Breath testing & the Intox EC/IR II
A GERD breath test DWI defense, or one based on mouthwash, dentures or diabetes, is only as good as the science and the facts behind it. Breath instruments are built to measure alcohol from deep in the lungs, so anything that puts alcohol in the mouth or throat can push a reading up. North Carolina guards against that with a 15-minute observation period and two-sample testing. Here is what the research shows, where those safeguards hold, and where they can fail.
Key takeaways
- Mouth alcohol from a recent drink, mouthwash, a burp or regurgitation can raise a breath reading, but it fades quickly, which is why NC requires 15 minutes of observation first.
- Controlled studies found GERD rarely causes falsely high evidential breath results, but one study did document elevated, irreproducible readings in some GERD subjects shortly after drinking.
- Dentures did not meaningfully affect results after a 20-minute wait in a forensic study; NC rules do not require removing dental devices.
- The Intox EC/IR II’s fuel cell, according to its manufacturer, does not respond to acetone, the compound most associated with diabetes and fasting.
What is mouth alcohol and why does it matter for a breath test?
Mouth alcohol is alcohol sitting in the mouth, throat or esophagus rather than coming from the lungs. A breath instrument assumes the sample is deep-lung (alveolar) air in equilibrium with the blood, so alcohol picked up on the way out of the mouth can make the reading higher than the person’s true alcohol concentration.
Common sources include a drink taken minutes before the test, alcohol-based mouthwash or breath spray, a burp or belch that brings stomach vapor up, and regurgitation or vomiting. The effect is temporary because alcohol in the mouth evaporates and is absorbed quickly. The legal system’s main answer is time: wait long enough that any mouth alcohol is gone, and watch the person so nothing new is introduced. Our guide to the Intox EC/IR II explains how the instrument decides when it has a deep-lung sample.
A period during which a chemical analyst observes the person to be tested to determine that the person has not ingested alcohol or other fluids, regurgitated, vomited, eaten, or smoked in the 15 minutes immediately before collection of a breath specimen. The analyst may observe while conducting the instrument’s operational procedures. Dental devices or oral jewelry need not be removed.
Read the rule (NC Administrative Code) ↗A breath result is admissible under G.S. 20-139.1(b) only if the test followed these DHHS rules and the analyst held a current permit. Our article on the 15-minute observation period covers that rule in depth.
Mouthwash, breath spray and a recent drink: how fast mouth alcohol fades
Mouthwash can produce a very high reading if the test is given immediately, but the effect drops off within minutes. That is the scientific basis for the observation period.
In a study published in JAMA (Modell, Taylor and Lee, 1993), ten people rinsed with Listerine, Scope or Lavoris and were tested on an Alco-Sensor III at 2, 4, 6, 10 and 15 minutes. Two minutes after rinsing, average readings were the equivalent of about 0.24 for Listerine, 0.17 for Scope and 0.036 for Lavoris. The values decayed exponentially, and by ten minutes the averages and the highest individual values were well below the 0.08 range for all three brands. The authors concluded mouthwash would not threaten accuracy under normal circumstances, but that using it immediately before a test could significantly raise the result.

The practical takeaway: mouthwash or breath spray matters when it was used close to the test, especially if the observation period was short, interrupted or poorly documented. A product used an hour before a stop is very unlikely to explain an evidential result. Roadside portable screening tests are a different story, because they are often given minutes after the stop; see what the roadside Alco-Sensor can and can’t prove.
Can GERD or acid reflux cause a false breath test result?
It can in limited circumstances, but the research says it is uncommon. GERD (gastroesophageal reflux disease) lets stomach contents move up into the esophagus, and if there is unabsorbed alcohol in the stomach, alcohol vapor could reach the breath sample.

Two controlled studies are most often cited. In a Swedish study (Kechagias et al., Journal of Forensic Sciences, 1999), ten people with severe GERD awaiting anti-reflux surgery drank a moderate dose of alcohol, and researchers compared breath and blood every 5 to 10 minutes for four hours, even using an abdominal compression belt to provoke reflux. Four subjects definitely experienced reflux, but it did not produce widely deviant breath readings, and the authors concluded that a falsely increased evidential result from reflux was highly improbable.
A later study (Booker and Renfroe, Journal of Forensic Sciences, 2015) dosed fifteen subjects, ten with GERD, to blood alcohol levels above 0.15 and compared blood and breath every 20 minutes for eight hours. After absorption, breath and blood agreed closely. But three subjects showed elevated breath alcohol readings, up to 0.105, during the absorptive phase, apparently from gastric alcohol passing the lower esophageal sphincter rather than from burping or regurgitation. Those contaminated samples appeared only when the stomach held a high alcohol concentration, and they were not reproducible in size.
Read together, the research suggests a GERD argument is strongest when drinking was recent (alcohol still in the stomach), the person has a documented reflux diagnosis, the two breath samples disagree or required a third sample, and the result is close to a legal line. It is weak when drinking ended hours earlier and the two samples agree closely. Because irreproducible readings are the signature of contamination, the second sample matters; North Carolina requires at least two that agree within 0.02 under G.S. 20-139.1(b3).
Burping, belching and vomiting during the observation period
Regurgitation and vomiting are specifically listed in the observation rule, so if they happen within 15 minutes of a sample, the rule has not been met. Burps are harder: they may be quiet, and they are not always noticed or written down.
The analyst does not have to stare at the person without blinking. In State v. Roberts (2014), the officer observed the defendant for 21 minutes but briefly lost direct sight while handling the defendant’s right to a witness; the Court of Appeals held the rule was still satisfied. On the other hand, in State v. Forney (2024), an officer noticed chewing gum in the defendant’s mouth after the first sample and took the second sample without restarting the observation period. The lead opinion said that violated the rules, but the three judges did not agree on that point, and all agreed the conviction stood because the other evidence of impairment was strong. Body-camera and booking-room video is often the only way to see what happened in those 15 minutes; see how to get dashcam and bodycam footage.
Dentures, bridges, braces and oral piercings
North Carolina’s rule says dental devices and oral jewelry need not be removed for a breath test, and the best forensic research supports that position once the waiting period has passed.
In a study by the Wisconsin State Laboratory of Hygiene (Harding et al., Journal of Forensic Sciences, 1992), 24 alcohol-free denture wearers swished brandy for two minutes, spat it out, and were tested on an Intoxilyzer 5000 with dentures removed, held loosely, and secured with adhesive. Beyond 20 minutes after spitting, mouth alcohol made no significant contribution to the readings, with only trace values of 0.01 or less in two subjects. The authors concluded dentures, with or without adhesive, do not significantly affect results if a 20-minute deprivation period is observed.

Note the detail: the study’s safe point was 20 minutes, while North Carolina’s observation period is 15. The difference rarely matters in practice, because most NC tests involve more than 15 minutes between the last drink and the first sample. It can matter when someone drank right up to the stop and was tested quickly. The timeline is the key; see our breath-test timeline case study.
Diabetes, ketosis, low-carb diets and acetone
Diabetes affects DWI cases more through behavior and odor than through the breath number. The Intox EC/IR II measures alcohol with a fuel cell that, according to its manufacturer, does not respond to acetone, the compound produced in diabetic ketosis and by fasting or very-low-carbohydrate diets.
Acetone and dieting have a long history in breath-test litigation. The U.S. Supreme Court noted in California v. Trombetta (1984), a case about an Intoxilyzer, that defendants could challenge results with evidence of chemicals that appear in the blood of people who are dieting. North Carolina’s instrument measures alcohol with a fuel cell, and its manufacturer states the sensor “does not respond to acetone or other substances found in the human breath after a fifteen minute deprivation period.” That is a manufacturer claim about specific substances, not about every chemical; unusual exposures such as industrial solvents should be raised with a toxicology expert.
Diabetes still matters in three ways. Diabetic ketoacidosis can produce a fruity or sweet breath odor an officer may mistake for alcohol. Low blood sugar can cause confusion, slurred speech and poor coordination that look like impairment, which is why NHTSA trains officers to ask drivers whether they are diabetic. And a medical emergency can explain driving behavior. These points go to the officer’s impairment opinion and probable cause rather than the breath number.

How North Carolina courts handle these challenges
North Carolina courts treat the observation period and sequential-sample rules as real requirements, but they read them practically and ask whether any violation actually mattered. The cases below frame how mouth-alcohol and medical arguments are litigated.
After gum was found in the defendant’s mouth between samples (both 0.11), the lead opinion said the observation period should have been restarted; no majority agreed on that point, and the conviction was affirmed because the other evidence of impairment made any error harmless.
Read the UNC School of Government analysis ↗The observation rule does not require the analyst to watch without ever looking away; 21 minutes of observation with only brief lapses in direct sight satisfied the rule, and the motion to suppress was properly denied.
Read the opinion ↗Explained that sequential testing exists to ensure accuracy because factors like extraneous alcohol in the mouth, body temperature and breathing pattern can affect a single reading; readings of 0.20 and 0.19 with insufficient samples between them were valid.
Read the opinion ↗Police need not preserve breath samples, partly because defendants can attack results in other ways: faulty calibration, extraneous interference (the parties cited radio waves and chemicals in dieters’ blood), and operator error shown through cross-examination.
Read the opinion ↗Where the defendant was continuously observed, an insufficient sample and instrument reset between two Intox EC/IR II results did not require a new observation period; the results were sequential and admissible.
Read the UNC School of Government summary ↗More decisions are collected in our NC DWI case-law library. Because a breath result can be the entire case on the 0.08 theory, our breath-testing records guide explains which documents to request.
Documenting a medical or mouth-alcohol issue
These challenges succeed or fail on records. Gather what existed before the arrest, and preserve what shows the 15 minutes before each sample.
- Diagnosis records for GERD, hiatal hernia or other reflux conditions
- Prescription history for reflux medication (for example, proton-pump inhibitors)
- Dental records describing dentures, bridges or other appliances
- Diabetes records, glucose meter or continuous-monitor data for that day
- Exact time of the last drink and the last mouthwash, spray, gum or food
- Breath ticket showing each sample time and result, and any third sample
- Video of the testing room covering the observation period
- Analyst’s observation-period start time and any notes of burping or vomiting

A blood test can also help. Under G.S. 20-16.2(a), the implied-consent rights include the right to have a qualified person of your own choosing administer an additional chemical test; see implied-consent rights, line by line. A blood result is not affected by mouth alcohol, though it raises its own issues explained in blood and toxicology evidence. Timing matters for any result; see rising BAC and retrograde extrapolation.
Mouth alcohol is a timing problem. The question is never just “do you have GERD?” It is what happened in the fifteen minutes before each breath.Adkins & Purkey
From our book
Charged with DWI in North Carolina?
Our book explains the science of breath testing, the observation period and the medical issues that can affect both the test and the officer’s impairment opinion.
Get the ebook at Barnes & Noble ↗ Amazon Kindle ↗ Amazon paperback (print) ↗
GERD, mouth alcohol and breath tests: FAQ
Can GERD cause a false positive on a DWI breath test?
It is possible but uncommon. A 1999 study of severe GERD patients found reflux did not produce widely deviant evidential readings, while a 2015 study found elevated, irreproducible readings in three of fifteen subjects shortly after heavy drinking, when alcohol was still in the stomach.
Does mouthwash affect a breath test?
Yes, briefly. In a 1993 JAMA study, readings two minutes after rinsing averaged about 0.24 for Listerine, but fell well below 0.08 within 10 minutes. North Carolina’s 15-minute observation period is designed to let mouth alcohol dissipate.
Do I have to remove dentures for a breath test in North Carolina?
No. 10A NCAC 41B .0101(6) says dental devices and oral jewelry need not be removed, and a 1992 forensic study found dentures did not significantly affect results after a 20-minute wait.
Can diabetes affect a breath test?
The Intox EC/IR II’s manufacturer states its fuel cell does not respond to acetone, the compound associated with diabetic ketosis and fasting. Diabetes is more likely to matter because low blood sugar and ketoacidosis can mimic signs of impairment.
What if I burped or vomited before my breath test?
The observation rule requires that the person not have regurgitated or vomited in the 15 minutes before a sample. If that happened and the period was not restarted, it can be a basis to challenge the result. Video and the analyst’s notes are key evidence.
Adkins & Purkey, PLLC
Talk to a North Carolina DWI lawyer
If reflux, mouthwash, dental work, diabetes or a burp before the test may have affected your breath result, the timeline and the testing-room video need to be gathered quickly. Offices in Huntersville (Lake Norman) and Locust (Stanly County), serving Charlotte, Mecklenburg, Iredell, Cabarrus, Stanly and courts across North Carolina.
General North Carolina legal information as of November 2026, not legal advice for any specific case. Reading this article does not create an attorney-client relationship. Attorney advertising.



