EVIDENCE LIBRARY
Can a Zero Breath Alcohol Reading Exclude Next-Morning Driving Deficits?
A 52-driver naturalistic study found poorer simulated driving after drinking, including among 26 people with zero breath readings. Residual-alcohol groups differed on some outcomes, and sleep and blood alcohol were not directly established.
A zero breath reading did not exclude next-morning deficits on this simulated commute. In a study of 52 young drivers, several driving and divided-attention measures were poorer after a night of drinking, including in the subgroup whose breath instrument read zero. The findings do not establish that every hungover person is impaired or that a zero breath reading proves zero blood alcohol.
A morning commute rather than a laboratory reaction test
Alford and colleagues combined data from three naturalistic studies. Each participant completed a morning drive after no alcohol and another after a social drinking night. Condition and scenario order were counterbalanced.
The approximately 20-minute simulation mixed rural and urban roads with traffic lights, parked vehicles, pedestrians and other hazards. Drivers also responded to changing symbols as a divided-attention task.
The 52 participants were aged 18–25, with one to six years of driving experience. They chose their own drinking and other nighttime activities. This preserved real-world variation but limited control over dose, sleep and other influences.
Zero and residual alcohol were breath-based categories
Before driving, 26 participants had zero instrument readings. The other 26 had residual breath-derived BAC estimates averaging 0.047%, ranging from 0.01% to 0.08%. These categories were assigned after data collection, rather than randomly allocated.
No simultaneous blood assay confirmed the zero readings. Accordingly, “zero breath” is the supported measurement description; it is not proof of complete absence of ethanol from blood or the body. The blood–breath comparison analysis discusses the separate issue of interchangeability.
Several deficits occurred in the zero-reading group
In the zero-reading subgroup, mean divided-attention response time increased from 2.3 to 3.4 seconds after the drinking night. Mean driving speed increased from 29.1 to 33.7 mph. Time off road rose from 0.5% to 1.3%, and lane excursions increased from 9.7 to 20.2. These within-person contrasts were statistically significant.
Not every endpoint changed significantly. Lane-position variability did not show the same significant result, and incorrect or missed divided-attention responses and tickets were not significant within the zero-reading subgroup. Simulated collisions did not increase in that group.
Thus, the conclusion concerns selected aspects of performance, not a uniform worsening of every driving measure.
Residual alcohol did not produce an identical profile
The residual-alcohol group showed several similar next-day deficits. It also drove faster, exceeded the speed limit more often and received more simulated tickets than the zero-reading group.
Within the residual group, simulated accidents and collisions increased significantly from 1.1 to 2.5. That result should not be generalized to the zero-reading group, where the corresponding contrast was not significant and the mean moved in the opposite direction.
Because participants were not randomized to residual status, between-group differences cannot isolate the causal contribution of residual ethanol from drinking history, timing, sleep or other differences.
Feeling better did not reliably establish better driving
Participants reported greater workload and effort after drinking. Some correlations linked lower reported hangover or fatigue with faster driving, but significant associations were limited and differed between groups. The study did not validate a personal symptom threshold for safe driving.
Nor did it directly calibrate hangover impairment against an acute alcohol condition in these same participants. Comparisons with BAC effects from other studies should not be converted into a claim that a particular hangover is equivalent to a specific BAC.
Limits on a case-specific conclusion
Sleep was not recorded, some participants could not recall exact drinking amounts, and the sample consisted of relatively inexperienced drivers who regularly experienced hangovers. The results cannot separate all effects of sleep loss from other next-day consequences or establish causation in a specific crash.
The supported lesson is that alcohol detection and functional recovery are separate questions. For a different next-day inference, the site’s hangover severity and elimination-rate analysis explains why symptoms cannot supply an individual BAC reconstruction.
Primary source
Alford C and colleagues. The Impact of Alcohol Hangover on Simulated Driving Performance during a ‘Commute to Work’—Zero and Residual Alcohol Effects Compared. Journal of Clinical Medicine. 2020;9:1435. https://doi.org/10.3390/jcm9051435.