EVIDENCE LIBRARY
Does Urine Alcohol Prove a Hangover? What a Morning-After Study Measured
A 36-person study found residual morning urine ethanol, but no significant correlation with overall hangover severity. Learn what the urine result cannot prove.
Short answer: A measured urine ethanol concentration cannot, by itself, establish whether someone has a hangover, how severe that person’s symptoms are, what their blood alcohol concentration (BAC) was, or whether they were impaired. In a small naturalistic human study, residual ethanol was found in 31 of 36 morning-after urine specimens, but the single overall hangover-severity rating did not correlate significantly with urine ethanol concentration. Certain individual symptoms correlated with urine ethanol in the subgroup that reported hangovers. Those group-level associations are not a validated diagnostic or forensic conversion rule.
This distinction matters when a morning-after urine test is offered as proof of an individual’s clinical condition, concentration at an earlier time, ability to drive, or cause of an event. A laboratory measurement is one kind of evidence; symptom reporting, functional testing, and a reconstruction of past alcohol exposure are different kinds.
What question did the human study actually test?
Van de Loo and colleagues studied the relationship between ethanol measured in urine at 9:30 a.m. and self-reported hangover symptoms following ordinary nights of drinking. Their work, published online in 2016 and in the January 2017 issue of Psychopharmacology, is an observational study—not a controlled dose-response trial or validation of a legal impairment test. The full report is available at DOI: 10.1007/s00213-016-4437-0.
There were 36 healthy social drinkers, aged 18–30, recruited in two equal groups. Eighteen reported experiencing hangovers after heavy drinking; the other 18 reported that they generally did not. The researchers called the latter the hangover-immune group, but that label describes participants’ reports rather than a demonstrated biological immunity. Each participant attended a post-drinking test day and a separate control day when no alcohol had been consumed the day before.
Drinking was not administered or directly observed by the investigators. Participants chose their own drinks, quantities, timing, and settings. They later reported their drinking history, and the investigators estimated peak alcohol concentrations from those reports and a formula. They did not obtain a measured peak BAC during the evening.
What was measured, and what was only estimated?
On the test mornings, the investigators collected urine and measured its ethanol content by headspace gas chromatography with flame-ionization detection (GC-FID). A calibrated chemical method can quantify ethanol in the submitted urine specimen. It does not automatically quantify alcohol in blood, recreate the maximum concentration from the prior night, or assess neurologic performance.
Participants separately rated overall hangover severity and individual symptoms such as headache, nausea, concentration difficulty, and sleepiness. These were subjective symptom scores, not instrumented driving, attention, balance, or decision-making tests. Because the drinking episodes were unsupervised, the amount and timing of alcohol exposure were also different from a carefully controlled dosing study.
| Evidence | How the researchers obtained it | What it can and cannot establish |
|---|---|---|
| Morning urine ethanol | Measured urine specimen using headspace GC-FID | Amount of ethanol in that urine sample; not a simultaneous blood BAC or impairment score |
| Estimated previous-night peak alcohol concentration | Calculation from recalled drinks, timing, sex, and body weight | A model-derived approximation; not a directly measured peak BAC |
| Overall hangover severity | Participant self-rating on a 0–10 scale | Reported symptom burden; not objectively demonstrated performance loss |
| Individual symptom ratings | Participant-reported nausea, sleepiness, headache, and other symptoms | Perceived symptoms; not proof that urine ethanol caused any one symptom |
| Alcohol elimination or metabolism rate | Not directly measured as a serial individual concentration-time profile | Cannot establish whether a particular participant was a fast or slow metabolizer |
For a general explanation of why a specimen and its analytical method must be kept together, see our guide to alcohol testing by headspace GC-FID. For concentration changes and their limits, see pharmacokinetics in forensic toxicology.
What did the study find about residual urine alcohol?
On the post-drinking morning, ethanol was present in 31 of 36 urine samples (86.1%). Across all 36 participants, morning urine ethanol concentrations were higher after drinking than on the control day (the Results section reports p = 0.002). The symptomatic group also had higher morning urine ethanol than the group that reported no hangover, with a between-group result of p = 0.027.
These findings show that residual urine ethanol and the study’s group classifications were associated under the conditions observed. They do not mean that 86.1% of participants had an objectively diagnosed hangover, that every urine-positive individual was symptomatic, or that the urine result determined the severity of any particular person’s symptoms. The investigators expressly reported that the single overall hangover-severity score did not significantly correlate with urine ethanol concentration.
The most informative results concern individual symptoms. Among the 18 participants who experienced hangovers, the study reported positive correlations between urine ethanol and several symptom ratings. For example, the correlation for concentration problems was r = 0.633 and for sleepiness r = 0.646. These were relationships between two measured or reported variables across people, not evidence that a specific urine concentration predicts a specific person’s cognitive performance. Among the 18 self-described hangover-immune participants, none of the urine-ethanol correlations with individual hangover symptoms reached statistical significance.
Why a positive urine ethanol result is not a hangover diagnosis
Urine ethanol reflects alcohol in the collected urine matrix. A hangover is a multidimensional collection of symptoms after drinking. They overlap in some circumstances, but the two concepts are not interchangeable. Even in this selected sample, the relationship depended on how symptoms were grouped and whether individual symptoms or a single overall rating were examined.
There is no diagnostic urine ethanol cutoff established by this study for saying “hangover present,” “hangover absent,” or “hangover severity 7 of 10.” Nor did the study derive a sensitivity, specificity, positive predictive value, or externally validated classification rule. Calling a positive urine result a diagnostic test for hangover would assign a clinical performance characteristic the researchers did not measure.
Conversely, an undetected or low urine ethanol concentration is not proof that an individual could not be experiencing next-day symptoms. A person’s report of feeling poorly and an analytical measurement of ethanol represent different observations, collected with different tools and limitations.
Can morning urine alcohol be converted into earlier blood BAC?
Not from this study. The alcohol concentration of a urine sample is not a direct measurement of simultaneous blood ethanol. The concentration in collected urine may reflect alcohol entering the bladder over an earlier interval, and the timing of the last void can affect the interpretive window. The authors specifically noted that they had not recorded whether participants urinated overnight or early that morning.
The study likewise did not standardize elapsed time from the last drink to the 9:30 a.m. collection, hydration, or amount of water consumed. Those unknowns matter even before considering questions about dose, absorption, elimination, and analytical uncertainty.
Its previous-night peak alcohol values were estimated from drinking reports. They are not serial blood data from which an individual elimination slope can be verified. The fact that the two groups did not differ statistically in reported drink count or estimated peak concentration does not prove matched physiologic exposures for every individual.
A valid attempt to interpret a case would first identify specimen type, collection and voiding history, analytical method, reporting units, preservation, and the question being asked. Our alcohol and volatiles evidence guide discusses that specimen-first approach. A numerical urine-to-blood conversion, retrograde BAC calculation, or time-of-drinking conclusion would require independent supporting evidence; this observational symptom study does not provide it.
Does the study show who was impaired the next morning?
No. Participants rated subjective experiences, including concentration problems, but the researchers did not administer a validated driving test, structured roadside field-sobriety test, psychomotor vigilance battery, reaction-time assessment, or simulated driving task as an outcome of this investigation. An association between a urine concentration and a self-reported concentration problem is not interchangeable with a measured deficit on one of those tests.
Similarly, the study did not establish that urine ethanol caused symptoms or that a symptom caused a crash, error, workplace incident, or medical event. Symptoms can be affected by sleep, beverage consumption, individual variability, and other factors. The paper cannot substantiate an individualized conclusion about fitness to drive, a statutory BAC threshold, tolerance, or fault.
Could “hangover immunity” simply mean faster ethanol elimination?
The authors proposed differential metabolism as one possible explanation for differences between people who reported hangovers and those who did not. That suggestion is a hypothesis, not a demonstrated mechanism in the study. There were no directly observed drinking-session peaks and no sequential measured blood concentration profiles adequate to calculate individual elimination rates. Without those measures, the results cannot classify particular people as unusually rapid metabolizers.
Other possibilities also remain unresolved by this design. The researchers discussed ethanol metabolites, including acetaldehyde, as possible contributors. Their urine chromatograms showed acetaldehyde and acetone peaks, but they did not validate either compound in this trial as a biomarker that diagnoses hangover or explains symptom causation.
Statistical caution also matters. Symptom correlations were examined in small subgroups across many outcomes. Multiple testing makes individual nominally significant p-values more vulnerable to chance findings; the paper was not a prospective external validation study. The strength of a subgroup correlation cannot be assumed to transfer to another population or a single litigated case.
What should be checked before using this study in a forensic opinion?
| Question in dispute | Relevant evidence | Boundary of the 36-person study |
|---|---|---|
| Was ethanol detected in the morning sample? | Specimen identity, collection time, validated urine ethanol analysis and quality controls | The paper used morning urine GC-FID; its statistical association is not a substitute for the case laboratory record |
| What was the person’s blood BAC at an earlier event? | Reliable blood or breath measurements, documented timings and an independently justified kinetic analysis | Previous-night peak values in the paper were model estimates, not measured peaks |
| Was the person experiencing a hangover? | Contemporaneous symptom evidence and a defensible clinical framework | No validated urine concentration threshold for an individual hangover diagnosis |
| Was the person cognitively or motor impaired? | Relevant task performance, medical examination and temporal context | No objective driving or psychomotor performance endpoint was measured |
| Did alcohol or hangover cause an event? | Complete event facts, alternative explanations, timing and reliable causal analysis | Naturalistic correlations cannot establish individual event causation |
Interpreting a result requires the primary case evidence, not simply a citation to a group-level association. The useful question is not “Was this urine alcohol positive?” alone; it is “What exactly was measured, when, and what inference is scientifically supported by that observation?”
Frequently asked questions
Can a urine alcohol test prove that someone has a hangover?
No. The study found group associations but did not validate a urine ethanol level that diagnoses hangover in an individual.
Does a higher urine ethanol concentration mean proportionally worse hangover severity?
No proportional prediction rule was established. In fact, the single overall hangover-severity rating did not significantly correlate with urine ethanol concentration.
Why were 31 of 36 morning samples positive?
The participants had recently consumed alcohol in naturalistic settings, and ethanol remained detectable in many morning urine specimens. The 31/36 figure describes residual urine ethanol detection in this sample—not the proportion with proven impairment.
Were previous-night peak BACs measured in the participants?
No. Peak alcohol concentrations were estimated from self-reported drinking patterns and a formula. They were not directly measured during the drinking episodes.
Does feeling hungover establish unsafe driving performance?
Not on the basis of this paper. It assessed self-reported symptoms, not actual driving performance or a standardized psychomotor test. Functional impairment requires separate evidence.
Can urine alcohol be converted to a blood BAC or a time of drinking?
This study supplies no validated individual conversion or timing method. Specimen differences and unknown voiding, hydration, and sampling intervals limit that inference.
Did the paper prove that some people eliminate ethanol faster and are immune to hangovers?
No. Faster metabolism was discussed as a possible explanation, but elimination rates were not directly determined in the study’s participants.
Bottom line: measured urine ethanol and a person’s condition are different questions
In the Van de Loo et al. study, next-morning urine ethanol was often detectable and was associated with certain self-reported symptoms in a small subgroup. That evidence is scientifically interesting but narrow. It does not validate a hangover diagnosis from urine ethanol, a measured blood BAC, an earlier-event reconstruction, a driving impairment inference, or a causal opinion about an individual incident. A forensic interpretation should state which propositions are direct measurements, which are estimates, and which require independent evidence.
Primary scientific source: Van de Loo A, Mackus M, Korte-Bouws G, Brookhuis K, Garssen J, Verster JC. “Urine ethanol concentration and alcohol hangover severity.” Psychopharmacology. 2017;234:73–77. Published online September 28, 2016. https://doi.org/10.1007/s00213-016-4437-0.
Scientific education only. Conclusions about an individual specimen or event require the complete case record and a qualified case-specific assessment.