1
STATE OF MINNESOTA
IN SUPREME COURT
A24-1440
Court of Appeals Derek Alexander Knapp, Appellant,
vs.
Commissioner of Public Safety, Respondent.
Thissen, J.
Took no part, Aligada, Jr., J. Filed: November 12, 2026 Office of Appellate Courts ________________________ Robert M. Christensen, Robert M. Christensen, P.L.C., Minneapolis, Minnesota, for appellant.
Keith Ellison, Attorney General, Nicholas Moen, Assistant Attorney General, Saint Paul, Minnesota, for respondent.
William A. Lemons, Minnesota County Attorneys Association, Saint Paul, Minnesota, for amicus curiae Minnesota County Attorneys Association.
Charles A. Ramsay, Ramsay Law Firm, P.L.L.C., Roseville, Minnesota, for amicus curiae National College for DUI Defense.
________________________
S Y L L A B U S
In assessing the admissibility of a breath-test result challenged on the ground that an inadequate observation period rendered the test result unreliable, an imperfect observation period is not a per se reason to exclude the breath-test result. Instead, the
2
district court must consider all record evidence bearing on whether the breath sample was contaminated and must admit the test result if a preponderance of the evidence shows that the breath sample was not contaminated.
Vacated and remanded.
O P I N I O N
THISSEN, Justice. In this case, we are asked to determine how district courts should assess the admissibility of a breath-test result in an implied-consent hearing when the driver challenges the sufficiency of the pre-test observation period. Because breath-testing instruments like the one used in this case are designed to measure "deep lung air,"
residual alcohol in the mouth presents a contamination risk. A pre-test observation period is intended to prevent (or detect) mouth alcohol contamination. Here, respondent Commissioner of Public Safety (the Commissioner) revoked appellant Derek Alexander Knapp's driver's license after Knapp was arrested on suspicion of driving while impaired (DWI) and consented to a breath test that showed his alcohol concentration exceeded the legal limit. Knapp challenged the revocation on the sole ground that the arresting officer conducted an inadequate pre-test observation period. The district court determined that the observation period was inadequate and that Knapp's test result was inadmissible. The court of appeals reversed, holding that the district court abused its discretion in ruling the test result was inadmissible based solely on evidence of an imperfect observation period.
3
We now clarify the proper standard: An imperfect observation period is not a per se reason to exclude a breath-test result. Rather, the risk that a breath sample was contaminated could support a district court determination that a breath-test result taken from that sample is inadmissible. The district court must consider all record evidence bearing on whether a breath sample was contaminated and admit the test result if a preponderance of the evidence shows that the breath sample was not contaminated. Because neither the court of appeals nor the district court applied this standard, we vacate and remand to the district court to apply this standard here.
FACTS
Under Minnesota Statutes § 169A.20, subdivision 1(5), "[i]t is a crime for any person to drive, operate, or be in physical control of any motor vehicle" when "the person's alcohol concentration at the time, or as measured within two hours of the time, of driving, operating, or being in physical control of the motor vehicle is 0.08 or more."
Alcohol concentration is defined as the number of grams of alcohol per 100 milliliters of blood, 210 liters of breath, or 67 milliliters of urine. Minn. Stat. § 169A.03, subd. 2. This case concerns the administration of a test to determine the alcohol concentration in a breath sample.
Breath Test and Pre-Test Observation Period
Breath-testing instruments approved by the Commissioner, like the one used in this case—the DataMaster DMT-G (DMT)—are designed to measure the alcohol concentration in exhaled "deep lung air." Joseph Fetzik et al., Minn. Dep't of Pub. Safety,
4
Bureau of Crim. Apprehension, DMT Operator Training Manual 14-15 (v.4.2 2025).1 Deep lung air provides a basis to determine alcohol concentration because ingested alcohol is absorbed into the bloodstream, circulated through the body, and excreted into the air in deep lung tissue. Id. at 91. When a person who has consumed alcohol exhales into a breath-testing instrument, the instrument measures the alcohol concentration of the exhalation, or breath sample. See id. To accurately measure a person's alcohol concentration, it is therefore essential that the alcohol in the breath sample be from deep lung air. Residual alcohol in the mouth, known as mouth alcohol, presents a contamination risk because this alcohol was not absorbed into the bloodstream but is nevertheless present in exhaled breath. Id. at 14-15, 48, 102. The presence of mouth alcohol can result in inaccurate, elevated alcohol-concentration measurements. Id. Mouth alcohol can result from recent drinking or liquid alcohol from the stomach introduced into the mouth by vomiting, belching, or regurgitating. Id. A pre-test observation period is one step in the breath-testing procedure intended to prevent or detect mouth alcohol contamination. Id. at 48, 58.
5
Implied-Consent Hearing
The following undisputed facts are based on the evidence presented at Knapp's implied-consent hearing. On February 10, 2023, a Washington County sheriff's deputy arrested Knapp on suspicion of DWI and transported him to the Washington County jail to conduct an evidentiary breath test on a DMT. When the deputy and Knapp arrived at the jail at midnight, a corrections officer asked the deputy to wait with Knapp outside the intake area while another arrestee was processed. The deputy and Knapp waited in the deputy's squad car. Knapp sat in the back seat and remained handcuffed. The deputy sat in the front seat, which was separated from the back seat by a clear plexiglass divider with an opening in the middle. After asking Knapp to open his mouth and lift his tongue, the deputy gave Knapp the breath-test advisory. When the advisory concluded two minutes later—about 12:02 a.m.—the deputy escorted Knapp to the jail's intake area. Inside the intake area, a corrections officer searched Knapp while the deputy spoke to the DMT operator who would administer Knapp's breath test. The deputy told the DMT operator that he began the 15-minute observation period at 12:00 a.m., and requested that the DMT be administered at 12:15 a.m. The deputy spoke to the DMT operator for another minute and then walked to a computer terminal to begin paperwork. Footage from the deputy's body-worn camera shows that the deputy and Knapp remained in the same room and conversed intermittently, but the distance between the two is unclear. At around 12:06 a.m., the corrections officer finished searching Knapp and placed him in a chair next to the deputy. For the next three to four minutes, the deputy's body-worn camera video shows that Knapp remained seated next to the deputy, and the
6
deputy and Knapp conversed while the deputy entered information into a computer. For reasons that are not clear in the record, the video from the deputy's camera ended at 12:10 a.m.
Beginning at 12:15 a.m., the DMT operator ran a series of successful diagnostic checks on the DMT, which included four air-blank tests and a control sample. At 12:17 a.m., more than 15 minutes after Knapp entered the intake area, Knapp provided the first of two breath samples. The DMT reported that Knapp's alcohol concentration was .08.
Based on the test result, the Commissioner revoked Knapp's driver's license. See Minn. Stat. §§ 169A.20, subd. 1(5); 169A.52, subds. 2(1), 4. Knapp petitioned for review of the revocation, and the district court held an implied-consent hearing. See Minn. Stat. § 169A.53, subd. 2. At the implied-consent hearing, Knapp clarified that he was challenging the admissibility of the DMT test results and that his argument focused on a lack of an adequate observation period.
The deputy who arrested Knapp testified at the hearing. His testimony included a description of the training he received on conducting an observation period. The deputy stated that he was trained that an officer must conduct a 15-minute observation period prior to administering a DMT test. The officer also testified that his training taught him that he must check the test subject's mouth to ensure it does not contain any vomit or obstructions, and observe the test subject for "burping, belching and vomiting." The deputy was trained not to become distracted and to watch or remain in proximity to the driver for the duration of the observation period.
7
The deputy also testified about the observation period he conducted before Knapp's test. He testified that he began the observation period in his squad car, and that he could hear Knapp effectively and see him through the rearview mirror. The deputy stated that he did not observe Knapp burping, belching, or vomiting during the observation period.
The DMT operator testified that she was a trained and certified DMT operator at the time she administered Knapp's test. She stated that during Knapp's test the DMT successfully completed the required air blank tests and control sample test, and that the DMT accepted both of Knapp's breath samples. The DMT operator confirmed that the DMT reported Knapp's alcohol concentration as .08. She testified that she did not conduct Knapp's observation period, but that Knapp did not burp, belch, or vomit during the test.
Knapp did not offer any evidence. When the Commissioner moved to submit the DMT test result into evidence, however, Knapp's attorney suggested that the observation period had been inadequate and "object[ed] to the [c]ourt making a finding that [the test] is a valid .08 test." The district court noted the objection and stated that it would "receive [the test result] into evidence." The district court did not immediately rule on the Commissioner's revocation decision. Instead, the court instructed the parties to submit arguments in writing and concluded the hearing.
Following the hearing, the parties submitted written arguments to the district court. Knapp argued that the deputy's observation was inadequate and that an inadequate observation period alone renders a breath-test result unreliable and inadmissible. The
8
Commissioner, in contrast, argued that a breath test is admissible as long as the operator was certified and the test instrument was functioning normally. The Commissioner urged that any alleged defects or deviations in the testing procedure go to the evidentiary weight of a test result, not its admissibility.
The district court determined that the Commissioner failed to demonstrate that the deputy conducted an adequate observation period and, consequently, the test result was unreliable. In reaching its decision, the court found that when the deputy and Knapp were in the squad car waiting to enter the jail, the deputy was "facing away from [Knapp]" and
"had to rely on his rear-view mirror to see [Knapp]." The court also found the deputy's
"perspective was further limited by the Plexiglass barrier" between the squad car's front and back seats. The court also noted that after the deputy and Knapp entered the jail, the deputy's "location [was] unclear relative to [Knapp]." Accordingly, the district court rescinded Knapp's license revocation.2
9
Appeal
The Commissioner appealed, and the court of appeals reversed. Knapp v. Comm'r of Pub. Safety, 23 N.W.3d 847, 857 (Minn. App. 2025). It concluded that the district court abused its discretion by "determining that the DMT test-results evidence was unreliable and by rescinding the revocation of Knapp's driving privileges based solely on evidence of an imperfect pretest observation period, without evidence that Knapp had ingested something or otherwise experienced a bodily function during the observation period that affected the test results." Id. at 856.
In reaching its decision, the court of appeals applied a three-part burden-shifting test developed in a series of court of appeals cases. Id. at 853-56. Under that test, " '[t]he commissioner must make a prima facie case that the test is reliable and that its administration in the particular instance conformed to the procedure necessary to ensure reliability.' " Id. at 853 (quoting Vondrachek v. Comm'r of Pub. Safety, 906 N.W.2d 262, 272 (Minn. App. 2017)). If the Commissioner meets this burden, the driver " 'must produce evidence suggesting why the test was untrustworthy.' " Id. at 855 (quoting Falaas v. Comm'r of Pub. Safety, 388 N.W.2d 40, 42 (Minn. App. 1986)). At this step,
" 'a deficiency in observation of the subject does not of itself invalidate the test,' " rather of a chemical or scientific test must establish that the test itself is reliable and that its administration in the particular instance conformed to the procedure necessary to ensure reliability. Without a foundation guaranteeing the test's reliability, the test result is not probative as a measurement and hence is irrelevant." (citations omitted)). Because the district court rescinded Knapp's license revocation on the basis that the test result lacked a foundation of reliability, we review its order as an admissibility determination.
10
it " 'merely gives the driver an opportunity to suggest why such a failure makes the test results unreliable.' " Id. (quoting Melin v. Comm'r of Pub. Safety, 384 N.W.2d 474, 476 (Minn. App. 1986)). Furthermore, " '[t]he argument that something may have occurred during observation to affect the test result is speculation and should not be used without supporting evidence as the basis for rescinding a revocation.' " Id. at 856 (quoting Falaas, 388 N.W.2d at 42). Instead, under court of appeals precedent, the driver "must prove that he ingested or regurgitated something that affected the test results." Id.; see also Hounsell v. Comm'r of Pub. Safety, 401 N.W.2d 94, 96 (Minn. App. 1987) ("Even if a driver can prove that he had something in his mouth, he must also demonstrate that the substance actually affected his test results."). If the driver meets this burden, the burden shifts back to the Commissioner to demonstrate that the defect the driver alleges did not affect the test's reliability. See Noren v. Comm'r of Pub. Safety, 363 N.W.2d 315, 318 (Minn. App. 1985).
We granted review.
ANALYSIS
We are asked to decide whether the results of Knapp's breath test are admissible and whether the fact-finder may consider them in his driver's license revocation proceeding. Knapp argues that the breath-test results are inadmissible because the Commissioner failed to show that the test was conducted according to the required procedures. Knapp's argument focuses on the adequacy of the observation period. He does not challenge the DMT's reliability, nor does he claim that the DMT operator operated the DMT improperly. Determining whether Knapp's breath-test result is
11
admissible requires us to clarify the standard that district courts should apply in assessing the admissibility of breath-test results.
We review a district court's evidentiary rulings for an abuse of discretion. State v. Carbo, 6 N.W.3d 114, 123 (Minn. 2024). A district court abuses its discretion when it bases its ruling "on an erroneous view of the law" or makes a ruling "against logic and the facts in the record." Riley v. State, 792 N.W.2d 831, 833 (Minn. 2011). We review a district court's factual findings for clear error. State v. Carridine, 812 N.W.2d 130, 141 (Minn. 2012). In this case, neither the Commissioner nor Knapp challenges the district court's factual findings. The Commissioner's arguments—while not conceding that the observation period was defective—focus on the legal standard for admissibility of the test results and not the facts surrounding the test itself.3
Like all evidence, breath-test results must satisfy the basic requirements of the rules of evidence. First, the test result must be relevant. Minn. R. Evid. 402. " 'Relevant evidence' means evidence having any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it
12
would be without the evidence." Minn. R. Evid. 401. Relevant evidence "may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence." Minn. R. Evid. 403.
Additionally, the party offering a chemical or scientific test result, such as a breath-test result, must establish that (1) "the test itself is reliable" and (2) "its administration in the particular instance conformed to the procedure necessary to ensure reliability." State v. Dille, 258 N.W.2d 565, 567 (Minn. 1977); see also Doe v. Archdiocese of St. Paul, 817 N.W.2d 150, 165 (Minn. 2012) (explaining that
"foundational reliability requires the proponent of a … test [to] establish that the test itself is reliable and that its administration in the particular instance conformed to the procedure necessary to ensure reliability" (alteration in original) (citation omitted) (internal quotation marks omitted)).
In the context of a breath test, the first foundational reliability question—is the test itself reliable?—focuses on whether a breath-test instrument, when properly operated, reliably and accurately analyzes a breath sample and reports an alcohol-concentration measure based on its contents. The second foundational reliability question—whether the test's administration in the particular instance conformed to the procedure necessary to ensure reliability—focuses on whether the breath test at issue was conducted according to the procedure designed to provide a test result that reliably reflects the alcohol concentration in the driver's deep lung breath. Knapp's challenge to the breath test in this
13
case—that his breath sample was potentially contaminated—concerns this second question.
A.
We begin our analysis by clarifying how a district court should approach this second foundational reliability question as to breath test procedure and reliability, and start by reviewing our precedent on the admissibility of breath-test results. For background, we review our decision in State, City of St. Louis Park v. Quinn, 182 N.W.2d 843 (Minn. 1971), in which we considered a challenge to the reliability of breath-test instruments. We then turn to our decisions in State v. Dille, 258 N.W.2d 565, and State, Department of Public Safety v. Habisch, 313 N.W.2d 13 (Minn. 1981), which provide the framework for our analysis. Next, we review In re Source Code Evidentiary Hearings in Implied Consent Matters (Source Code), 816 N.W.2d 525 (Minn. 2012), where we established the evidentiary standard a district court must apply in determining whether a particular test result is reliable.
We also address the parties' arguments regarding Minnesota Statutes section 634.16—a statute that, if certain conditions are met, eliminates the need for expert testimony to establish that a breath-testing instrument provides a trustworthy and reliable measure of alcohol concentration—and our decision in State v. Loving, 775 N.W.2d 872 (Minn. 2009).
Finally, based on this analysis, we clarify how a district court should consider the risk of a contaminated breath-test sample when assessing a test result's admissibility.
14
1.
We begin, for background and context, with our decision in Quinn, which concerned the reliability of breath-test instruments. In Quinn, the defendant was convicted of driving under the influence after a breath test showed that his alcohol concentration exceeded the legal limit. 182 N.W.2d at 844. On appeal, Quinn argued that the district court erred in admitting the test result because, at trial, the State did not offer expert testimony on "the theory and mathematical formula behind the machine and the formula that translates volume of alcohol by weight in the breath to volume of alcohol by weight in the blood."4Id. at 845. According to Quinn, without such expert testimony his test result lacked foundation and was inadmissible. Id. The question before us was whether, in every case, the State needed an expert to testify about how the breath-test instrument assessed the concentration of breath alcohol and converted that concentration into a blood-alcohol-content concentration.
We rejected Quinn's argument and held that expert testimony was unnecessary to establish how the breath-test instrument measured breath alcohol and converted that
15
measurement to blood alcohol because that process has been accepted as reliable. Id. Thus, we concluded, there was no basis to refuse to admit an instrument's test result for lack of such expert testimony as long as the State established that the test operator was
"properly certified" and that the instrument "was in proper working order and the chemicals were in proper condition." Id.5We observed that "[i]t is generally held that the alcoholic content of the blood may be reliably determined by [a breath] test, and testimony of the reading obtained upon a properly conducted test may be admitted without antecedent expert testimony that the reading is a trustworthy index of alcohol in the blood." Id. (citing 7 Am. Jur. 2d Automobiles and Highway Traffic § 332; State v. Miller, 165 A.2d 829 (N.J. Super. Ct. App. Div. 1960); State v. Johnson, 199 A.2d 809 (N.J. 1964)). In other words, Quinn focused on what we now consider the first requirement for establishing foundational reliability—whether the test itself is reliable. The Commissioner incorrectly reads Quinn to stand for the much broader principle that a breath-test result is admissible as long as it was obtained by a certified operator using a properly functioning test instrument, regardless of whether the test was properly administered. The Commissioner argues that alleged defects in administering a test go to the weight of the test result, not its admissibility. The Commissioner's position ignores the fact that a properly functioning breath-test instrument will not produce a reliable analysis of a person's alcohol concentration if the breath sample used to perform that
16
analysis is contaminated. Quinn does not address the distinct question of whether a defendant may challenge a test result on the grounds that the sample tested was contaminated.
2.
We considered the sample contamination question in Dille, a case addressing the results of a blood test for alcohol concentration. 258 N.W.2d at 566. Among other things, Dille argued that his particular test result lacked foundational reliability because his blood sample may have been contaminated. Id. at 567-68. Dille pointed out that the State did not present direct evidence that the needle and bottle used to collect his blood sample were sterilized and that trial testimony indicated his sample bottle contained a white substance that the State failed to identify. Id.
We started our analysis by observing that "[t]he proponent of a chemical or scientific test must establish that the test itself is reliable and that its administration in the particular instance conformed to the procedure necessary to ensure reliability."6Id. at 567 (emphasis added); see also id. at 568 n.2 ("The burden of persuading the jury that the test results were accurate remains with the prosecution."). Under that standard, we agreed with Dille that the State bears "the burden to prove that contaminants have not adversely affected the reliability of the test." Id. at 568.
We concluded, however, that the State met its burden in Dille's case because the record contained sufficient evidence that Dille's sample was not contaminated. Id. We
17
noted that, although the State did not present direct evidence that the needle and bottle were sterilized, the record showed that the sample was taken in a hospital, by a medical technologist, that the technologist testified that the needle had never been used before, and that the sample bottle was supplied by the Bureau of Criminal Apprehension (the BCA). Id. We also noted that the Washington Supreme Court confronted a similar challenge in State v. Erdman, 391 P.2d 518 (Wash. 1964). In Erdman, it was undisputed that the appellant's sample bottle contained a white substance. The prosecution in that case did not establish the substance's identity with direct evidence, but the chief toxicologist testified that he did not detect any inaccuracies when testing the defendant's sample and that it was general practice to place sodium fluoride—a white substance— into bottles prepared for blood samples. Id. at 519-20. The Erdman court concluded that this testimony "constituted prima facie proof as to the identity of the powder and the consequent accuracy of the blood-alcohol test." Id. at 519. We likened the testimony in Erdman to the testimony in Dille's case that the BCA provided the sample bottle and its contents as part of a blood-testing kit. Dille, 258 N.W.2d at 568. We concluded that this record evidence constituted sufficient "indicia of reliability" (what we referred to as
"prima facie proof"7) that the sample tested was Dille's uncontaminated blood—
7 Our use of the term "prima facie proof" in Dille has caused confusion about the test that courts should apply in assessing the admissibility of breath-test results. Drawing on that language, the court of appeals has developed a three-part burden-shifting test for admitting alcohol-concentration-test results. See Kramer v. Comm'r of Pub. Safety,
18
notwithstanding the lack of direct evidence for each step in the testing procedure, including sterilization. Id. In other words, the State's failure to present evidence of every protocol necessary for an alcohol-concentration-test result to be reliable, including steps to ensure the uncontaminated nature of a tested sample, does not mean that a test result burden-shifting test applied and focusing our analysis of the admissibility of breath-test results on the ultimate question of whether a test result is reliable and relevant). We consider the burden-shifting test to be unnecessarily complicated and have concerns that it may improperly shift the burden to prove inadmissibility to the driver. In particular, we are concerned with recent court of appeals cases suggesting, as in this case, that an implied-consent petitioner or DWI defendant bears the burden of proof to show that a sample was contaminated. See Knapp, 23 N.W.3d at 856 (concluding Knapp's breath-test result was admissible because Knapp "presented no evidence to suggest that something occurred during the observation period that affected the test results"); Junker v. Comm'r of Pub. Safety, 925 N.W.2d 661, 663 (Minn. App. 2019) ("To impeach the credibility of breath-test results, a driver must prove that he burped during the observation period and must also demonstrate that the burping actually affected the test results."); Vondrachek, 906 N.W.2d at 273 (stating that an implied-consent petitioner
"has the burden of presenting evidence that he ingested or regurgitated a substance that would affect the results," and holding that appellant failed to meet that burden because he
"presented no evidence of any alcohol or other substance in his mouth that would have affected the results"). As discussed more fully below, that suggestion is contrary to our holding in Dille, where we said that "the prosecution has the burden to prove that contaminants have not adversely affected the reliability of the test." 258 N.W.2d at 568 (emphasis added). For this reason, we refer to the Commissioner's burden as establishing that the breath sample was not contaminated.
We further clarify that we did not endorse a burden-shifting test in Dille when we said that, upon the Commissioner establishing "prima facie proof" of a test's reliability,
" 'some burden must be placed upon the defendant to come forward with some proof, evidence or indication that impugns the accuracy of the test.' " Id. at 568 (quoting Erdman, 391 P.2d at 519-20). Rather, our statement reflects that a district court's decision on whether a breath test's administration conformed to the procedures necessary to ensure the test result's reliability requires evaluating all the circumstances bearing on that question.
Accordingly, we decline to adopt the burden-shifting framework the court of appeals applied, and, to the extent that prior court of appeals opinions are inconsistent with our decision today, those opinions are overruled. Going forward, courts should apply the more straightforward test we set forth in this opinion to assess the admissibility of alcohol-concentration-test results.
19
must be suppressed. Such a test result may still be admitted so long as there is sufficient evidence in the record that the test result was based on an uncontaminated sample.8 We further observe that the question in Dille was not whether the State could prove that each required step in the alcohol-concentration-test procedure was performed adequately. Rather, the question was whether there was sufficient evidence to conclude, in light of the process used to gather the sample, that the sample that was tested accurately represented the driver's blood. Consistent with Dille, Knapp himself acknowledges before us that the breath-testing process—including the observation period—need not be perfect to be reliable and that "[m]inor or de minimis lapses [during an observation period] do not defeat admissibility."
In Habisch, we applied reasoning similar to that we applied in Dille. Habisch involved a breath-testing procedure that called for the test operator to measure the alcohol concentration of a sample of the test subject's breath and then validate the test's accuracy by measuring samples with a known or expected alcohol concentration. 313 N.W.2d at
14. The validation samples in Habisch's case included a sample of the air in the room, which should contain no alcohol, and a sample of "simulator solution," which, in Habisch's case, contained an alcohol concentration of .11.9Id. at 14-15.
20
At Habisch's trial, the test operator testified that the room air and simulator solution tests gave the expected results. Id. at 15. Another officer testified, however, that the standard procedure called for replacing the simulator solution every 30 days because the solution can lose its properties over time and cease to be a valid indicator of the instrument's accuracy. Id. He testified that the simulator solution used in Habisch's test had not been replaced in 62 days. Id.
Habisch argued that failing to replace the simulator solution on schedule rendered his test result unreliable. Id. at 14. We disagreed, concluding, as we did in Dille, that despite the lack of evidence that a particular procedure was followed, there was sufficient other evidence in the record supporting the test's reliability. Id. at 16. First, we pointed out that the record included the test instrument's logbook, which showed that the simulator solution properly read .11 the day it was received, and it consistently read .11 during and after Habisch's test. Id. at 15-16. These readings demonstrated that, despite its age, the simulator solution had not lost its potency. Second, we noted the two separate validation tests performed—the room air and simulator solution tests—and concluded that two separate validation tests providing expected results " 'would seem to be almost incontrovertible proof not only that the chemicals are proper but that the instrument is in working order.' " Id. at 16 (quoting L. Poindexter Watts, Some Observations on Police- Administered Tests for Intoxication, 45 N.C. L. Rev. 34, 87 (1966)).
(2) first subject sample, (3) Air Blank, (4) Control Sample, (5) Air Blank, (6) second subject sample, (7) Air Blank. DMT Operator Training Manual 50-51.
21
Thus, in Habisch, as in Dille, the petitioner challenged his test results based on alleged flaws in a particular step in the testing process. In neither case did we conclude that any flaw in administering a breath test precludes establishing its foundational reliability or renders the test result per se inadmissible. Rather, in both cases we applied the standard that a test result is admissible if there is sufficient record evidence showing that the reliability purposes that a protocol serves—i.e., its role in preventing test results that are not an accurate measure of the alcohol concentration in the driver's blood or breath—was met despite any error in administration. And in both cases, we found that standard had been met.
3.
We next review our decision in Source Code, 816 N.W.2d 525, a case in which we established the evidentiary standard a district court must apply in determining whether there is sufficient evidence—what we termed "indicia of reliability" in Dille—to show that a test result is reliable. In Source Code, the appellants10challenged the reliability of test results obtained from Intoxilyzer 5000EN breath-testing instruments, arguing, among other things, that defects in the Intoxilyzer's computer source code11caused the
22
instrument to report inaccurate values for measured breath alcohol. Id. at 530-31. The district court held an evidentiary hearing and determined that the test results were reliable and unaffected by actual or alleged problems with the instrument's source code. Id. at 537; see In re Source Code Evidentiary Hearings in Implied Consent Matters, Nos. 70- CV-09-19459; 70-CR-09-19749, Order (Dakota Cnty. Dist. Ct. filed Mar. 8, 2011). On appeal, appellants argued that the district court committed reversible error because it did not articulate the evidentiary standard it applied in determining that the numerical values the instrument reported were reliable and unaffected by actual or alleged problems with the source code. Source Code, 816 N.W.2d at 537.
Before us, the Source Code parties agreed that whether the Intoxilyzer was reliable and unaffected by actual or alleged defects in its source code was a preliminary question of admissibility. Id. We observed that, under Minn. R. Evid. 104(a), "[p]reliminary questions concerning … the admissibility of evidence shall be determined by the court,"
but that we had not previously articulated the evidentiary standard of proof required to establish a preliminary condition of admissibility under the rule. Source Code,
816 N.W.2d at 537-38. Persuaded by the Supreme Court's reasoning in Bourjaily v. United States, 483 U.S. 171, 175 (1987), we held that district courts must use a as the ["]master processor["]). When the code is compiled or assembled and linked, it is converted into a form that is executable by the microprocessors. The microprocessors execute the instructions contained in the converted source code to perform the instrument functions.
816 N.W.2d at 527 n.1.
23
preponderance of the evidence standard when determining whether a preliminary condition of admissibility has been fulfilled.12Source Code, 816 N.W.2d at 539.
4.
Next, we examine Minnesota Statutes section 634.16, a statute enacted in 1984, that codifies, in part, our holding in Quinn. Section 634.16 provides:
In any civil or criminal hearing or trial, the results of a breath test, when performed by a person who has been fully trained in the use of an infrared or other approved breath-testing instrument, as defined in section 169A.03, subdivision 11, pursuant to training given or approved by the commissioner of public safety or the commissioner's acting agent, are admissible in evidence without antecedent expert testimony that an infrared or other approved breath-testing instrument provides a trustworthy and reliable measure of the alcohol in the breath.
Minn. Stat. § 634.16.
The parties offer alternative interpretations of this statute. Both Knapp and the Commissioner suggest that the statute addresses the foundational reliability—and hence
24
the admissibility—of breath-test results generally rather than being limited to the narrow question of the foundational reliability of the test instrument. But the parties draw distinct conclusions from that basic premise.
Knapp argues that section 634.16 only allows a test result to be admitted without antecedent expert testimony establishing that the breath-testing instrument provided a trustworthy and reliable measure of alcohol concentration in the breath sample if all protocols necessary to establish reliability were properly performed in the particular case. And Knapp asserts that an adequate observation period is one of those protocols.13The Commissioner agrees that, under the statute, test result admissibility requires considering whether the breath test was properly administered in the particular case, but argues that this consideration is limited to administration or use of the instrument. The Commissioner asserts that the observation period is not a protocol related to use of the instrument. Consequently, according to the Commissioner, if the record shows that a trained operator properly followed all steps needed to operate an approved instrument—a fact that is uncontested here—the breath-test results that the instrument generates are admissible.
We start with the text and structure of the statute. In re SIRS Appeal by Best Care, LLC, 26 N.W.3d 459, 471 (Minn. 2025). Section 634.16 is structured with a narrow focus
25
on one aspect of breath test admissibility: whether a breath-test instrument accurately determines a breath sample's alcohol-concentration measure. This is the question regarding the foundational reliability standard that we established in Quinn, where we held that breath-test results were admissible without antecedent expert testimony that the breath-test technology in use at the time accurately converted breath-alcohol measurements into blood-alcohol-concentration measurements. 182 N.W.2d at 845.14The requirement to establish a breath test's foundational reliability—and concomitantly its admissibility—is a technical, scientific question that is beyond the typical experience and knowledge of most lay jurors and is thus subject to expert opinion testimony. Minn. R. Evid. 701, 702. In section 634.16, however, the Legislature explicitly eliminated the need for antecedent expert testimony to establish that a breath-testing instrument provides a trustworthy and reliable measure of alcohol in the breath as long as particular conditions exist: the instrument is approved by the Commissioner, see Minn. Stat. § 169A.03, subd. 11 (defining an "approved breath-testing instrument"); and the instrument is operated by a person trained (in accordance with Commissioner-approved training) on the use of the breath-testing instrument. Minn. Stat. § 634.16; see In re Comm'r of Pub.
26
Safety, 735 N.W.2d 706, 711 (Minn. 2007) (referring to the statute as a presumption of admissibility). That is all the statute does.
Accordingly, we disagree with the premise of both parties' arguments—that the statute addresses breath-test result foundational reliability and admissibility generally. The language of the statute is specifically limited to whether the "breath-testing instrument provides a trustworthy and reliable measure of the alcohol in the breath."
Minn. Stat. § 634.16 (emphasis added). Further, within the context of that narrow question, the statute simply provides that, if the conditions in the statute are satisfied, antecedent expert testimony is not required to establish that the instrument provides reliable results measuring alcohol concentration. Any other concerns about the reliability of the breath-test result as a measure of the alcohol concentration of deep lung breath— including about the manner in which the breath test was administered in a specific case— remain foundational admissibility questions that the district court must consider.15The words "antecedent expert testimony" (which mirror the language in Quinn) support this interpretation because establishing the foundational fact that an instrument, if properly operated, accurately analyzes a breath sample and reports its results as an alcohol-
27
concentration measurement is where expert testimony may be necessary and helpful for a fact-finder without the statute providing a means to presume accuracy. We also disagree with Knapp's argument that the statute requires, as a condition of eliminating the need for expert testimony, that a breath test be "performed … pursuant to training given or approved by the commissioner of public safety or the commissioner's acting agent." Minn. Stat. § 634.16. The statute does not say that. The phrase "pursuant to training given or approved by the commissioner" modifies the phrase "by a person who has been fully trained in the use of [the instrument]." See Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 152-53 (2012) (discussing the nearest-reasonable-referent canon, which states that when the syntax involves something other than a parallel series of nouns or verbs, a postpositive modifier normally applies only to the nearest reasonable referent). In other words, the language addresses the type of training required and establishes that the person operating a testing instrument must be fully trained pursuant to Commissioner-approved training.16
28
In sum, under Minn. Stat. § 634.16, for purposes of admissibility, it is possible to establish foundation that a breath-test instrument accurately and reliably analyzed a breath sample and reported an alcohol-concentration measurement for that sample without expert testimony as long as (1) the Commissioner has approved the instrument, and (2) the operator was fully trained pursuant to Commissioner-approved training.
5.
Finally, we review our decision in Loving, which the Commissioner points to in arguing that questions of contamination are questions of weight, not admissibility. In Loving, we considered whether the district court properly admitted expert testimony that identified gunshot residue on a murder defendant's coat. 775 N.W.2d at 877. The parties agreed on the reliability of the SEM/EDX analysis used to determine whether there was gunshot residue on the coat and the expert's identification of gunshot residue on the coat. Id. at 878. The defendant, however, argued that the gunshot-residue results were generally unreliable because the test results themselves do not demonstrate how or when the gunshot residue got on the coat and, on the particular facts of the case, it was separate, adequate breath samples in the proper sequence," Minn. Stat. § 169A.51, subd. 5(c); or (2) fail twice to provide "two separate, adequate breath samples within 0.02 alcohol concentration," Minn. Stat. § 169A.51, subd. 5(d)-(f).
Section 169A.51, subdivision 5, is not about evidentiary admissibility as is section
634.16 (although a test operator's failure to conduct a sequenced test in accordance with subdivision 5(a) would not support a determination of test failure or refusal). More specifically, a person's failure to provide two samples with alcohol-concentration results within the .02 differential set forth in subdivision 5(d) does not—and the Legislature did not intend it to—provide definitive proof that the samples were not contaminated with mouth alcohol and affect test-result admissibility under section 634.16.
29
impossible to determine how or when the gunshot residue got on the coat or whether the shooter actually wore the coat. Id. at 877-78.
We disagreed, reasoning as follows:
[T]he experts for the State and for [the defendant] agreed that the SEM/EDX method was reliable in testing for [gunshot residue] on clothing. In addition, both experts agreed that the test had accurately identified [gunshot residue] on the coat. Lastly, the expert performing the test followed approved procedures while testing for [gunshot residue]. Thus, there was sufficient foundational reliability for the evidence. [The defendant's] argument about the inadequacy of the test's meaning is an argument about the weight of the evidence, not the reliability of the evidence. The multiple possibilities of how or when [gunshot residue] could have been transferred to the coat do not make SEM/EDX analysis or the [gunshot residue] evidence unreliable or unreliable as to him.
Id. at 878.
Relying on Loving, the Commissioner argues that questions of contamination are not subject to admissibility challenges but instead are always questions of fact for the fact-finder. As this relates to Knapp, the Commissioner argues that whether Knapp's breath sample was deep lung breath or contaminated with mouth alcohol is a question about the meaning of the test. We conclude, however, that the issue in Loving differs from the issue in this case.
In Loving, the gunshot-residue-test results were admitted for the purpose of showing that gunshot residue was found on the defendant's coat, and there was no dispute that it was gunshot residue; the gunshot-residue test result did not purport to prove how the residue got on the coat. See id. at 879 (pointing out that the jury was informed of the limitations of the presence of gunshot residue and what conclusions could or could not be drawn therefrom). Here, the purpose of the breath-test result is to determine whether
30
Knapp's alcohol concentration at the relevant time was .08 or more, which is itself determinative of an element of the offense. If Knapp's breath sample was contaminated with mouth alcohol, the test result—designed to measure alcohol concentration from deep lung breath—may not have provided an accurate measure of his alcohol concentration. Finally, since Dille, we have analyzed contamination of a sample in the context of alcohol-concentration tests through a reliability lens. Dille, 258 N.W.2d at 567-68. Thus, the Commissioner's reliance upon Loving in arguing that questions of contamination are not subject to admissibility challenges is misplaced.
6.
In summary, based on our analysis of the relevant cases and Minn. Stat. § 634.16, we conclude that the risk that a breath sample was contaminated could support a district court determination that the breath-test result reached with that sample is inadmissible, but that such a determination must be based on all record evidence bearing on whether the sample was contaminated, with the proponent of admissibility bearing the burden under a preponderance of the evidence standard. An imperfect observation period is not a per se reason to exclude the breath-test results.
To sum up the prior analysis of our precedent and Minn. Stat. § 634.16 more completely, the proponent of admitting a breath-test result must show that the test result is relevant and probative, see Minn. R. Evid. 401-403, and that it has foundational reliability, see Dille, 258 N.W.2d at 567. For admissibility purposes, the proponent may establish foundation that the breath-test instrument accurately and reliably analyzed a breath sample and reported an alcohol-concentration measurement from that sample
31
without expert testimony if two conditions are met: (1) the instrument has been approved by the Commissioner, and (2) the person operating the instrument was fully trained pursuant to Commissioner-approved training.17Minn. Stat. § 634.16; Quinn, 182 N.W.2d at 845.
The proponent of admitting a breath test must also establish foundation that the breath test was properly administered in the particular case, including that the breath test provided a reliable measure of the alcohol content of the driver's deep lung breath. Dille,
258 N.W.2d at 567. Under our alcohol-content-test-result-admissibility precedent, flaws in administering a breath test do not decisively establish that the breath-test result is inadmissible because it lacks foundational reliability. Id. at 567-68; Habisch, 313 N.W.2d at 16. Rather, the test result may be admissible if sufficient record evidence shows that the reliability purposes that the allegedly flawed protocol serves—here, preventing test results that are not a measure of the alcohol concentration in the driver's deep lung breath—were met. Dille, 258 N.W.2d at 568.
In assessing whether a breath-test result is admissible, and specifically whether the breath sample tested was not contaminated so as to provide a reliable measure of the alcohol content of the driver's deep lung breath, a district court should consider all the evidence bearing on that question. Ultimately, when a defendant challenges a test sample on the ground that it may have been contaminated, the district court should assess
32
admissibility under a preponderance of evidence standard, asking whether it is more likely than not that the test results reflect a reliable measure of the alcohol concentration of a driver's deep lung breath.18See Source Code, 816 N.W.2d at 539.
B.
Having explained the proper standard, we now consider the court of appeals' and district court's analysis here. Knapp's only challenge to the admissibility of the breath- test result in his case is that the result lacks foundational reliability because of an inadequate observation period.19Knapp argues that the court of appeals erred when it concluded that the breath-test result was admissible. In addressing whether Knapp is correct, we ask whether all the evidence bearing on the question establishes, by a preponderance of the evidence, that the contamination the observation period is meant to prevent did not occur. The purpose of an observation period, like the sterilization of a needle and bottle in Dille, is to prevent sample contamination. Thus, the question in this case is whether all pertinent evidence shows, by a preponderance of evidence, that Knapp's breath sample accurately tested deep lung air—i.e., it was not contaminated. If the answer is "Yes," then the district court should admit the test-result evidence.
33
In answering this question here, we decline to adopt the court of appeals' position that Knapp's failure to introduce direct affirmative evidence of mouth alcohol means the breath test is admissible. As discussed above, the burden is on the Commissioner to prove by a preponderance of the evidence that Knapp's breath sample was not contaminated. Under our precedent on the admissibility of test results for alcohol concentration, we also reject Knapp's position that an inadequate observation period, standing alone, is dispositive of admissibility. Inadequacies in an observation period are one factor to consider in assessing, under the preponderance of the evidence standard, whether it is more likely than not that a breath sample was not contaminated.20 Here, the district court assessed whether to admit Knapp's test result based solely on the adequacy of the observation period, finding the test result unreliable because "[t]he record … does not establish that [the deputy] continuously observed [Knapp] for the duration of the observation period." Thus, the district court abused its discretion because it based its decision on the wrong legal standard. See Riley, 792 N.W.2d at 833 (stating a district court abuses its discretion when it bases its ruling "on an erroneous view of the law"). Of course, the district court's findings about the observation period are relevant under the correct standard—whether a preponderance of evidence establishes that Knapp's breath sample was not contaminated. For instance, the district court found that in the squad car, "the deputy was facing away from [Knapp]," "had to rely on his rear-view
34
mirror to see [Knapp]," and that "his perspective was further limited by the Plexiglass barrier." The district court also found that, inside the jail, the deputy's "location is unclear relative to [Knapp]." Based on these findings, the district court determined that the record does not establish that the deputy observed Knapp "such that he was able to observe burping or belching 15 minutes prior to the administration" of the test. The record, however, includes other evidence bearing on whether Knapp's breath sample was contaminated. For instance, the deputy testified that from the time of arrival at the Washington County Jail to the time the breath test was administered, he could see and hear Knapp, and he did not observe Knapp burping, belching, or vomiting. In addition, the record shows that Knapp did not provide his first breath sample until 12:17 a.m.—15 minutes after Knapp and the deputy entered the jail—a fact which may diminish the significance of the deputy observing Knapp in the squad car. Video footage shows that once Knapp was inside the jail he was in proximity to the deputy for several minutes.
Based on the above, we conclude that the district court abused its discretion when it determined that Knapp's test result was unreliable because the court did not apply the appropriate analysis in assessing the test result's foundational reliability and, consequently, based its decision on an erroneous view of the law. The district court remains, however, in the best position to make that determination under the proper standard articulated in this opinion. Therefore, we remand to the district court to determine whether Knapp's breath-test results are admissible under the standard we articulate today.
35
CONCLUSION
For the foregoing reasons, we vacate the decision of the court of appeals, and in turn vacate and remand to the district court for proceedings consistent with this opinion. We leave to the district court's discretion whether to reopen the record on remand. Vacated and remanded.
ALIGADA, Jr., J., not having been a member of the court at the time of submission, took no part in the consideration or decision of this case.