potential error rate, whether it has been subjected to scientific testing and replication, and the degree to which it is generally accepted within the relevant expert community. A methodology that the testing laboratory itself retired five months after it was used on the defendant's sample fails the error rate and general acceptance inquiries in the most direct way imaginable. The lab's own institutional judgment, expressed through its operational decision to abandon the method, was more candid than its expert witnesses were permitted to be on the stand.
Michigan practitioners should note that the same gatekeeping analysis applies under MRE 702, which Michigan amended effective January 1, 2004, to expressly incorporate the Daubert reliability standard. MCL 600.2955 further codifies the Daubert factors for cases involving death or personal injury and requires courts to consider the known or potential error rate and whether the methodology is generally accepted within the relevant expert community. A Michigan court presented with a blood test result generated by a single-column method that the testing laboratory has since discarded should be asked, directly and persistently, to apply those statutory factors to the laboratory's own institutional verdict on its former technique.
The HGN Test: A Textbook Example of How Not to Do It
The blood test is not the only thing wrong with Rudloff. The court also upheld the admission of the Horizontal Gaze Nystagmus test results, and that ruling is, if anything, harder to defend than the blood test ruling.
The HGN test is a standardized psychophysical assessment with a very specific administration protocol. The National Highway Traffic Safety Administration manual prescribes the number of passes required for each component of the test because the protocol is what produces the validated correlation between observed clues and blood alcohol concentration. The test was validated empirically, using the prescribed procedure. When the procedure is not followed, the validated correlation no longer applies in the same way, and the officer's observations cannot be reliably mapped to the NHTSA's correlation tables.
Deputy Murphy testified that she did not perform all of the required medical passes. She performed four passes on one eye and three on the other, rather than the equal and complete number the protocol demands. She acknowledged this deviation on the stand.
This is not a minor variation from ideal field conditions of the type the NHTSA manual contemplates when it notes that slight deviations may affect evidentiary weight. The NHTSA manual's acknowledgment that tests administered under less than ideal conditions will "generally serve as valid and useful indicators of impairment" refers to environmental factors, such as uneven pavement, poor lighting, or traffic noise, not to the officer's failure to complete the required number of standardized stimulus passes. The passes exist to control for physiological variables and to ensure that nystagmus is evaluated under consistent conditions. Cutting them short is not an adaptation to difficult field conditions. It is a failure to administer the test.
The NHTSA manual states: "The procedures outlined in this manual describe how the [field sobriety tests] are to be administered under ideal conditions. We recognize that the [tests] will not always be administered under ideal conditions in the field, because such conditions will not always exist. Even when administered under less than ideal conditions, they will generally serve as valid and useful indicators of impairment. Slight variations from the ideal . . . may have some affect [sic] on the evidentiary weight given to the results. However, this does not necessarily make the [tests] invalid."
The court read that language as covering what happened in Rudloff. That reading is a stretch. The passage addresses the conditions under which the test is administered, not the completeness of the administration itself. A test performed in a rainstorm on a gravel shoulder, but executed fully and correctly, falls within the manual's contemplation. A test in which the officer simply skips passes does not.
The court also leaned on the standard from State v. Prescott, 280 Neb. 96 (2010), that an officer may testify to HGN results if she has been adequately trained and conducted the testing in accordance with that training. The difficulty with that standard, applied to these facts, is that conducting a test in accordance with one's training presupposes that the training was followed. If the training required a certain number of passes and the officer performed fewer, the test was not conducted in accordance with the training. The court papered over this logical gap by accepting the deputy's testimony that she performed the test "the way that she had been trained and always practiced," without examining whether that practice was itself compliant with the NHTSA protocol her training was based on.
What Michigan Defense Attorneys Should Take From This Case
The Rudloff decision is instructive not because it got the science right, but because it shows how courts can resolve genuinely serious forensic challenges against defendants when defense counsel does not press those challenges with sufficient precision and depth.
On the blood test issue, the lesson is that a Daubert challenge to a superseded methodology must go beyond arguing that a better method now exists. Defense counsel must affirmatively demonstrate that the retired method carried a known, quantifiable error risk that the new method was designed to eliminate, and that this error risk was present in the specific analysis at issue. That argument requires documentary discovery of the laboratory's standard operating procedures, quality control records, proficiency test results, and internal communications regarding the transition to the new method. It also requires an expert who is prepared to explain, in concrete analytical chemistry terms, how the absence of a confirmatory second column creates an undetectable risk of false elevation in the reported blood alcohol concentration.
On the HGN issue, the lesson is that deviations from the NHTSA protocol must be characterized precisely. Defense counsel should distinguish clearly between environmental deviations, which the manual acknowledges as inevitable, and procedural omissions, which go to whether the test was administered at all in its validated form. The argument is not that the test was performed under difficult conditions. The argument is that an incomplete test is not the test. A result obtained from an incomplete procedure cannot be mapped to a validated correlation table designed for a complete procedure, and admitting that result as proof of impairment asks the jury to rely on science that was not actually performed.
Michigan courts applying MRE 702 and MCL 600.2955 are obligated to function as genuine gatekeepers, not rubber stamps. When a laboratory has abandoned a testing method, when an officer has skipped required steps in a standardized protocol, and when the State's own experts acknowledge residual doubt about the accuracy of the results, the gatekeeper should be closing the door, not waving the evidence through. Rudloff is a case study in what happens when that door is left open.
