// Workers AI · dad joke modeWhy was the driver under arrest? He was high on the list.

"DUID" redirects here. For the DHCPv6 identifier, see DHCPv6 § DHCP unique identifier.

Roadside sign in Victoria, Australia, warning drivers about mobile drug testing

Drug-impaired driving, also called driving under the influence of drugs (DUID) or drug driving, is the operation of a motor vehicle while a psychoactive drug other than - or in addition to - alcohol reduces the driver's ability to drive safely. The conduct is a criminal or administrative offense in many countries.

Drugs implicated include illicit substances, prescribed and over-the-counter medicines, and combinations of those substances with alcohol. Laboratory and epidemiological studies have documented effects on attention, tracking, reaction time and risk-taking for several drug classes. The relationship between a measured concentration in blood or oral fluid and crash risk is much clearer for alcohol than for most other drugs.

The World Health Organization treats driving under the influence of alcohol or other psychoactive substances as a major crash risk factor. It notes that the increase in crash risk varies by substance; amphetamines, for example, have been associated with about a fivefold rise in fatal-crash risk relative to non-use. Roadside surveys measure how often drugs are present in drivers, not whether those drivers are impaired. In the 2013–2014 U.S. National Roadside Survey, about one in five sampled drivers tested positive for at least one drug; the European DRUID project found lower average rates in general traffic, with large regional differences and with alcohol still the substance most often found in seriously injured and killed drivers.

Legal systems use more than one method of proof. An impairment statute requires evidence that the drug affected driving. Zero tolerance and numeric per se statutes treat a laboratory result as the offense. Some jurisdictions allow a laboratory result to support an inference of impairment without making the number conclusive. Testing and reporting practices vary widely, which limits comparison of arrest and fatality statistics across places and years.

History

Impaired-driving statutes began as alcohol offenses. Early 20th-century codes typically required proof that the driver was “intoxicated” or “under the influence,” judged from observed behavior. Chemical testing shifted that focus. Indiana and Maine adopted presumptive blood alcohol content levels in 1939; the Uniform Vehicle Code later incorporated similar chemical-test provisions. New York enacted the first U.S. implied consent law in 1952. Robert Borkenstein’s Breathalyzer (1954) made roadside alcohol measurement practical, and by 1973 every U.S. state had an implied-consent statute.

Drug-impaired driving was slower to acquire its own tools. Officers routinely encountered drivers who appeared impaired but registered little or no alcohol. In the early 1970s two Los Angeles Police Department sergeants, working with physicians and psychologists, developed a standardized evaluation that became the Drug Evaluation and Classification (DEC) protocol. The department formally recognized a Drug Recognition Expert (DRE) program in 1979. The National Highway Traffic Safety Administration funded multi-state pilots in 1987–1988; the International Association of Chiefs of Police later coordinated national and international expansion.

From the 1990s onward, jurisdictions added zero-tolerance or numeric drug rules on top of impairment offenses, first in parts of Europe and then more widely after oral-fluid screening became commercially available. Victoria, Australia, introduced random roadside oral-fluid testing in December 2004, modeled on random breath testing and initially targeting methamphetamine, MDMA and THC. Sweden adopted a scheduled-drug zero-tolerance rule in 1999; Norway set graded blood limits for twenty drugs in 2012, calibrated to degrees of alcohol impairment. Canada added regulated blood-drug concentrations in 2018. England and Wales added specified-drug concentration offenses in 2015. Germany replaced a near-zero judicial THC practice with a 3.5 ng/mL serum limit in 2024 after recreational cannabis reform.

Effects by drug class

NHTSA's DEC program groups impairing drugs into seven categories used by Drug Recognition Experts: central nervous system (CNS) depressants, CNS stimulants, hallucinogens, dissociative anesthetics, narcotic analgesics, inhalants, and cannabis. Experimental and epidemiological work does not treat those categories as equal in crash risk.

Alcohol remains the benchmark. Crash risk rises in a reproducible way as blood alcohol content rises. In NHTSA's Virginia Beach case–control study, drivers at 0.08% BAC were about four times as likely to crash as sober drivers, and about twelve times as likely at 0.15%.

CNS depressants other than alcohol include benzodiazepines and Z-drugs such as zolpidem and zopiclone. They impair divided attention and increase lane-position variability in experimental studies. In DRUID roadside surveys they were among the medicines most often detected in northern Europe, typically in older daytime drivers. Benzodiazepines have also been among the most frequently detected medicines in some crash series when laboratories test for them.

Opioids, whether illicit or prescribed (the DEC "narcotic analgesic" category), can slow reaction time and produce drowsiness. Medicinal opioids appeared more often in northern European roadside samples than illicit opiates did. In the U.S. National Roadside Survey, prescription and over-the-counter medications that can affect driving were found in 7.3% of weekend nighttime drivers - a separate group from illegal-drug positives.

Stimulants such as cocaine and amphetamines can improve some laboratory measures of alertness at low doses and degrade judgment and impulse control; fatigue can follow the acute phase. DRUID associated amphetamines with elevated injury risk, especially in combination with other substances. Cocaine was the second most common illicit drug in several southern European roadside samples after cannabis.

Cannabis / delta-9-THC. A 2022 meta-analysis of 57 experimental driving studies found that cannabis impaired lateral control (lane weaving and excursions) and was associated with lower driving speed; combined cannabis and alcohol produced larger decrements than either substance alone, and cannabis effects were described as similar in magnitude to low blood-alcohol concentrations. A related meta-analytic review of acute Δ9-THC effects estimated that most driving-related cognitive skills recovered within about five hours after inhaling 20 mg of THC in occasional users, with oral doses taking longer; regular consumers showed smaller average deficits. Blood THC falls quickly after inhalation while effects may continue, and regular consumers can have residual blood THC after acute effects have ended. Systematic reviews have found no reliable linear relationship between a single blood-THC value and driving performance. Observational syntheses report low-certainty associations between cannabis use and crash fatality or injury; they do not resolve dose or timing.

Hallucinogens, dissociative anesthetics (phencyclidine, ketamine) and inhalants are less common in roadside samples but can produce marked perceptual and motor disturbance. Several countries set "any detectable level" limits for LSD, PCP, ketamine or heroin's 6-monoacetylmorphine rather than a numeric impairment threshold.

Combinations of alcohol with another drug, or of two drugs, produced some of the highest odds ratios for serious injury or death in DRUID's case–control analyses - higher than most single-drug findings except high BAC.

Epidemiology

Presence is not impairment

A positive toxicology test shows that an analyte was present above a laboratory cut-off in the specimen drawn. It does not by itself establish that the person was impaired at the time of driving, or that the drug caused a crash. NHTSA states that distinction in its roadside-survey and crash-risk reports.

Roadside surveys

The 2013–2014 National Roadside Survey sampled volunteer drivers at 300 locations in the continental United States. Using oral fluid and blood, about 22% of both daytime and nighttime drivers tested positive for at least one of 98 drugs (illegal, prescription or over-the-counter). Delta-9-THC was the most frequent drug (8.7% of daytime and 12.7% of nighttime drivers). Among weekend nighttime drivers, 15.2% were positive for an illegal drug and 7.3% for a prescription or over-the-counter medication that can affect driving. On a panel comparable to 2007, nighttime drug prevalence rose from 16.3% to 20.1%. Over the same decades, weekend-night drivers at or above 0.08% BrAC fell from 7.5% in 1973 to 1.5% in 2013–2014.

The European DRUID project sampled about 50,000 drivers in 13 countries. Estimated prevalence in general traffic was 3.48% for alcohol (≥0.1 g/L), 1.90% for illicit drugs, 1.36% for selected medicines, 0.39% for drug–drug combinations and 0.37% for alcohol plus a drug or medicine. THC was the most common illicit drug in traffic overall. Benzodiazepines and medicinal opioids were more characteristic of northern Europe and of older female daytime drivers. Southern and western Europe had higher roadside rates of alcohol, cocaine and cannabis. Eastern European samples were lower on average.

Comparable random-sample figures are scarce for Latin America, Africa and much of Asia. Enforcement in those regions more often relies on observed impairment, crash investigation, or alcohol screening; oral-fluid programs are uncommon outside high-income jurisdictions.

Crash-involved drivers

Among seriously injured or killed drivers in DRUID hospital and fatality studies, alcohol alone was the substance most often detected, followed by alcohol combined with another substance. Combined use was far more common in crash-involved drivers than in roadside traffic.

NHTSA's Virginia Beach case–control study compared more than 3,000 crash-involved drivers with 6,000 controls matched on location, time and direction of travel. Unadjusted odds of crash involvement were 1.21 for drivers positive for illegal drugs and 1.25 for THC. After adjustment for age, sex, ethnicity and alcohol, the drug associations were no longer statistically significant; alcohol's association remained large. The study was conducted in one metropolitan area, most crashes were non-fatal, and the drug measure was presence rather than dose or documented impairment.

Charged-driver and fatal-crash files

Arrest files are not a sample of all drivers. Where states publish case-processing reports, the share of filings that include a drug test, and the share that are alcohol-only versus mixed, vary with laboratory capacity. Conviction patterns in a single state cannot be treated as national effects of a particular statutory wording.

The U.S. Fatality Analysis Reporting System (FARS) is a census of crashes with a death within 30 days. Its alcohol fields are widely used for national estimates. NHTSA has warned that FARS drug fields are not comparable: who is tested varies by state and by whether the driver died; laboratories use different matrices, panels and cut-offs; until a 2018 redesign the file stored at most three non-alcohol drugs under a hierarchy that under-counted combinations; missing drug results are not missing at random and are not imputed as missing BACs are; and a recorded positive test is presence, not impairment or causation. Year-to-year changes in "drug-positive fatal-crash drivers" therefore mix changes in use with changes in testing and reporting.

Detection and enforcement

Roadside investigation

Alcohol can be measured at the roadside with a preliminary or evidential breath test. No equivalent, court-ready device exists for the full range of other drugs. A typical investigation in countries with chemical-test statutes uses observation of driving and of the driver; psychomotor tests such as standardized field sobriety tests (SFSTs); a breath test to measure or exclude alcohol; in some places a roadside oral-fluid screen; and a blood test, urine or oral-fluid draw for immunoassay screening and confirmatory mass spectrometry.

SFSTs were validated primarily as an alcohol battery. Agencies also use them when drugs are suspected. Roadside oral-fluid devices are screening tools; a screen is not a quantitative blood result. Blood delta-9-THC can fall substantially in the first one to two hours after inhalation, so the interval between the stop and the draw affects numeric THC statutes more than it affects alcohol per se cases.

Implied consent and chemical tests

U.S. implied-consent statutes treat a driver's license as conditioned on submitting to a chemical test after a lawful impaired-driving arrest. Civil license suspension for refusal is widespread. In Birchfield v. North Dakota (2016) the Supreme Court of the United States held that the Fourth Amendment allows a warrantless breath test as a search incident to a lawful arrest, but not a warrantless blood draw, and that a state may not impose a criminal penalty for refusing a warrantless blood test on an implied-consent theory. Blood remains the usual confirmatory matrix for drugs other than alcohol; agencies typically seek a warrant, rely on a recognized exigency, or use a statute that imposes only civil consequences for refusal. Other common-law jurisdictions use analogous “fail to provide” offenses rather than the U.S. implied-consent label.

Oral-fluid programs

As of a 2025 National Conference of State Legislatures summary, 24 U.S. states authorized some use of oral fluid in DUI cases, but most did not collect it in routine practice. Alabama and Indiana were described as having active roadside screening programs; Michigan has operated DRE-administered pilots under statute. Alabama's state laboratory program uses roadside screens for cannabis, cocaine, methamphetamine, amphetamine, opioids and benzodiazepines, followed by evidentiary oral-fluid confirmation.

Australia operates the most extensive random roadside oral-fluid regime. Victoria began testing in December 2004. A 2025 federal Office of Road Safety summary stated that every state and territory uses roadside saliva screening, with laboratory confirmation in all jurisdictions except the Northern Territory, and that THC, MDMA and methamphetamine are screened nationwide. Transport for NSW describes mobile drug testing for ecstasy, cannabis, cocaine and methamphetamine and states that a confirmed presence of those prescribed illicit drugs is an offense separate from driving under the influence of any drug.

Drug Recognition Experts

The DEC program originated in the Los Angeles Police Department in the 1970s. A certified DRE follows a 12-step protocol: breath alcohol; interview of the arresting officer; preliminary examination; eye examinations; divided-attention psychomotor tests; vital signs; dark-room eye examination; muscle tone; injection-site inspection; statements of the suspect; the DRE's opinion of the drug category; and a toxicology request. Field studies have reported that DREs often name a category later confirmed by toxicology. Those studies generally start with drivers already suspected of impairment and use laboratory presence, not a measured driving deficit, as the reference standard.

Commercial and occupational driving

Occupational rules are often stricter than general traffic codes. In the United States, Federal Motor Carrier Safety Administration regulations prohibit commercial driver's license holders from using Schedule I substances and from using other controlled substances unless a licensed practitioner who knows the driver's history has advised that the drug will not affect safe operation. DOT testing covers marijuana, cocaine, opioids, amphetamines and PCP, with pre-employment, random, reasonable-suspicion, post-accident, return-to-duty and follow-up tests. Many countries impose lower or zero alcohol limits on professional drivers; some also apply lower drug thresholds or more frequent testing to bus, taxi and heavy-vehicle operators.

Legal approaches

Statutes generally fall into four groups, which jurisdictions often combine:

• Impairment / under the influence. The prosecution must prove that a drug, or alcohol plus a drug, affected the ability to drive. Some codes require that the person be "incapable of safely operating" a vehicle; others require impairment "to the slightest degree" or that driving be "less safe."
• Zero tolerance. Driving with any measurable amount of a listed drug - or, in some codes, an inactive metabolite - is the offense.
• Numeric per se. Driving with a concentration at or above a listed value is the offense. Alcohol's 0.08% BAC (0.05% in Utah) is the familiar example. A 2023 review of illicit-drug limits in 19 jurisdictions found large variation and limited harmonization; Canada, Denmark, New Zealand and Norway also use graded sanction tiers.
• Permissible inference. A stated concentration allows a fact-finder to infer impairment. Colorado uses this device for 5 ng/mL or more of delta-9-THC in whole blood on a DUI charge.

Delta-9-THC is the principal intoxicating cannabinoid named in most numeric blood limits. 11-Hydroxy-THC is an active metabolite; THC-COOH is an inactive metabolite used as a marker of past use. A metabolite-only urine result can reflect use outside an acute-impairment window.

United States

Scope of state law

Every U.S. state and the District of Columbia prohibits driving under the influence of drugs. Older "DUID law" counts usually referred to zero-tolerance or numeric add-on statutes, not to the existence of an impairment offense.

GHSA, using state highway-safety office reviews current through June 2025, reported 16 states with a zero-tolerance law for one or more drugs and five with a numeric per se law for one or more drugs, and about 18 states with either form of add-on rule for cannabis. LAPPA's August 2025 survey groups some jurisdictions differently: it folds Colorado's inference in with numeric THC limits and lists additional states on the cannabis rows. The current state code controls.

LAPPA's August 2025 survey described cannabis add-on rules as follows. Other compilations differ on edge cases.

Cannabis DUID add-on rules according to LAPPA (August 2025)

Rule type in that survey	What the add-on typically does	States listed by LAPPA (2025)
Zero tolerance for THC and/or a metabolite	Any measurable amount of delta-9-THC, "marijuana," a Schedule I substance, or in some codes an inactive metabolite, is itself unlawful. Some of these states have medical-use exceptions.	Arizona, Delaware, Georgia, Indiana, Iowa, Michigan, Oklahoma, Pennsylvania, South Dakota, Utah; Wisconsin
Numeric or hybrid numeric rules (LAPPA's grouping)	A stated ng/mL value is itself unlawful, or is used with other facts. LAPPA includes Colorado's 5 ng/mL inference in this group; GHSA lists Colorado separately as inference-only.	Illinois (5 ng/mL delta-9-THC in whole blood; 10 ng/mL in other bodily substances); Montana (5 ng/mL); Nevada (2 ng/mL THC and 5 ng/mL metabolite in blood; felony DUID context); Ohio (2 ng/mL "marijuana" in whole blood and higher urine and metabolite figures); Washington (5 ng/mL delta-9-THC in blood, generally within two hours of driving); Colorado (5 ng/mL inference on DUI)
Impairment only for cannabis	Presence of THC is evidence. The prosecution must still prove actual impairment.	Remaining states in that survey, including California, New York, Massachusetts and Oregon

Charging practice and offense wording

Whether statutory wording changes conviction rates is not established nationally. A 2010 NHTSA review of drug per se laws found that many states did not separate alcohol and drug arrests, that conviction files often omitted toxicology, and that prosecutors in mixed cases often charged only the alcohol count. The review concluded that available data were insufficient to say whether those laws increase arrests or convictions.

Colorado is the U.S. jurisdiction whose charging files have been analyzed in most detail on this point. State law creates two impairment misdemeanors with different elements. Driving under the influence (DUI) requires that alcohol, a drug, or both affected the person to a degree that the person is "substantially incapable" of safely operating a vehicle. Driving while ability impaired (DWAI) requires that the same substances affected the person "to the slightest degree" so that the person is less able than usual to drive safely. A laboratory finding of 5 ng/mL or more of delta-9-THC in whole blood creates a permissible inference of DUI only; it does not apply to DWAI. Comprehensive drug testing covered 37% of 2020 filings in the state's published case-processing reports.

Edward C. Wood analyzed those official tables in a 2024 peer-reviewed paper. Wood is president of the advocacy organization DUID Victim Voices; the study uses the Colorado Division of Criminal Justice reports rather than a new sample of drivers. He estimated that about half of 2020 filings were alcohol-only and half were drug-related, and that combined (polydrug) use was the largest drug-related subset. Overall conviction rates in that year were 91% for alcohol-only cases, 90% for polydrug cases and 72% for THC-only cases. Across three years of tables, the average DUI conviction rate for THC-only defendants at or above 5 ng/mL was 65%, compared with 92% for alcohol-only defendants at or above 0.08% BAC. THC-only defendants below 5 ng/mL were convicted of DUI in 9% of cases on that three-year average, while DWAI conviction rates for THC-only defendants remained about 99% both above and below 5 ng/mL. The same files show DWAI used as a plea disposition in many alcohol cases, so a DWAI conviction does not by itself mean the driver was only slightly impaired.

Wood concluded that, in those Colorado files, a non-zero THC inference together with defining DUI as being "incapable of safe driving" was associated with fewer convictions on the higher DUI charge, especially below 5 ng/mL, even when DWAI convictions remained high. That is an analysis of one state's charging practice. It is not a national evaluation of per se or inference statutes, and NHTSA's 2010 multi-state review did not find sufficient data to measure those laws' effect on arrests or convictions.

Other countries

Canada

Since December 2018, drug-impaired operation has been charged under Criminal Code section 320.14. The section makes it an offense to operate a conveyance while ability is impaired to any degree by alcohol, a drug, or a combination; to have a blood alcohol concentration of 80 mg/100 mL or more within two hours of driving; to have a blood drug concentration at or above a level set by regulation within two hours of driving; or to have a regulated combination of alcohol and a drug. See also impaired driving in Canada.

The Blood Drug Concentration Regulations create two THC tiers: 2 ng/mL but less than 5 ng/mL within two hours of driving is a summary offense; 5 ng/mL or more is the hybrid offense. A combination of 50 mg/100 mL alcohol plus 2.5 ng/mL THC is a separate hybrid offense. LSD, psilocybin, psilocin, PCP, ketamine, cocaine, methamphetamine and 6-monoacetylmorphine are prohibited at any detectable level; GHB has a 5 mg/L limit. Police may demand oral-fluid screening and, on prescribed grounds, blood. Refusal is an offense. Provinces add license suspensions.

European Union and Nordic countries

Member states of the European Union use impairment offenses, low administrative thresholds for illicit drugs, and, in some countries, numeric limits. DRUID recommended that policy continue to treat high-BAC alcohol and alcohol–drug combinations as the highest-risk patterns. It did not create a single EU offense.

Nordic practice is among the strictest. Sweden and Finland treat any quantifiable amount of a scheduled drug in blood as unlawful, with a prescription defense when the medicine was taken as directed; a supratherapeutic concentration can still support a charge. Denmark and Norway use threshold limits; Norway's 2012 scheme set impairment limits for twenty drugs intended to correspond to 0.02% BAC, with higher graded-sanction tiers for thirteen of them. After the stricter Nordic drug-driving rules, the number of suspects apprehended rose sharply in some of those countries.

Germany in 2024 set a 3.5 ng/mL THC limit in blood serum as an administrative offense, with a zero limit for novice drivers and drivers under 21 and a medical-use privilege when the driver is not impaired. Combining cannabis with alcohol attracts a higher fine.

United Kingdom

From March 2, 2015, England and Wales added specified-drug concentration offenses under section 5A of the Road Traffic Act 1988 on top of the existing impairment offense. Regulations set blood limits in micrograms per litre. The Department for Transport described eight of the limits as a near-zero-tolerance approach to drugs associated with illicit use (including 2 µg/L for delta-9-THC and 10 µg/L for cocaine) and eight as higher, risk-based limits for drugs associated with medical use (for example 80 µg/L morphine and 550 µg/L diazepam); amphetamine has a separate 250 µg/L limit. A statutory medical defense can apply when a specified drug was lawfully prescribed and taken as directed and the driver was not impaired.

Ireland

Ireland prohibits driving while ability is impaired by a drug. The Road Safety Authority has conducted roadside impairment testing of suspected drug-impaired drivers since 2014.

Australia

Every Australian state and territory operates roadside oral-fluid screening. Victoria introduced the first program in December 2004. The core roadside panel is THC, methamphetamine and MDMA; some jurisdictions also screen for cocaine. A laboratory-confirmed presence of a prescribed illicit drug is typically an offense aimed at recent use rather than at a validated blood-impairment curve. A separate impairment offense covers any drug, including medicines. Public-health advice in some states tells cannabis consumers not to drive for a stated number of hours after use. A prescription for medical cannabis is generally not a defense to a presence offense.

Hong Kong and New Zealand

Hong Kong prohibits driving while incapable of proper control because of a drug and, separately, driving with any concentration of listed illicit drugs. Police may require an impairment test at a station. New Zealand's Land Transport Act 1998 prohibits driving while impaired with evidence of a qualifying drug in blood; qualifying drugs include specified Class A, B and C controlled drugs under the Misuse of Drugs Act 1975 and some medicines. New Zealand is one of the jurisdictions that also uses graded sanction limits for some drugs.

Latin America, Africa and Asia

Comparative roadside-drug data are scarce. Most middle-income jurisdictions prohibit driving while under the influence of a narcotic or intoxicating drug as an impairment offense, often without a published numeric THC line and without random oral-fluid programs.

Brazil's Código de Trânsito Brasileiro makes it an administrative infraction to drive under the influence of alcohol or any other psychoactive substance that causes dependence (art. 165) and a crime to drive with psychomotor capacity altered by those substances (art. 306). Article 277 authorizes tests, clinical examination or other technical means to establish influence of alcohol or another such substance; refusal is separately penalized. Japan's Road Traffic Act prohibits driving in a state that makes normal driving unlikely because of narcotics, cannabis, opium, stimulants or specified poisons (art. 66) and punishes that conduct with imprisonment of up to five years or a fine of up to one million yen (art. 117-2). South Africa's National Road Traffic Act 93 of 1996, section 65(1), prohibits driving or occupying the driver's seat of a running motor vehicle while under the influence of intoxicating liquor or a drug having a narcotic effect. Numeric alcohol limits are set in the same section; later amendments that would add explicit THC concentrations remain uncommenced pending proclamation. India and China rely primarily on alcohol limits (0.03% and 0.02% BAC respectively for ordinary drivers) and on general impairment or dangerous-driving provisions for other drugs; systematic public drug-driving statistics are limited.

Medical and prescribed use

Prescribing information for many CNS-active medicines warns patients not to drive until they know how the drug affects them. Some countries mark packages of impairing medicines with a warning symbol. Nordic zero-tolerance rules typically exempt a drug taken in accordance with a valid prescription unless the concentration is supratherapeutic or the driver is observably unfit.

Cannabis is the sharpest conflict. In Australia, a lawful medicinal-cannabis prescription is generally not a defense to a roadside presence offense; patients using THC products can test positive hours after last use, and guidance is to avoid driving after THC and to prefer CBD-only products where clinically appropriate. England and Wales and Germany provide a statutory medical defense or privilege when the product was prescribed, taken as directed, and the driver is not impaired. Several U.S. medical-cannabis states keep impairment-only DUID rules; others apply zero-tolerance or numeric THC limits with uneven patient exceptions. Residual blood THC in daily patients is the practical problem those numeric lines create.

Prevention and effectiveness

Documented responses include high-visibility enforcement, Advanced Roadside Impaired Driving Enforcement (ARIDE) training for officers who are not DREs, public-information campaigns, license suspension, drug courts, and, mainly for alcohol, ignition interlocks.

A 2021 systematic review of interventions against “drugged driving” found high-certainty evidence that cannabis packaging with health warnings increases knowledge; moderate-certainty evidence that roadside drug testing can reduce driving after cannabis among users and that motivational interviewing can reduce the behavior in youth or prior offenders; and low or very-low certainty that media campaigns, criminalization, license withdrawal or per se drug laws change fatalities or injuries. Alcohol ignition interlocks reduce drink-driving recidivism while installed; residual effects after removal are smaller and program participation is often low. Interlocks do not detect other drugs. Pharmacy counseling and package warnings are widely recommended; their isolated effect on crash rates is not well measured.

Equity and enforcement

Drug-driving enforcement is filtered through ordinary traffic stop practice. A large analysis of U.S. police stops found higher search rates for Black and Hispanic drivers and a lower contraband “hit rate,” a pattern consistent with a lower search threshold; recreational-marijuana legalization reduced drug-related search volume but did not eliminate the gap. That study measured stops and searches, not DUID charging or conviction outcomes. Comparable published analyses specific to drug-impaired driving charges outside the United States are limited. Presence-based oral-fluid regimes can also burden medical-cannabis patients and people who used days earlier, independent of current deficit.

Debate over numeric drug limits

Highway-safety agencies and prosecutors have supported zero-tolerance or numeric add-on statutes as a way to prosecute cases in which drug effects may fade before a blood draw and in which a jury may give less weight to a DRE evaluation than to a BAC printout. GHSA's impaired-driving policy endorses zero tolerance for illicit drugs and for under-21 drivers, and it states that available research has not identified a scientifically sound per se limit for cannabis.

Toxicologists and experimental researchers have argued that a THC nanogram line does not perform the function BAC performs for alcohol. The AAA Foundation's 2016 analysis of cannabis-positive DUI arrestees found that candidate thresholds from 1 to 10 ng/mL misclassified large numbers of drivers relative to SFST performance; the Foundation concluded that a quantitative THC per se threshold "cannot be scientifically supported." Later reviews reached the same concentration-versus-impairment point: blood THC is a weak correlate of deficit, especially in regular consumers, even though experimental work shows that cannabis does impair lateral control.

Canada, Norway, Germany and several Australian states adopted numeric or roadside-screen regimes after reviewing that literature, treating a recent-use marker or an administrable line as a workable rule rather than as a precise measure of deficit. U.S. states remain divided between impairment-only cannabis statutes and add-on numeric or zero-tolerance rules.

See also

• Driving under the influence
• Drunk driving in the United States
• Impaired driving in Canada
• Cannabis and impaired driving
• Cannabis drug testing
• Drug Recognition Expert
• Field sobriety testing
• Implied consent
• Birchfield v. North Dakota
• Breathalyzer
• Blood alcohol content
• Fatality Analysis Reporting System
• Ignition interlock device
• Road traffic safety

External links

• Governors Highway Safety Association, Drug-Impaired Driving
• Legislative Analysis and Public Policy Association, Drugged Driving: Summary of State Laws
• NHTSA, Drug and Alcohol Crash Risk Study
• EUDA / EMCDDA summary of the DRUID project
• WHO fact sheet, Road traffic injuries