Breathalyzer results can be thrown off by a surprisingly wide range of factors that have nothing to do with how much you drank. Residual alcohol trapped in your mouth, acid reflux pushing stomach vapors upward, elevated body temperature, certain medical conditions, and even the hand sanitizer used by the person administering the test can all distort readings. Some of these effects are large enough to push a result above the legal limit when your actual blood alcohol is below it, and the science behind each one is more specific than most people realize.
Residual Mouth Alcohol
The most well-known cause of an artificially high reading is alcohol lingering in your mouth rather than coming from your lungs. When you burp, use mouthwash, or have recently taken a sip of an alcoholic drink, ethanol vapor sitting in your oral cavity gets swept into the breath sample alongside the deeper air from your lungs. Because the instrument measures all the ethanol in the sample without distinguishing its source, even a small amount of mouth alcohol can inflate the number substantially.
This is exactly why law enforcement protocols require a waiting period before testing. Research on denture wearers, who were once thought to trap mouth alcohol in the surfaces around their dental plates, found that after a 20-minute deprivation period, residual mouth alcohol made no meaningful contribution to the reading. Only trace amounts were detected in a couple of subjects, and denture adhesives did not change that outcome.1Journal of Forensic Sciences. The Effect of Dentures and Denture Adhesives on Mouth Alcohol Retention The 15-to-20-minute observation window used by most jurisdictions exists specifically to let mouth alcohol dissipate. If that window is cut short or the officer does not notice a belch during the wait, the result can be unreliable.
Acid Reflux and Gastric Alcohol
Gastroesophageal reflux disease, commonly known as GERD, creates a direct pathway for alcohol vapor to travel from the stomach into the throat and mouth without a visible belch. In a controlled study, subjects diagnosed with GERD were dosed to blood alcohol levels above 0.150 g/dL. Three of those subjects produced elevated breath readings as high as 0.105 g/dL during the absorptive phase, apparently caused by gastric alcohol leaking through the lower esophageal sphincter without any obvious burp or regurgitation.2PubMed. The Effects of Gastroesophageal Reflux Disease on Forensic Breath Alcohol Testing
A few details from that study are worth noting. The contaminated breath samples appeared only when there was a high concentration of unabsorbed alcohol still in the stomach, which is most likely in the early period after drinking. And the effect was inconsistent: the elevated readings were not proportional to how much gastric alcohol remained, and they were irreproducible in magnitude from one test to the next. That inconsistency is itself a red flag. If two back-to-back breath samples differ wildly, mouth or stomach contamination is a likely explanation, and most modern testing protocols require two readings that agree within a narrow range for exactly this reason.
Body Temperature
Breathalyzers are calibrated around an assumed body temperature of about 34°C at the mouth, reflecting the cooling that occurs as air travels from the lungs outward. When your actual body temperature is higher than normal, more ethanol crosses from the blood into the air in your lungs, and the machine reads that extra ethanol as a higher blood alcohol level than you truly have.
In a study that raised subjects’ core body temperature by about 2.5°C using a warm water bath, breath alcohol readings climbed as much as 23% above their actual blood alcohol concentration. The researchers calculated that each degree Celsius of elevated body temperature inflated the breath reading by roughly 8.6% relative to blood alcohol.3PubMed. Effect of hyperthermia on breath-alcohol analysis A fever of even one degree above normal can therefore shift the result enough to matter near a legal threshold. The blood alcohol itself was unaffected by the temperature change; only the breath reading moved. Research on the blood-to-breath ratio has confirmed this relationship, showing that subjects with higher breath and body temperatures consistently had lower partition ratios, meaning more ethanol appeared in their breath per unit of blood alcohol.4PubMed Central. Reflections on variability in the blood–breath ratio of ethanol and its importance when evidential breath-alcohol instruments are used in law enforcement
This matters for practical scenarios beyond illness. Vigorous physical exertion, sitting in a hot car, or even anxiety-driven increases in body temperature can nudge the reading upward. The effect is not dramatic at small temperature differences, but it is real and well-documented.
How Breathing Patterns Affect the Result
The way you blow into the device changes what it measures. Breath alcohol concentration is not uniform across the entire exhalation. Air that comes out first, from the upper airways, carries less ethanol than the deep alveolar air that exits last. A long, steady blow delivers more of that deeper, ethanol-rich air, producing a higher reading than a shallow puff.
Researchers found that hyperventilation before the test significantly lowered the breath alcohol reading compared to a normal blow. Immediately after hyperventilating, readings dropped measurably, and the suppression was still detectable ten minutes later.5PubMed Central. Manipulation of Breath Alcohol Tests: Can Specific Techniques Alter Breath Alcohol Content? A poor-effort blow, where the subject did not fully exhale, also produced lower readings. Separate work confirmed the same pattern: shallow breathing and hyperventilation both led to underestimation of breath alcohol concentration.6PubMed. Influence from breathing pattern on alcohol and tracer gas expirograms–implications for alcolock use
The practical takeaway cuts both ways. Someone who hyperventilates (whether deliberately or from nervousness) could produce a reading that underestimates their actual blood alcohol. And someone who holds their breath before blowing could in theory concentrate the ethanol in their lungs and produce a higher reading. Modern evidential instruments try to account for this by requiring a minimum volume and a sustained blow, but the breathing pattern still introduces variability.
Hand Sanitizer and Environmental Alcohol Vapor
This one catches people off guard. Alcohol-based hand sanitizer, which typically contains 60-70% ethanol, can generate detectable breath alcohol readings in the person being tested and even in nearby bystanders. In a study that divided participants into groups based on proximity, subjects who applied hand sanitizer to their own hands registered a median breath reading of about 0.05 g/dL. Those who stood in an enclosed space with multiple people using sanitizer around them registered a median of roughly 0.12 g/dL, well above the legal limit in most jurisdictions, despite having consumed no alcohol at all.7PubMed. Common hand sanitizer may distort readings of breathalyzer tests in the absence of acute intoxication
A separate study in an emergency department setting found that alcohol was detected in breath one minute after healthcare workers used alcohol-based hand hygiene products, regardless of whether the product was a foam or gel, and even when the worker wore gloves afterward. The researchers recommended that staff administering breath tests use soap and water instead of alcohol-based sanitizers.8PubMed. Effects of alcohol-based hand hygiene solutions on breath alcohol detection in the emergency department The effect is temporary, lasting only a few minutes, but it demonstrates how sensitive these instruments are to airborne ethanol from any source. If the testing officer sanitized their hands moments before holding the mouthpiece, or if the test happens in a small room where sanitizer was recently used, the reading can be contaminated.
Diabetic Ketoacidosis and Ketone Interference
People with uncontrolled diabetes, especially those in diabetic ketoacidosis (DKA), produce high levels of acetone as a byproduct of fat metabolism. That acetone is the source of the “fruity breath” sometimes noticed in DKA patients. The problem for breathalyzer accuracy is that the body can convert acetone into isopropanol, a type of alcohol that some breath testing instruments detect and misidentify as ethanol.9PubMed. Early Detection of Diabetic Ketoacidosis by Breathalyzer in a Sailor Reporting for Duty
A documented case involved a U.S. Navy sailor who registered a positive breathalyzer reading while reporting for duty, despite not having consumed alcohol. The elevated reading was attributed to the metabolic ketoacidosis the sailor was experiencing. This interference is most relevant for fuel-cell-based portable breath testers, which can be less specific in distinguishing ethanol from other alcohols. Infrared-based evidential instruments used at police stations are generally better at filtering out non-ethanol compounds, but they are not immune to interference at very high acetone concentrations.
This does not only affect people with Type 1 diabetes. Certain diabetes medications, particularly SGLT2 inhibitors, can trigger DKA even without dramatically elevated blood sugar, a scenario called euglycemic DKA.10Journal of Breath Research. Breath acetone as a potential marker in clinical practice People following very low-carbohydrate or ketogenic diets also produce elevated acetone, though typically at far lower levels than those seen in DKA. Whether those lower levels produce meaningful breathalyzer interference depends on the instrument’s specificity, and the evidence on that question is thinner than many internet claims suggest.
Auto-Brewery Syndrome
In rare cases, a person’s gut microbiome ferments carbohydrates into ethanol without any alcohol being consumed. This condition, called auto-brewery syndrome (also known as gut fermentation syndrome), produces genuine endogenous alcohol that is absorbed into the bloodstream and exhaled in the breath exactly like consumed alcohol would be.11PubMed Central. Auto-Brewery Syndrome: A Clinical Dilemma
The condition is genuinely rare, with only a handful of well-documented cases in the medical literature. It tends to arise in people who have had intestinal surgery, take long courses of antibiotics that alter gut flora, or have fungal overgrowth conditions. What makes it legally relevant is that the ethanol produced is real ethanol, not an interfering substance that the instrument misreads. A breathalyzer in this scenario is reading actual alcohol. The “false” part is that the person did not drink. Whether this constitutes a legal defense depends on the jurisdiction, but from a scientific standpoint, the breath test is accurately measuring something the person did not choose to consume.
Asthma Inhalers and Medications
Some metered-dose inhalers contain small amounts of ethanol as a co-solvent or use propellant gases that can register on breath testing instruments. A study of 69 asthma patients using standard bronchodilator inhalers found that these devices could generate false positive breath alcohol readings in some patients. The effect was short-lived and could be prevented by waiting 5 to 10 minutes between inhaler use and testing.12PubMed. A comparison of standard inhalers for asthma with and without alcohol as the propellant on the measurement of alcohol in breath
Beyond inhalers, a number of liquid medications, cough syrups, and oral sprays contain ethanol as a solvent. The effect is similar to mouthwash: it creates residual mouth alcohol that dissipates within 15 to 20 minutes. If you have used any alcohol-containing oral product shortly before a breath test and the observation period was not properly followed, the reading may be inflated.
Occupational Solvent Exposure
Workers exposed to organic solvents like paint thinner sometimes worry that inhaled vapors will trigger a positive breath test. The evidence here is reassuring. In a controlled study, volunteers inhaled white spirit vapor at workplace-relevant concentrations, including one scenario where they painted with gloss paint in an unventilated room at concentrations reaching 185 parts per million for 20 minutes. After all exposures, breath tests on evidential instruments produced readings that were extremely small and never exceeded a trivial threshold for breath samples taken more than 10 minutes after the exposure ended.13PubMed. The response of evidential breath alcohol testing instruments with subjects exposed to organic solvents and gases. II. White spirit and nonane
Modern infrared breath testing instruments are designed to distinguish ethanol’s absorption signature from that of other volatile compounds. While some older or less sophisticated devices might have been more vulnerable to solvent interference, the current generation of evidential instruments appears to handle this well. The risk is highest with cheap portable screening devices that use less specific detection methods.
Lung Disease and Providing a Valid Sample
People with chronic lung conditions face a different kind of problem: they may not be able to blow hard enough or long enough to provide what the instrument considers a valid breath sample. When a person with severely reduced lung function cannot deliver the required volume or flow rate, the test either fails entirely or captures a sample that is not truly representative of deep lung air.
In a study using the Dräger 6510 evidential instrument, all asthma and COPD participants were able to provide a valid breath specimen, with an overall failure rate of just 1.3%. The only participant who failed was in the interstitial lung disease group with moderate severity.14Journal of Forensic and Legal Medicine. A clinical investigation into the ability of subjects with a lung disease to provide breath specimens using the Dräger 6510 A separate study using a different instrument found a higher failure rate: about 24% of lung disease volunteers could not provide a specimen, and the key predictor of failure was how severely their forced expiratory volume was reduced.15PubMed. A clinical investigation into the ability of subjects with lung disease to provide breath specimens using the EvidenzerIRL evidential breath analyser in alcohol intoxicant driving in criminal justice evidence
The variation between instruments matters here. Different evidential devices have different acceptance criteria for what constitutes a valid blow. A device that demands higher flow rates or longer exhalation will fail more people with lung disease, while one with more relaxed thresholds may accept a sample that is not fully representative of alveolar air. For someone with moderate-to-severe lung disease, a blood test is a more reliable way to determine alcohol level.
How Often Breath and Blood Results Disagree
Given all these potential interferences, a natural question is how well breathalyzer results match actual blood alcohol levels in everyday law enforcement use. A study that compared over 400 paired breath and blood results from Wisconsin drivers found that the breathalyzer result was lower than the blood result by more than 0.01 g/dL about 61% of the time, within 0.01 g/dL about 33% of the time, and higher than the blood result by more than 0.01 g/dL only 6% of the time.16PubMed. Breathalyzer accuracy in actual law enforcement practice: a comparison of blood- and breath-alcohol results in Wisconsin drivers
In other words, breathalyzers underestimate blood alcohol far more often than they overestimate it. The mean breath result in that dataset was 0.16 g/210 L, while the mean blood result was 0.176% w/v. This built-in tendency to read low is partly by design: the assumed blood-to-breath ratio used by most instruments is set conservatively to reduce the chance of a false high reading. But “reduces the chance” is not the same as “eliminates the chance.” That 6% of cases where the breath reading exceeded the blood result represents real-world instances where the instrument overestimated someone’s alcohol level, and the causes discussed throughout this article are the likely explanations.
The Blood-to-Breath Ratio Problem
Breathalyzers do not directly measure alcohol in your blood. They measure alcohol in your breath and then multiply by a fixed conversion factor to estimate blood alcohol concentration. In most countries, this factor is set at 2,100:1 or 2,300:1, meaning the instrument assumes that 2,100 (or 2,300) milliliters of breath contains the same amount of ethanol as 1 milliliter of blood. The trouble is that this ratio varies from person to person and even within the same person at different times.
A controlled drinking study involving 100 healthy volunteers from three ethnic groups found a mean blood-to-breath ratio of about 2,382 in the post-absorptive phase, with a standard deviation of 119. That spread means some individuals naturally have ratios well below the assumed conversion factor, and for them, the breathalyzer will consistently overestimate blood alcohol. Others have ratios well above the assumed value and will be underestimated.4PubMed Central. Reflections on variability in the blood–breath ratio of ethanol and its importance when evidential breath-alcohol instruments are used in law enforcement Body temperature, breath temperature, and how long the person exhaled all influenced the ratio. Longer exhalation times pushed the ratio lower, meaning longer blows produced higher breath readings per unit of blood alcohol.
This is not a flaw that can be fixed by better engineering. It is an inherent limitation of converting between two biological compartments whose relationship depends on physiology that the instrument cannot measure. The fixed ratio is a compromise: it works reasonably well for most people most of the time, but it introduces a margin of error that no amount of calibration can eliminate.
Foods and Beverages That Create Ambiguity
Claims about foods causing false breathalyzer readings are widespread online, with bread, ripe fruit, and energy drinks commonly named. The reality is more mundane. Any food or drink that contains trace ethanol can produce residual mouth alcohol immediately after consumption, but this dissipates within the standard 15-to-20-minute waiting period, just as mouthwash or denture adhesive residue does.
Fermented beverages like kombucha occupy a gray area. Regular kombucha typically contains under 0.5% alcohol by volume, which is low enough to be sold as non-alcoholic in most places. A study of kombucha consumption found that alcohol biomarkers appeared in the urine of most participants after drinking regular kombucha, confirming that some ethanol is absorbed.17Oxford Academic (Journal of Analytical Toxicology). Evaluation of Alcohol Markers in Urine and Oral Fluid after Regular and Hard Kombucha Consumption Whether this tiny absorbed dose would register on a breath test depends on how much you drank and how recently. A single bottle of regular kombucha is unlikely to move the needle past the observation-period safeguard. Hard kombucha, which can contain 5-8% alcohol, is a different story and is effectively an alcoholic drink.
The pattern across all dietary causes is the same: they either produce short-lived mouth alcohol that the waiting period handles, or they introduce genuinely absorbed ethanol that the instrument correctly detects. The “false reading from bread” scenario that circulates in legal defense folklore has very little scientific support when proper testing protocols are followed.