Driving drunk is a specific offence in almost every country, with a legislated threshold and a roadside test to enforce it. Driving drowsy is dangerous in much the same way, and in most jurisdictions it is not a specific offence at all.
The gap is not indifference. It is a measurement problem.
The short answer
Very few jurisdictions have a drowsiness-specific driving offence. Most prosecute drowsy driving under existing dangerous, careless, or negligent driving provisions, and only after a crash has occurred. There is no roadside test for drowsiness equivalent to a breathalyser, because unlike alcohol there is no substance to measure and no legislated threshold to measure it against.
Why alcohol is easy to legislate and drowsiness is not
A blood alcohol limit works as law because four things line up.
- There is a substance in the body that can be detected directly.
- Its concentration can be measured quickly, cheaply, and repeatably at the roadside.
- A threshold has been legislated: 0.05% or 0.08% depending on the jurisdiction, so the question at the roadside is binary.
- The measurement is defensible in court, with an established chain of evidence and instrument calibration standards.
Drowsiness satisfies none of these. There is no substance. There is no legislated threshold. And until recently there was no measurement that could be taken in the moment rather than reconstructed afterwards.
The result is that enforcement runs backwards. Rather than testing a driver and then establishing an offence, an investigator establishes a crash and then works back to whether drowsiness was involved, usually from the driver’s own account, phone records, work rosters, or the absence of any other explanation.
Reconstruction after the fact does not work, and the size of the failure is known
Official United States crash statistics attributed 2.4% of fatal crashes to a drowsy driver.3 Researchers at the AAA Foundation for Traffic Safety, examining the same body of crash records, put the figure at 16.5%.4 A later study using dash-cam footage from 3,593 drivers found drowsiness in 8.8% to 9.5% of all crashes, roughly eight times the federal estimate.5
Police-reported data undercounts drowsiness by a factor of around seven. This is not a criticism of investigators. It is a direct consequence of asking someone to establish, after a crash, a state that leaves no physical trace and that the only witness may not remember.
Australian figures are consistent: Fatigue is involved in approximately 16% of fatal crashes and 13% of serious injury crashes, rising to around 30% of fatal crashes on rural roads.6 Two independent datasets in two countries arriving at 16% is the strongest indication available that the real figure sits far above what enforcement records show.
Which jurisdictions have a drowsiness-specific law?
| Jurisdiction | What the law does | How it is triggered |
|---|---|---|
| New Jersey, USA | Maggie’s Law (2003) amended the state’s vehicular homicide statute so that driving without sleep can constitute recklessness | Proof that the driver fell asleep, or had been without sleep for more than 24 consecutive hours, may give rise to an inference of reckless driving1 |
| Most other jurisdictions | No drowsiness-specific offence. Drowsy driving is prosecuted under general dangerous, careless, or negligent driving provisions | After a crash, on the balance of the evidence available to investigators |
| Australia (heavy vehicles) | The Heavy Vehicle National Law regulates work and rest hours rather than drowsiness itself, with accreditation pathways for operators who manage fatigue risk differently | Compliance auditing and work diary records, not driver testing |
New Jersey is widely described as the only US state to define drowsy driving as recklessness within a vehicular homicide statute. The law was named for Maggie McDonnell, killed by a driver who had been awake for around 30 hours.
Its enforcement history illustrates the underlying problem rather than solving it. The burden falls on the prosecution to establish that the driver had gone more than 24 hours without sleep: a fact that generally exists only in the driver’s own recollection.
How heavy-vehicle regulation takes a different route
Where drowsiness cannot be measured, regulators have regulated the conditions that produce it instead. Australia’s Heavy Vehicle National Law is a clear example.
Under standard hours, a heavy vehicle driver may not work more than 12 hours in any 24-hour period or more than 144 hours in any 14 days, and must take a minimum of seven hours continuous rest in every 24-hour period.2 Operators who want more flexibility can seek accreditation to operate under alternative work and rest arrangements, on the basis that they demonstrate greater accountability for managing fatigue risk.
The Australian accreditation framework changed on 1 August 2026
Amendments to the Heavy Vehicle National Law commenced on 1 August 2026. Under the new Heavy Vehicle Accreditation scheme, Alternative Compliance Accreditation – Fatigue (ACA-Fatigue) replaces the previous Basic Fatigue Management and Advanced Fatigue Management modules for new applicants.7
Under ACA-Fatigue the regulator works with an operator to set Alternative Compliance Hours suited to their operation, which may be used in place of standard hours. Operators holding existing BFM or AFM accreditation may continue under its conditions until it expires, is cancelled, or they choose to transition.
The direction of the change is worth noting. The framework is moving further toward operators demonstrating how they manage fatigue risk, rather than simply complying with fixed hours, which raises the question of what evidence an operator can actually produce.
This approach is sound and it has a known limitation. Work and rest hours are a proxy. A compliant roster guarantees that a driver had the opportunity to sleep. It does not establish that they slept, or that the sleep they took was restorative. Two drivers on identical, fully compliant rosters can present entirely different levels of impairment.
That is the space an objective measure occupies: Not replacing the hours rules, but telling an operator which specific driver is impaired within them.
What would a roadside drowsiness test actually require?
This is the question that matters, and it has a concrete answer. To function as an enforcement instrument, a drowsiness measure would need four things.
- An objective physiological signal. Something measured from the body rather than reported by the driver. Self-assessment cannot serve, because the capacity to judge one’s own impairment degrades along with everything else.
- A defensible threshold. Legislators would need to set a level equivalent in function to 0.05% BAC: a point at which risk is deemed unacceptable. That is a policy decision, but it requires an underlying scale to set it on.
- Reproducibility without individual calibration. A test that requires a per-person baseline is unusable at the roadside. The measure has to mean the same thing for every driver.
- A legislated offence to attach to. Even a perfect measurement is inert without a provision making the measured state unlawful.
The first and third of these now exist. The Johns Drowsiness Scale (JDS™) scores impairment from drowsiness on a 0 to 10 scale, derived from eyelid movements, and requires no individual calibration: a score of 4.5 means the same thing for every driver. Risk of a performance failure rises substantially above that level.
The second and fourth are legislative choices, and they are the ones that have not been made.
Why enforcement is not the only reason this matters
Roadside testing may never arrive. There is a reasonable argument that it should not be the priority.
Alcohol enforcement works partly because it deters. A driver who knows they may be tested makes a different decision before setting out. Drowsiness does not work the same way: the driver who is dangerous is frequently the one least able to recognise it, so the deterrent effect of a roadside test would be weaker than for alcohol.
Where objective measurement changes outcomes today is inside operations, not at the roadside. An employer with real-time drowsiness data does not need a legislated threshold to act. They can intervene during the shift, before the driver reaches a public road at all.
Frequently asked questions
Is drowsy driving illegal?
In most jurisdictions there is no drowsiness-specific offence, but drowsy driving that causes a crash is routinely prosecuted under general dangerous, careless, or negligent driving provisions. New Jersey is the notable exception, having amended its vehicular homicide statute so that driving without sleep can constitute recklessness.
What is Maggie’s Law?
A New Jersey law passed in 2003, named for Maggie McDonnell. It amended the state’s vehicular homicide statute so that proof a driver fell asleep, or had been without sleep for more than 24 consecutive hours, may give rise to an inference of reckless driving.
Is there a breathalyser for drowsiness?
No. Alcohol enforcement works because there is a substance in the body, a fast roadside measurement, and a legislated threshold. Drowsiness has no substance to detect and no legislated threshold, so there is nothing for a roadside device to test against.
How do police prove a crash was caused by drowsiness?
Generally by inference after the fact: the driver’s own account, work rosters, phone or vehicle data, the absence of braking, and the elimination of other explanations. It is a reconstruction rather than a measurement, which is why drowsiness is widely believed to be under-recorded in crash statistics.
Can drowsiness be measured objectively at all?
Yes, in operational settings. The Johns Drowsiness Scale scores impairment from eyelid movements on a 0 to 10 scale without requiring individual calibration. What does not exist is a legislated threshold or an offence for such a measure to attach to.
Do work and rest hour rules prevent drowsy driving?
They reduce the conditions that produce it, which is meaningful. They cannot confirm that a compliant driver actually slept, or slept well. Two drivers on identical compliant rosters can carry very different levels of impairment.
Regulating a state nobody can currently test for
Drowsy driving law is where it is because legislators have been asked to prohibit something that could not be measured. The sensible response was to regulate the conditions instead (hours, rest, and accreditation) and to prosecute the consequences after the fact.
Objective measurement does not change the law. It does change what an operator can do without waiting for it.
Contact Optalert to discuss what real-time drowsiness measurement would show about the drivers in your operation.
References
- New Jersey Revised Statutes, N.J.S.A. 2C:11-5 (vehicular homicide), as amended 2003 by the legislation known as Maggie’s Law.
- National Heavy Vehicle Regulator, work and rest requirements, Heavy Vehicle National Law.
- National Highway Traffic Safety Administration, Traffic Safety Facts, 2008. 2.4% of fatal crashes involving a drowsy driver.
- AAA Foundation for Traffic Safety. Asleep at the Wheel: The Prevalence and Impact of Drowsy Driving. 2010, analysing NHTSA crash data 1999–2008. aaafoundation.org
- AAA Foundation for Traffic Safety. Prevalence of Drowsy Driving Crashes: Estimates from a Large-Scale Naturalistic Driving Study. 2018. aaafoundation.org
- Austroads. Guide to Road Safety Part 5: Road Safety for Rural and Remote Areas.
- National Heavy Vehicle Regulator. Alternative Compliance Accreditation – Fatigue (ACA-Fatigue). Available from 1 August 2026 under the amended Heavy Vehicle National Law. nhvr.gov.au
