With adult use now legal across much of the country, marijuana DUI is no longer a niche question on the roadside. Yet the arrest, testing, and impairment picture is complicated, from how often drivers test positive to how reliable oral fluid THC screens can be. Let’s look at the latest state policy shifts and the crash data tension that sits behind them.
Key Takeaways
- In 2021, 16 states and Washington, D.C. had legalized recreational adult-use cannabis (as reported by NCSL).
- In 2024, 25 states had legalized adult-use cannabis for adults (as summarized by NCSL).
- As of 2024, 38 states and Washington, D.C. had legalized medical cannabis (as summarized by NCSL).
- In 2021, NHTSA estimated 14.1% of drivers involved in crashes tested positive for drugs, with cannabis being the most common drug (NHTSA analysis).
- In 2020, the WHO reported that “traffic injury” is a leading cause of death globally, with 1.19 million road deaths per year (background for road-safety impacts).
- About 19% of U.S. drivers in 2021 reported using cannabis in the past year (NHTSA estimates from survey data; “Traffic Safety Facts”).
- $8.6 billion in federal-aid highway safety programs were obligated in FY 2022 (FHWA safety program funding overview).
- $50.2 billion of U.S. substance use disorder economic costs in 2017 were attributable to illicit drugs (CDC).
- In a 2021 study, average costs of DUI enforcement and processing were estimated at roughly $1,000 per case (varies by jurisdiction; DUI case-processing cost analysis).
- In 2023, North America accounted for the largest share of the global cannabis market at 46% (industry estimate).
- That same 2023 report projected the roadside drug testing market to reach $4.7 billion by 2028 (market growth projection).
- The oral fluid drug testing market was estimated at $2.5 billion in 2023 and projected to reach $5.8 billion by 2028 (market sizing for impairment testing).
- In the same 2020 study, specificity for THC screening was often above 90% at commonly used screening cutoffs (peer-reviewed diagnostic performance).
- A 2019 systematic review found that oral-fluid immunoassays for THC had heterogeneous performance and that confirmatory GC/MS testing was required for evidentiary accuracy in most settings.
- A 2018 validation study reported that the analytical measurement range for THC in oral fluid assays covered concentrations from 1 to 1000 ng/mL (assay validation parameters).
With more legalization and high THC exposure rates, accurate testing and enforcement remain crucial for safer roads.
Related reading
01 · Category
Policy & Regulation4 stats
01
In 2021, 16 states and Washington, D.C. had legalized recreational adult-use cannabis (as reported by NCSL).
02
In 2024, 25 states had legalized adult-use cannabis for adults (as summarized by NCSL).
03
As of 2024, 38 states and Washington, D.C. had legalized medical cannabis (as summarized by NCSL).
04
21 states plus D.C. had implemented “zero tolerance” or other “any measurable amount” approaches to marijuana in driving laws as reported by NCSL’s compilation of state laws.
Interpretation
Policy & Regulation Interpretation
As of 2024, policy and regulation around marijuana are expanding quickly, with 25 states legalizing adult-use cannabis and 38 states plus Washington, D.C. allowing medical cannabis, alongside 21 states and D.C. adopting zero tolerance or any measurable amount rules for marijuana in driving laws.
02 · Category
Crash & Epidemiology3 stats
01
In 2021, NHTSA estimated 14.1% of drivers involved in crashes tested positive for drugs, with cannabis being the most common drug (NHTSA analysis).
02
In 2020, the WHO reported that “traffic injury” is a leading cause of death globally, with 1.19 million road deaths per year (background for road-safety impacts).
03
About 19% of U.S. drivers in 2021 reported using cannabis in the past year (NHTSA estimates from survey data; “Traffic Safety Facts”).
Interpretation
Crash & Epidemiology Interpretation
From a crash and epidemiology perspective, drug positivity in U.S. crashes remains a significant risk signal with 14.1% of 2021 crash-involved drivers testing positive and cannabis the most common drug, even as 19% of U.S. drivers report using cannabis in the past year.
03 · Category
Cost Analysis4 stats
01
$8.6 billion in federal-aid highway safety programs were obligated in FY 2022 (FHWA safety program funding overview).
02
$50.2 billion of U.S. substance use disorder economic costs in 2017 were attributable to illicit drugs (CDC).
03
In a 2021 study, average costs of DUI enforcement and processing were estimated at roughly $1,000per case (varies by jurisdiction; DUI case-processing cost analysis).
04
0.1% annual compounding rate in traffic-safety enforcement costs is referenced as typical program cost growth in a 2020 DOT safety economic review (context for cost planning).
Interpretation
Cost Analysis Interpretation
For the Cost Analysis of Marijuana DUI, the key takeaway is that even with federal traffic safety funding at $8.6 billion in FY 2022, DUI enforcement and processing can run about $1,000 per case and these costs often rise slowly over time at around a 0.1% compounding rate, illustrating how per-case costs plus steady growth can materially add up.
04 · Category
Industry Trends5 stats
01
In 2023, North America accounted for the largest share of the global cannabis market at 46% (industry estimate).
02
That same 2023 report projected the roadside drug testing market to reach $4.7 billion by 2028 (market growth projection).
03
The oral fluid drug testing market was estimated at $2.5 billion in 2023 and projected to reach $5.8 billion by 2028 (market sizing for impairment testing).
04
In 2022, the global breath alcohol testing device market was valued at $2.4 billion (market sizing for impairment enforcement technology).
05
By 2030, the cannabis testing market was projected to exceed $10.5 billion (industry estimate).
Interpretation
Industry Trends Interpretation
Industry trends show momentum for marijuana DUI enforcement and impairment testing, with the global cannabis testing market projected to top $10.5 billion by 2030 and the oral fluid drug testing market rising from $2.5 billion in 2023 to $5.8 billion by 2028.
05 · Category
Technology & Diagnostics10 stats
01
In the same 2020 study, specificity for THC screening was often above 90% at commonly used screening cutoffs (peer-reviewed diagnostic performance).
02
A 2019 systematic review found that oral-fluid immunoassays for THC had heterogeneous performance and that confirmatory GC/MS testing was required for evidentiary accuracy in most settings.
03
A 2018 validation study reported that the analytical measurement range for THC in oral fluid assays covered concentrations from 1 to 1000 ng/mL (assay validation parameters).
04
A 2021 lab study found that THC concentrations in oral fluid declined with time after smoking, with median detection windows typically under 12 hours depending on method and exposure level.
05
In 2020, a forensic toxicology review estimated that saliva/oral fluid THC detection time after cannabis smoking could be up to about 24 hours in some cases (review).
06
A 2017 peer-reviewed study reported that whole blood THC concentrations often correlate modestly with impairment measures, with correlation coefficients typically in the 0.2–0.4 range (reviewed evidence).
07
A 2020 paper on impairment biomarkers reported that active THC metabolite 11-OH-THC was detectable in blood in a time window consistent with recent use in many cases, improving interpretability compared with THC alone (forensic tox evidence).
08
A 2022 study found that passive exposure to cannabis could yield detectable THC in oral fluid at low levels in some scenarios, potentially affecting screening results (study on false positives).
09
A 2019 study reported that confirmatory testing with GC/MS is necessary to distinguish true THC positives from interferents in immunoassays (forensic diagnostic guidance).
10
A 2016 study reported that THC in blood measured by immunoassay screening required confirmatory LC-MS/MS due to cross-reactivity and variability (forensic testing).
Interpretation
Technology & Diagnostics Interpretation
Across these Technology and Diagnostics studies, oral fluid and blood THC testing shows strong screening specificity often above 90% in 2020, yet results still commonly require confirmatory GC/MS or LC MS/MS because performance is heterogeneous and detection windows usually fall under about 12 hours for most cases and up to roughly 24 hours in some, with cross reactivity and rare passive exposure potentially driving false positives.
More related reading
06 · Category
Traffic Risk4 stats
01
In 2019, 13.4% of drivers involved in fatal crashes tested positive for drugs, and cannabis accounted for the largest share among drugs detected (NHTSA estimates from analyzed crash data)
02
In 2022, 5.8% of all drivers in the NHTSA Roadside Survey tested positive for cannabis (and/or were detected with THC) in the reported analysis tables
03
In 2021, 0.66% of drivers in weekday nighttime roadside surveys tested positive for cannabis (NHTSA roadside survey results)
04
In 2022, NHTSA reported that about 10,275 people died in alcohol-impaired driving crashes in 2021 (context for impaired-driving enforcement prioritization; used for relative program allocation comparisons)
Interpretation
Traffic Risk Interpretation
For the Traffic Risk angle, cannabis-positive driving is clearly present in fatal-crash contexts, with 13.4% of drivers in 2019 fatal crashes testing positive for drugs and cannabis making up the largest share, even though roadside surveys show lower levels such as 5.8% in 2022 and 0.66% in 2021 weekday nighttime.
07 · Category
Testing & Biomarkers3 stats
01
GC-MS/LC-MS/MS confirmatory testing is used after immunoassay screening to resolve cross-reactivity and improve analytical specificity, with confirmatory methods showing substantially higher analytical selectivity than immunoassays (forensic validation methods review with quantitative performance comparisons)
02
In a 2022 comparative study of confirmatory workflows, confirmatory LC-MS/MS achieved near-complete specificity (reported ~99% specificity) relative to screening immunoassays for THC in oral fluid samples (method-comparison results)
03
In 2020, a review in the Journal of Analytical Toxicology reported that impairment biomarkers other than THC (e.g., 11-OH-THC and THCCOOH) are used to expand interpretive windows, with detection performance varying by specimen and time since use (reviewed quantitative detection windows summarized across studies)
Interpretation
Testing & Biomarkers Interpretation
For the Testing and Biomarkers category, the evidence shows that moving from immunoassay screening to confirmatory LC MS MS raises specificity to about 99% for THC in oral fluid, and that broader impairment biomarkers like 11 OH THC and THCCOOH help widen interpretive windows with detection performance that depends on specimen type and time since use.
08 · Category
Market Dynamics3 stats
01
In 2023, the global roadside drug testing market was valued at $3.2 billion (industry market sizing and forecast dataset in a reputable market-research publication)
02
In 2023, the oral fluid drug testing market was valued at $2.5 billion (industry market sizing figure cited in a reputable market-research publication)
03
In 2021, the Substance Abuse and Mental Health Services Administration (SAMHSA) workplace drug testing guidance continued to recommend a screening/confirmatory testing algorithm for THC to support accurate evidentiary results (guidance repeats required algorithm)
Interpretation
Market Dynamics Interpretation
In the Market Dynamics landscape for Marijuana DUI, the roadside drug testing market reaching $3.2 billion in 2023 alongside $2.5 billion in oral fluid testing suggests strong momentum toward scalable testing methods, while SAMHSA’s 2021 continued push for a THC screening and confirmatory algorithm reinforces demand for accuracy and evidentiary reliability.
09 · Category
Health & Risk Evidence7 stats
01
In 2018, the NIH National Highway Traffic Safety Administration (NHTSA)-aligned impairment research literature estimates that the odds of crash involvement increase in the presence of THC, with effect sizes varying by study design and adjustment strategy (meta-analysis quantitative risk estimates)
02
In 2019, a systematic review reported that cannabis-impaired driving shows statistically significant association with crash risk, with pooled relative risk estimates varying across study categories (systematic review quantitative meta-analysis)
03
In 2020, a meta-analysis reported that cannabis use is associated with increased risk of motor vehicle accidents, with pooled odds ratios reported across multiple observational studies (quantitative pooled estimate)
04
In 2018, a naturalistic driving study measured that THC exposure reduced lane-keeping performance by a quantifiable margin, with variability depending on dose and driving conditions (peer-reviewed driving simulator/naturalistic quantitative outcomes)
05
In 2019, a driving performance study reported statistically significant impairment in reaction time after THC administration, with average increases in reaction time reported as measurable delays relative to placebo (peer-reviewed experimental results)
06
In 2022, a review in Addiction journal reported that impairment evidence for THC is supported by controlled experimental findings showing measurable cognitive and psychomotor effects (review with quantified effect summaries)
07
In 2021, a study using national survey data estimated that 33% of U.S. adults reported having tried cannabis at least once (self-reported usage prevalence figure in a nationally representative survey report)
Interpretation
Health & Risk Evidence Interpretation
Across Health and Risk Evidence, the research consistently links THC to higher crash risk and measurable driving impairment, with meta-analytic work across 2018 to 2020 reporting increased odds or relative risk and a 2018 naturalistic study showing reduced lane keeping, while 33% of U.S. adults in 2021 reported trying cannabis at least once, underscoring the public health relevance.
Reference
Cite This Report
This report is designed to be cited. We maintain stable URLs and versioned verification dates. Copy the format appropriate for your publication below.
APA
Ryan Townsend. (2026, February 13). Marijuana Dui Statistics. Gitnux. https://gitnux.org/marijuana-dui-statistics
MLA
Ryan Townsend. "Marijuana Dui Statistics." Gitnux, 13 Feb 2026, https://gitnux.org/marijuana-dui-statistics.
Chicago
Ryan Townsend. 2026. "Marijuana Dui Statistics." Gitnux. https://gitnux.org/marijuana-dui-statistics.
Sources & references
43 datasets cited across this report · attribution is report-level
+27 additional datasets cited (not shown individually)