Edible cannabis use on simulated driving performance

Abstract

Objective:

To assess driving performance after consuming edible cannabis using a driving simulator, examining frequency of use, THC dose, and rural versus urban settings.

Methods:

Adults in Denver, Colorado (N = 88), between 25 and 55 years old, were recruited from November 2023 to July 2024 and assigned to groups based on past 90-day cannabis use: daily (N = 29), occasional (N = 30), or no recent use (i.e., comparison group; N = 29). This within-subjects study compared driving performance (i.e., speed, lane departures per minute, and standard deviation of lateral placement (SDLP)) using a validated driving simulator (miniSim^™) at baseline and two post-consumption tests (52 and 119 min). Linear mixed models were used to evaluate performance.

Results:

Daily-use group drove slower than the occasional-use group (decrease in average speed from pretest of 2.49 mph (urban, p < 0.01 post-1); 1.80 mph (p = 0.02 post-2)) and the comparison group (decrease in average speed from pretest of 2.70 mph (urban, p < 0.01 post-1); 2.68 mph (urban, p < 0.01 post-2)). Lane departures increased in the occasional-use group from 0.17 at pretest to 0.47 in post-1 and 0.40 in post-2 (p < 0.01). Their rural SDLP increased: post-1 (29.37 cm, p < 0.01), post-2 (29.81 cm, p < 0.01), versus pretest (25.45 cm). This group had significantly greater change in lane departures than both the comparison group (0.26, p < 0.01) and daily-use group (0.34, p < 0.01) in post-1. In post-2, the change remained greater than the daily-use group (0.20, p = 0.02).

Conclusion:

Our study found significant changes in driving performance following edible cannabis use, including findings related to effect duration, use frequency, and road settings. The occasional-use group showed greater impairment than daily-use group, suggesting tolerance contributes to outcomes beyond THC concentration alone.

Introduction

Cannabis can be consumed through various methods, including smoking, vaping, dabbing, and oral ingestion in forms such as flower, oils, concentrates, topical creams, and edibles (Steigerwald et al. 2018; Wadsworth et al. 2022). The method and form of cannabis administration influence the onset, duration, and intensity of its effects (Barrus et al. 2016; Poyatos et al. 2020; Zipursky et al. 2020), which can also impact driving performance and related risks.

Prevalence of edible cannabis

While smoking remains the most common method of cannabis consumption, edibles are increasingly popular among adults in the United States (US) (Schauer et al. 2016; Steigerwald et al. 2018). In states where cannabis is legal for recreational or medical use, such as Colorado, rates of edible cannabis consumption are particularly high compared to states where cannabis is not legalized (Barrus et al. 2016; Goodman et al. 2020). Edible cannabis is available in a variety of forms, including baked goods, candies, chocolates, gummies, spreads, sublingual drops, pills, mouth sprays, and beverages (Barrus et al. 2016; Zipursky et al. 2020).

Differences in onset and duration: Inhaled versus edible cannabis

Inhaled cannabis and edible cannabis have distinct effects, and timing of those effects. Inhaled cannabis, such as through smoking or vaping, produces immediate effects, typically peaking within 20–30 min and declining within 2–3 h (Grotenhermen 2003). In contrast, edible cannabis has a delayed onset, with effects beginning 30–90 min after ingestion, peaking between 2–4 h, and lasting from 3–12 h (Grotenhermen 2003). The duration and intensity of these effects depend on individual factors such as body weight, metabolism, and eating habits (Grotenhermen 2003).

The primary difference between inhaled and edible cannabis lies in the absorption and metabolism of cannabinoids. Inhaled cannabis allows delta-9-tetrahydrocannabinol (delta-9-THC or THC) to rapidly enter the bloodstream and reach the brain within minutes. In contrast, edible cannabis must pass through the digestive system, where THC is metabolized in the liver into 11-hydroxy-delta9-tetrahydrocannabinol (11-OH-THC), a more potent psychoactive metabolite of THC (Barrus et al. 2016). This metabolic process alters both the intensity and duration of the effects.

Additionally, the presence of different cannabinoids in edible cannabis may influence its overall effects and impairment potential. While THC is the primary psychoactive compound, other cannabinoids such as cannabinol (CBN), cannabigerol (CBG), and cannabidiol (CBD) have varying effects. For example, CBG and CBD are non-psychoactive, while CBN exhibits mild psychoactive properties (Poyatos et al. 2020). These variations contribute to differences in impairment potential among cannabis products. The digestion and metabolism of edibles also introduce significant variability in THC absorption, resulting in different effects on individuals (Poyatos et al. 2020).

Difference in adverse effects: Inhaled versus edible cannabis

The method of cannabis consumption influences both the type and likelihood of adverse effects. Edible cannabis use has been associated with higher rates of psychiatric and cardiovascular complications compared to inhaled cannabis (Zipursky et al. 2020). It is also linked to an increase in emergency department visits due to gastrointestinal issues and intoxication, often resulting from overconsumption caused by the delayed onset of effects (Barrus et al. 2016; Poyatos et al. 2020; Zipursky et al. 2020; Peralt et al. 2022), particularly among individuals with less experience. Thus, the frequency of cannabis use may contribute to variations in both its effects and associated adverse outcomes and should be further investigated. Additionally, the delayed onset and prolonged effects of edibles have been linked to persistent psychotic symptoms that can last for days after consumption (Barrus et al. 2016). These effects may lead to different impairments and harms related to impairment, such as decreased driving performance and increased crash risks compared to inhaled cannabis.

Edible cannabis and driving impairment

Studies have shown that inhaled cannabis impairs driving performance and increases the risk of crashes (Brooks-Russell et al. 2021; Lira et al. 2021; Sevigny 2021; Fares et al. 2022; Simmons et al. 2022). Despite the growing popularity of edible cannabis and its unique pharmacokinetic profile, research on its effects on driving remains limited (Zhao et al. 2024). Therefore, understanding how edible cannabis affects driving performance is essential for developing better guidance on driving after its use.

Given the distinct differences between edible and inhaled cannabis (e.g., onset, duration, and potential adverse effects), it is vital to gather more information on how edible cannabis specifically affects driving performance. Furthermore, comprehensive data on factors such as consumption frequency is necessary, as the frequency of cannabis use can influence both outcomes and effects (Colizzi and Bhattacharyya 2018; Ramaekers et al. 2020). In this study, we assessed the driving performance using a driving simulator of individuals who consumed edible cannabis. We also examined how factors such as consumption frequency of use (occasional vs. daily), cannabis dose, and road settings (rural vs. urban) interact to affect driving performance.

Methods

Recruitment

Recruitment efforts included community outreach and advertisements targeting both individuals who use cannabis and those who do not in Denver, Colorado, from November 2023 to July 2024. Recruited participants were screened for eligibility through a web-based prescreening survey.

Eligible participants were between 25 and 55 years old, had a valid US driver’s license for at least three years with reported driving experience (an average of at least four days per month within the past six months), and were willing and able to complete two in-person study visits. These visits included a simulated driving test and toxicology tests (urine drug screening, alcohol breathalyzer test, venous whole blood test, and pupillometry). Eligible participants were not pregnant or breastfeeding.

Exclusion criteria included substance use or medical conditions that could interfere with driving performance, such as a history of substance use disorder or addiction (other than cannabis), use of medications known to impair cognitive or motor function, reported alcohol consumption exceeding three drinks per day within the past 30 days, and a past or current history of significant neurological or psychiatric disorders. Individuals with any known sleep disorder, including untreated sleep apnea, nocturnal schedules (e.g., night shift workers), or those who had donated 450 mL or more of blood in the past two weeks were also excluded.

Eligible participants were assigned to their respective cannabis use groups based on their frequency of cannabis use: daily, occasional, or no recent use. Daily use was defined as using cannabis at least once per day on 29 out of the past 30 days, excluding the day of data collection visit as participants were asked to refrain from use on the day of data collection to control for any potential effects from cannabis use outside the administration of the study. Occasional use was defined as using cannabis an average of at least two days per month but no more than three days per week in the past 90 days. We included a comparison group of participants who reported no recent cannabis use (i.e., no use in the 90 days prior to the study); this group did not have to be cannabis naïve. The comparison group was included to account for potential learning effects of using a driving simulator by following the same testing procedures. All use group classifications were based on any mode of cannabis use but also required some regular edible use. Additionally, eligible participants in the cannabis use groups with a history of clinically significant adverse effects related to cannabis intoxication (e.g., lightheadedness, nausea, dizziness) were further excluded.

Eligible participants were instructed to refrain from using non-prescription psychotropic drugs, hallucinogens, and synthetic cannabinoids during the duration of the study. They were also asked to avoid consuming more than three alcoholic drinks the day before data collection visits, using any cannabis products for at least 8 h, and using any edible cannabis products for at least 12 h prior to the visits. We also suggested that participants eat light meals approximately 1 h before the study visit.

Data collection

This within-subjects design study compared driving performance using data from three driving simulator tests: a baseline pretest in a sober condition and two post-tests (post-1 and post-2) following edible cannabis consumption. Post-2 served as a delayed assessment to evaluate the effects over time.

Eligible participants were screened during the first in-person data collection visit to confirm eligibility criteria. The screening included a urine test (approximately 30 mL) for drug screening of 13 commonly misused illicit and prescription medications, an alcohol breathalyzer test (Lifeloc FC10^™), heart rate and blood pressure measurements, a visual acuity test, and a urine pregnancy test for female participants. Participants also completed an abbreviated driving simulator test to identify those who developed symptoms of simulator adaptation syndrome, such as lightheadedness, headache, nausea, or vomiting. Participants were excluded if they reported or tested positive for benzodiazepines, methadone, ketamine, or phencyclidine (with or without a prescription), tested positive for alcohol on the breathalyzer, or had recently started taking any of the following medications in the past 7 days: amphetamines, barbiturates, buprenorphine, methamphetamine, opiates, oxycodone, propoxyphene, or tricyclic antidepressants. Additional exclusion criteria included uncorrected visual acuity disorder (defined as binocular vision worse than 20/40) or a body mass index greater than 40 kg/m². Participants who met eligibility criteria were scheduled for the second data collection visit, which took place 3–10 days after the first visit.

During the second data collection visit, participants in the cannabis use groups (occasional and daily) were instructed to bring their own edible cannabis, which they typically use, to replicate the desired effect during the visit. The edible cannabis had to be procured from a licensed dispensary, with the package labeled with THC potency and no more than a 1 to 1 ratio of THC to CBD (i.e., not be pre-dominately CBD). All participants began the second visit with a baseline pretest, which included a drug effect questionnaire, blood collection, an alcohol breathalyzer test, and a driving simulator test.

After the pretest, participants in the cannabis use groups were given approximately five minutes to ingest the self-provided edible cannabis at the dose they typically use, followed by a relaxation period. Cannabis consumption was directly observed by research personnel, who noted details of the package-labeled dose, the start time of use, and the amount consumed by the participants. Both use groups completed the first blood collection (i.e., post-1 blood test) approximately 45 min after edible consumption, followed by the post-1 driving assessment, which took place about 52 min after consumption. The post-2 blood collection occurred approximately 90 min after consumption, followed by the post-2 driving assessment at approximately 119 min post-consumption. The final post-3 blood collection was taken around 180 min after edible cannabis consumption, after the two post-test driving assessments. In addition to the driving assessments, the post-1 and post-2 tests included the same assessments as the pretest, with the addition of heart rate and blood pressure measurements. The comparison group only had blood collected at baseline. The comparison group’s post-1 and post-2 tests included all assessments taken during the pretest, with the post-1 driving assessment occurring 5 min after the pretest, and the post-2 driving assessment approximately 57 min after the pretest.

Study participants were required to have a designated sober driver for transportation after the data collection visit. Written informed consent was obtained from all participants during the initial in-person visit. All study procedures and protocols were reviewed and approved by the Colorado Multiple Institutional Review Board.

Study measures

For this specific analysis, we assessed the data on driving performance based on the driving simulator test and edible cannabis use.

Edible cannabis use and blood samples

The participant’s frequency of cannabis consumption was determined based on self-reported data collected during the screening visit. The amount of THC ingested by participants was chosen by the participants. Details of the dose consumed was noted according to the dose labeled on the packaging of the edible cannabis.

Whole blood samples were tested for THC, 11-OH-THC, 11-nor-9-carboxy-Δ9-tetrahydrocannabinol (THC-COOH), THC-glucuronide, THC-COOH-glucuronide, CBN, CBG, and CBD. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) was used for cannabinoid analysis (Henthorn et al. 2024; Brooks-Russell et al. 2025). In this paper, we focused on THC and its metabolites (11-OH-THC and THC-COOH).

Driving simulator

Driving performance was measured using the validated miniSim^™ driving simulator, developed by the University of Iowa Driving Safety Research Institute (Brooks-Russell et al. 2021). The second data collection visit consisted of three driving sessions (pretest, post-1, and post-2), each drive lasting approximately 20 min. Participants completed the simulated drives during scheduled appointments at times convenient for their personal schedules and intended cannabis use. Appointments were available on both weekdays and weekends, ranging from morning to afternoon hours.

The simulation was conducted in a darkened room with blackout curtains to control lighting conditions. The simulated environment depicted a daytime driving scenario with clear, sunny weather and included both urban and rural settings. Each drive began in an urban area with a posted speed limit of 35 miles per hour (mph), followed by an urban arterial portion with a speed of 45 mph, and then transitioned to a rural setting with a posted speed limit of 55 mph approximately halfway through the 20-minute assessment.

The urban segment included ambient traffic, pedestrians, parked vehicles, and occasional cyclists, with surrounding features such as sidewalks and buildings ranging from one to several stories, generally set back from the road. Several intersections were encountered during this segment, though no turns were made. The rural segment also featured ambient traffic and included wooded areas, open grassy fields, and rolling hills. Certain portions of the drive involved secondary tasks (i.e., distractions) and subtle steering or braking challenges; these were recorded but are not analyzed in the current manuscript. The urban and rural segments were designed to provide representative driving environments; no attempts were made to equate the level of demand between the urban and rural driving tasks.

Each drive began with standardized, prerecorded instructions. Study personnel silently observed participants throughout the session, noting any spontaneous comments or unusual driving behaviors (e.g., unnecessary lane changes). If a participant exceeded or fell below the posted speed limit by more than 10 mph, the study personnel issued a single verbal reminder of the posted speed limit. All driving simulator tests randomized the sequence of events to minimize learning effects and were designed to assess the same driving performance skills.

Driving performance measurements included the standard deviation of lateral placement (SDLP), which is a measure of car weaving (Brooks-Russell et al. 2021), the rate of lane departures per minute, and average driving speed. Lane departures were calculated as the number of instances, during predetermined driving segments, in which the vehicle crossed either the center line or the edge line. Since our study used a within-subjects design, any tendency for participants to straddle the line would be consistent across all conditions.

Data analysis

The main outcomes of interest in this analysis were driving performance measures in both rural and urban settings, which were analyzed using separate linear mixed effects models. Specific outcomes were SDLP, rate of lane departures, and average driving speed. The fixed effects in each model included age, frequency of cannabis use (i.e., daily, occasional, or no recent use), time since consumption (pretest, post-1, and post-2), and the interaction between frequency of use and time since consumption. Preliminary analyses indicated that gender did not influence the results and was therefore not included as a covariate in the final model. Average driving speed was controlled for in the SDLP and rate of lane departure models. A subject-specific random intercept was included in all models to account for within-participant repeated measures. For each group, the difference in covariate-adjusted least square means between baseline and each post-consumption time-point was calculated and assessed for statistical significance. Contrasts examining the pre- versus post-consumption least squared mean differences between groups were computed to assess the association between cannabis use history and within-participant change in driving performance. A descriptive analysis was conducted for the cannabis use groups, focusing on cannabis use history, THC concentration, and demographic characteristics. Statistical significance was set at a p-value of 0.05. All analyses were conducted using R version 4.4.2 (R Core Team 2024).

Results

Participant characteristics

A total of 88 participants (40.9% male) were included in the final study examining edible cannabis use, with N = 30 (34.1%) reported using cannabis occasionally, 29 (32.9%) using daily, and additional 29 (32.9%) reporting no use in the past 90 days (i.e., the comparison group). The majority of the participants were between 25 and 35 years old. The largest groups by race/ethnicity and education level were non-Hispanic White and individuals with at least a college degree, respectively. All groups had an average of 15 to 18 years of driving experience. Refer to Table 1 for all demographic characteristics.

Table 1.

Demographic characteristics of participants by cannabis use groups.

Demographic Characteristic	Comparison Group1 (n = 29)	Occasional-Use1 (n = 30)	Daily-Use1 (n = 29)
Gender
Female	18 (62%)	13 (43.4%)	18 (62%)
Male	11 (38%)	15 (50%)	10 (34.5%)
Other	0 (0%)	2 (6.6%)	1 (3.5%)
Age Group
25-35	20 (69%)	18 (60%)	23 (79%)
36-55	9 (31%)	12 (40%)	6 (21%)
Weight (kg) Race	75.3 (52.2-152.0)	82.5 (51.5-121.4)	75.9 (45.8-10.5.1)
Race
White	26 (90%)	25 (83%)	19 (66%)
Other	3 (10%)	5 (17%)	10 (34%)
Ethnicity
Hispanic or Latino	3 (10%)	4 (13%)	7 (24%)
Not Hispanic or Latino	26 (90%)	26 (87%)	21 (72%)
No Response Education	0 (0%)	0 (0%)	1 (3.4%)
Education
High School and Some College	2 (6.9%)	2 (6.7%)	7 (24%)
Completed College Degree	13 (45%)	15 (50%)	17 (59%)
Graduate Degree	14 (48%)	13 (43%)	5 (17%)
Driving Experience, Years	16.69 (7.18)	18.23 (7.99)	15.53 (4.81)
Age at First Use of Cannabis, Years	21.00 (6.00)	21.43 (7.41)	17.62 (3.68)
Lifetime Cannabis Use on a Weekly Basis	8 (28%)	29 (97%)	29 (100%)

Note: Comparison group defined as no cannabis use in 90 days prior to the study. Lifetime weekly cannabis use is defined as having used cannabis on a weekly basis at any point during an individual’s lifetime.

¹n (%); Mean (SD or Range).

Cannabis use and effect

All participants in the occasional and daily edible use groups reported using cannabis at least weekly, and 28% of the comparison group reported using cannabis at least weekly at some point in their lifetime. The average age at first use of cannabis for the comparison, occasional-use, and daily-use groups was 21 years old, 21.4 years old, and 17.6 years old, respectively. For the occasional-use group, participants reported consuming cannabis an average of 7.72 days (SD 4.18) in the past 29/30 days, with usage occurring 1.80 days per week (SD 0.97) and an average of 1.04 times per day on days of cannabis use (SD 0.12). Most participants in the occasional-use group reported using cannabis between 9 and 18 h after waking up. For the daily-use group, participants reported using cannabis on an average of 28.61 days (SD 3.06) in the past 29/30 days, with usage occurring 6.68 days per week (SD 0.71) and an average of 2.39 times per day on days of cannabis use (SD 1.71). The daily-use group reported varying times of use after waking up, with the plurality (28%) reporting use 9-12 h after waking, followed by 21% using within 1 h after waking, and 17% using between 6 and 9 h of waking up.

During our study, all but four participants in the occasional-use group (87%) consumed edible cannabis in the form of gummies (see Table 2). The median THC concentration of the edibles consumed, as indicated on the packaging labels, was 7.50 mg (interquartile range [IQR]: 5.00 mg). The daily-use group also primarily consumed edibles in the form of gummies (90%), with two participants consuming them in powder form (5.6%) and one in taffy form (5.6%). The edibles consumed had a median THC concentration of 20.00 mg (IQR: 20.00 mg), as indicated on the packaging labels.

Table 2.

Cannabis use characteristics of daily and occasional cannabis use group.

Cannabis Use Characteristic	Occasional-Use1 (n = 30)	Daily-Use1 (n = 29)
In the Past 29/30 Days:
Number of Days Used	7.72 (4.18)	28.61 (3.06)
Number of Days Used Per Week	1.80 (0.97)	6.68 (0.71)
Times Used Per Day on Days of Cannabis Use	1.04 (0.12)	2.39 (1.71)
Time of Day Consumed After Waking:
Immediately upon waking	0 (0%)	1 (3.4%)
Within 1 h	0 (0%)	6 (21%)
1-3 h	1 (3.3%)	2 (6.9%)
3-6 h	1 (3.3%)	4 (14%)
6-9 h	1 (3.3%)	5 (17%)
9-12 h	11 (37%)	8 (28%)
12-18 h	16 (53%)	3 (10%)
Edible Cannabis Consumed During Study:
Type of Edible Cannabis
Gummies	26 (87%)	26 (90%)
Candy	1 (3.3%)	0 (0%)
Chocolate	1 (3.3%)	0 (0%)
Powder	1 (3.3%)	2 (6.9%)
Taffy	0 (0%)	1 (3.4%)
Mint	1 (3.3%)	0 (0%)
THC Concentration, mg2	7.50 (5.00, 10.00)	20.00 (10.00, 30.00)

Note:

¹n (%); Mean (SD);

²Median (IQR).

The whole blood samples tested reflect the concentration of THC and THC metabolites present in the use groups after edible cannabis consumption (see Table 3). The comparison group had a median level of 0.00 ng/mL (IQR: 0.00) for THC, 11-OH-THC, and THC-COOH, confirming no use in the past 90 days. The occasional-use group had a baseline median concentration of 0.00 ng/mL (IQR: 0.00) for THC and 11-OH-THC, and a minimal level of THC-COOH at 2.03 ng/mL (IQR: 2.56), reflecting less frequent use compared to the daily-use group. In contrast, the daily-use group showed measurable median concentrations of THC and its metabolites at baseline: 2.09 ng/mL (IQR: 4.32) of THC, 1.16 ng/mL (IQR: 2.03) of 11-OH-THC, and 45.80 ng/mL (IQR: 58.31) of THC-COOH, indicating a higher frequency of use. For both the occasional-use and daily-use groups, THC concentration levels increased from baseline in the post-1 blood test (approximately 45 min after consumption, before the post-1 driving assessment). The post-2 blood test (drawn approximately 90 min after consumption, after post-1 but before the post-2 driving assessment) showed a further increase in the concentration levels of THC and its metabolites.

Table 3.

Cannabinoids in participant whole blood sample test results at pretest and post-edible cannabis consumption by cannabis use groups.

	Comparison (n = 29) Median (IQR)	Occasional-Use (n = 30) Median (IQR)				Daily-Use (n = 29) Median (IQR)
	Baseline1	Baseline1	Post11	Post21	Post31	Baseline1	Post11	Post21	Post31
THC	0.00 (0.00)	0.00 (0.00)	1.04 (2.85)	1.88 (2.29)	1.14 (1.39)	2.09 (4.32)	7.04 (7.87)	9.38 (13.46)	5.68 (9.10)
11-OH-THC	0.00 (0.00)	0.00 (0.00)	1.01 (1.24)	1.43 (1.17)	1.04 (0.85)	1.16 (2.03)	4.33 (4.23)	5.12 (5.32)	3.67 (4.59)
THC-COOH	0.00 (0.00)	2.03 (2.56)	8.10 (9.43)	15.28 (12.37)	12.89 (11.26)	45.80 (58.31)	70.63 (85.87)	101.53 (80.67)	69.78 (85.85)

Note: Comparison group defined as no cannabis use in 90 days prior to the study. Baseline samples were collected at pretest, prior to edible cannabis consumption. Post1, Post2, and Post3 blood samples were collected approximately 45, 90, and 180 min after edible cannabis consumption, respectively.

¹Values represent Median (IQR) in ng/mL.

These results indicate that THC and THC metabolite concentrations increased post-consumption before the post-1 driving assessment and remained present before the post-2 driving assessment, with variations in concentration levels between the two post-consumption driving assessments. The post-3 blood test (drawn approximately 180 min after consumption, following the post-2 driving assessment) showed a decrease in concentration levels in both use groups compared to the post-2 blood test. This suggests that the driving assessments likely measured driving performance within the window of peak THC concentration levels post-consumption.

Driving performance by frequency of use

Average driving speed

In the comparison group, participants drove below the posted speed limit on average during the pretest, averaging 54.3 miles per hour (mph) in rural settings (55 mph speed limit) and 32.2 mph in urban settings (35 mph speed limit). Their speeds increased in both rural and urban settings during post-1 and post-2 compared to the pretest (all p < 0.01; see Table 4). In rural settings, the comparison group’s speed minimally increased from the pretest to above the speed limit, reaching 55.0 mph in both post-1 and post-2; while in urban settings, the speed increased but remained below the speed limit, with speeds of 34.0 mph in post-1 and 34.7 mph in post-2.

Table 4.

Adjusted model comparing driving performance after edible consumption by cannabis use group.

	Comparison (n = 29)					Occasional-Use (n = 30)					Daily-Use (n = 29)
	Pre	Post1	Post2	p1	p2	Pre	Post1	Post2	p1	p2	Pre	Post1	Post2	p1	p2
Speed (Rural)	54.33	55.00	55.08	<0.01	<0.01	54.44	54.52	54.89	0.69	0.03	54.54	54.73	54.79	0.38	0.24
Speed (Urban)	32.24	34.02	34.67	<0.01	<0.01	33.03	34.59	34.59	<0.01	<0.01	34.49	33.57	34.25	0.09	0.65
Lane Departure (Rural)	0.17	0.20	0.23	0.65	0.30	0.17	0.47	0.40	<0.01	<0.01	0.12	0.08	0.15	0.45	0.65
Lane Departure (Urban)	0.07	0.11	0.15	0.50	0.19	0.10	0.05	0.22	0.42	0.03	0.01	0.05	0.00	0.48	0.80
SDLP (Rural)	24.21	26.05	25.77	0.08	0.13	25.45	29.37	29.81	<0.01	<0.01	23.65	24.05	24.22	0.70	0.58
SDLP (Urban)	22.02	21.68	22.32	0.76	0.79	24.14	24.41	24.76	0.81	0.58	20.63	20.28	20.95	0.76	0.77

Note: Comparison group defined as no cannabis use in 90 days prior to the study. Speed is the average driving speed. The posted speed limit was 55 miles per hour and 35 mph for rural and urban settings, respectively. Lane departure measures the rate of lane departures per minute. SDLP is the Standard deviation of lateral placement. Pre refers to the outcome measurement from the pretest at baseline, before consumption in a sober state. Post1 and Post2 are measurements taken after edible cannabis consumption for the use groups (approximately 52 min and 119 min post-consumption, respectively). p1 and p2 are p-values comparing Pre to Post1 and Pre to Post2, respectively. Significant p-values (p < 0.05) are bolded.

A similar pattern of statistically significant increases in average speed at post-1 and post-2 compared to the pretest was observed in the occasional-use group in urban settings, and from pretest to post-2 in rural settings. However, these changes were minimal relative to the posted speed limits, even at the extreme quartiles of speed (10^th and 90^th percentiles). In urban settings, participants in the occasional-use group initially drove an average of 1.97 mph below the posted speed limit, then increased their speed to 0.41 mph below the speed limit at both post-tests (p < 0.01). In contrast, the daily-use group in both rural and urban settings, maintained average speeds slightly below the posted speed limit throughout all test periods, with no statistically significant changes from pretest to post-1 or post-2 (see Table 4).

When comparing cannabis use groups after edible consumption, changes in average speed from baseline to both post-consumption time points in the urban setting significantly differed between the daily-use and occasional-use groups, as well as between the daily-use and comparison groups. In the urban setting, the daily-use group decreased their speed, while the occasional-use and no- use groups increased theirs. This resulted in differences in average speed changes of 2.49 mph (p < 0.01) and 1.80 mph (p = 0.02) between the daily-use and occasional-use groups at post-1 and post-2, respectively (see Table 5). Similarly, differences in average speed change between the daily-use and comparison groups were 2.70 mph (p < 0.01) and 2.68 mph (p < 0.01) at post-1 and post-2, respectively. In the rural setting, a significant difference in speed change from baseline was observed between the occasional-use and comparison groups, with a difference of 0.59 mph at post-1 (p = 0.05). No statistically significant differences in speed change were observed between the daily-use and occasional-use groups, or between the daily-use and comparison groups in the rural setting; nor between the occasional-use and comparison groups in the urban setting at either post-1 or post-2.

Table 5.

Differences in driving performance from pretest to post-1 and post-2 following edible cannabis consumption, between cannabis use groups.

	Post-1 Test						Post-2 Test
	Occasional vs. Comparison		Occasional vs. Daily-Use		Daily vs. Comparison		Occasional vs. Comparison		Occasional vs. Daily-Use		Daily vs. Comparison
	Est	p	Est	p	Est	p	Est	p	Est	p	Est	p
Speed (Rural)	−0.59	0.05	−0.10	0.731	−0.49	0.10	−0.30	0.32	0.20	0.49	−0.50	0.09
Speed (Urban)	−0.22	0.77	2.49	<0.01	−2.70	<0.01	−0.87	0.25	1.80	0.02	−2.68	<0.01
Lane Departure (Rural)	0.26	<0.01	0.34	<0.01	−0.07	0.40	0.16	0.06	0.20	0.02	−0.04	0.67
Lane Departure (Urban)	−0.09	0.29	−0.09	0.29	0.00	0.99	0.05	0.57	0.14	0.09	−0.09	0.26
SDLP (Rural)	2.08	0.15	3.53	0.01	−1.44	0.32	2.81	0.05	3.80	<0.01	−0.99	0.50
SDLP (Urban)	0.61	0.70	0.61	0.70	0.00	1.00	0.31	0.84	0.29	0.85	0.02	0.99

Note: Comparison group (n = 29) defined as no cannabis use in 90 days prior to the study. Occasional-use group: n = 30; daily-use group: n = 29. Speed is the average driving speed. The posted speed limit was 55 miles per hour and 35 mph for rural and urban settings, respectively. Lane departure measures the rate of lane departures per minute. SDLP is the standard deviation of lateral placement.

Significant p-values (p < 0.05) are bolded.

Rate of lane departures per minute and SDLP

Differences in the rate of lane departures per minute and the SDLP from pretest to post-1 and post-2 were not statistically significant for the comparison or daily-use groups. However, the occasional-use group showed a statistically significant increase in both lane departures and SDLP in rural settings. The rate of lane departures per minute more than doubled, increasing from 0.17 at pretest to 0.47 in post-1 and 0.40 in post-2 (p < 0.01 for both; see Table 4). SDLP for the occasional-use group in the rural setting was also higher in post-1 and post-2, reaching 29.37 and 29.81 cm (both p < 0.01), respectively, compared to 25.45 cm at pretest. In the urban setting, the rate of lane departures in the occasional-use group increased significantly from baseline to post-2, rising from 0.10 to 0.22 per minute (p = 0.03).

When comparing use groups after edible cannabis consumption, the occasional-use group in rural settings showed a statistically significant difference in the change in the rate of lane departures from baseline to post-1, compared to both the comparison group (difference = 0.26, p < 0.01) and the daily-use group (difference = 0.34, p < 0.01). This difference between the occasional-use and daily-use groups remained significant at post-2, with a difference of 0.20 (p = 0.02) (see Table 5). Additionally, at both post-consumption time points, the occasional-use group exhibited a greater change in SDLP from baseline in rural settings, 3.53 cm (p < 0.01) at post-1 and 3.80 cm (p < 0.01) at post-2, compared to the daily-use group.

Discussion

This study examined driving performance, including average driving speed, SDLP, and rate of lane departures (per minute), after edible cannabis consumption among adults (ages 25–55) using a driving simulator. Groups were compared based on use frequency (i.e., comparison, occasional-use, and daily-use) and time since consumption (i.e., post-1 and post-2 compared to pretest). We also analyzed differences in THC and THC metabolite concentrations in whole blood samples taken before and after edible cannabis consumption, as well as differences in driving performance across road settings (i.e., rural vs. urban). We found statistically significant differences in driving performance after edible cannabis consumption, particularly within the occasional-use group, as well as differences between the use groups.

Statistically significant results were found when comparing speed from pretest to post-1 and post-2 for the occasional-use and comparison groups, while no statistically significant difference was observed in the daily-use group, which consistently drove below the speed limit. The average change in speed among the daily-use group, who drove more slowly, was significantly different from both the comparison and occasional-use groups in the urban setting. In the rural setting, a significant difference in the average change in speed between the comparison and occasional-use groups was observed from baseline to post-1. These findings align with existing literature, which suggests that individuals tend to drive at slower speeds after cannabis consumption, particularly among individuals who use cannabis more frequently (Brands et al. 2019; Brooks-Russell et al. 2021; Miller et al. 2022; Zhao et al. 2024). Despite the statistical significance, the actual reduction in speed in our study was not safety critical, as the changes were minimal (less than 3 mph, with most below 1 mph). Additionally, no group, on average, drove at speeds so low or high relative to the posted speed limit that they indicated impairment related to speed. Therefore, it is difficult to draw strong inferences regarding impairment based on speed changes after edible cannabis consumption.

Our findings on SDLP and lane departure rates were both statistically significant and clinically relevant. The results suggest that impairment following edible cannabis consumption was evident in SDLP and lane departure outcomes, particularly in rural settings for the occasional-use group. In rural settings, the occasional-use group showed statistically significant SDLP increases of 3.92 cm in post-1 and 4.36 cm in post-2 compared to pretest measurements. This change in SDLP is significant, as prior studies have indicated that an increase of 2.4 cm in SDLP is a notable indicator of impairment (Verster and Roth 2011; Manning et al. 2024). The occasional-use group in our study also exhibited a significant increase in lane departures per minute (0.47 in post-1 and 0.40 in post-2, compared to 0.17 in the pretest) and demonstrated higher rates of change in lane departures compared to the comparison group at post-1, and the daily-use groups at both post-1 and post-2 in rural settings. These findings align with previous literature showing increased SDLP and lane departures after cannabis use (Hartman et al. 2015; Bondallaz et al. 2016; Brooks-Russell et al. 2021; Miller et al. 2022). Overall, our findings reflect behavior patterns similar to those observed with inhaled cannabis, with differences in effect size.

In our study, we found multiple results that differentiate the occasional-use group from the daily-use group. Speed changes were observed in the daily-use group, especially in urban settings. In contrast, SDLP and lane departures were more prevalent among the occasional-use group and in rural settings, significantly affecting driving performance. No statistically significant changes in SDLP or lane departures were observed for the comparison or daily-use groups. The observation that the occasional-use group exhibited higher impairment in lane departures and SDLP compared to the daily-use group suggests that frequency of use may influence impairment levels. In rural settings, daily-use group had lower change of SDLP from pretest to both post-tests compared to the occasional-use group, further emphasizing the higher level of impairment related to SDLP performance among individuals who use cannabis less frequently. Other studies have similarly reported greater impairment in occasional-use groups compared to both comparison and daily-use groups (Brooks-Russell et al. 2021; Miller et al. 2022). McCartney et al. (McCartney et al. 2022) found that impairment was more pronounced in individuals who use cannabis occasionally, whereas individuals who use regularly (defined as weekly) showed no significant biomarker indication of impairment. This discrepancy may be attributed to varying tolerance levels (Desrosiers et al. 2015; Brooks-Russell et al. 2021; Powlowski et al. 2025). Differences in THC concentration levels further support this, as reflected in our baseline and post-1 blood test results. For example, the daily-use group had an average THC concentration of 5.41 ± 7.55 ng/mL at baseline and 10.63 ± 10.61 ng/mL at post-1, whereas the occasional-use group had 0.46 ng/mL at baseline and 2.58 ± 3.82 ng/mL at post-1. Burt et al. (Burt et al. 2021) found that THC concentrations of 2–5 ng/mL were associated with impairment, but greater impairment was observed in occasional-use groups despite the concentration levels.

When evaluating the interaction between THC dose, cannabis use frequency, and driving performance, no significant results were found. This suggests that driving performance may be more closely related to THC tolerance and use frequency rather than THC dose alone. Additionally, cannabis effects on driver impairment are influenced by pharmacokinetics, metabolism, and individual factors such as polysubstance use and food consumption prior to cannabis intake (Bondallaz et al. 2016; Dowd et al. 2023). Thus, THC product concentration or dose consumed alone should not be interpreted as a direct measure of impairment, as multiple factors interact. However, this does not necessarily mean that daily use results in less driving impairment or is safer than occasional use. Regular cannabis use has been associated with long-term cognitive and motor function impairments that may affect driving skills (Bondallaz et al. 2016). Since our study focused on the acute effects of edible cannabis, this distinction should be considered when interpreting our findings.

Further, the assessment of time from consumption is another important finding from our study. Unlike inhaled cannabis, where effects are immediate, edibles take 30 min to 3 h to manifest (Schlienz et al. 2020; Burt et al. 2021). We found statistically significant driving performance impairments at post-1 (approximately 52 min after consumption), which remained significant at post-2 (approximately 119 min after consumption) after consuming edible cannabis. This is consistent with another study on edible cannabis consumption and driving performance, where impairment was observed at 2 h post-consumption, but no longer had an effect at 4 and 6 h (Zhao et al. 2024). Our blood test results also showed increasing THC and metabolite concentrations from baseline to post-1 and post-2 (45- and 90-minutes post-consumption, respectively), followed by a decrease at post-3 (180 min post-consumption) in both occasional-use and daily-use groups. For the daily-use group, THC levels in blood samples tended to remain elevated compared to the occasional-use group, particularly for metabolites that persist longer in the body (Grotenhermen 2003; Burt et al. 2021). This supports the understanding that the effects of edible cannabis last longer than those of inhaled cannabis, and that both the method of cannabis consumption (i.e., inhaled versus edibles) and the frequency of use must be considered when developing guidelines for driving after use, rather than focusing solely on cannabis use as a whole and THC concentration alone.

Limitations

Our study is not without limitations. As this was an observational study, we did not quantify administered cannabis consumption, and self-dosed intake may have limited standardization. The documented cannabis dose could also be under- or overestimated, as it was based solely on package labels; however, measures were in place to verify that labels came from licensed facilities adhering to labeling guidelines. Participants were also observed by study personnel while administering cannabis to ensure accurate documentation of the dose used. However, the exact dosage was difficult to measure, potentially leading to individual variability in the effects of the dose. Differences in daily and occasional cannabis consumption, including intake amounts and product concentrations, may have influenced the results. Additionally, the recruited participants were mostly non-Hispanic white, and the study focused on individuals aged 25–55 years, which may limit generalizability. However, this age restriction was implemented to minimize the confounding effects of age and driving experience. Variations in driving performance may also have limited the ability to detect differences between user groups. The study was powered to detect a 3.6 cm SDLP difference, which may have underestimated or failed to detect smaller effects. While the driving simulator was validated and provided a valuable controlled assessment, it may not fully replicate the full range of real-world driving conditions such as navigation or crash avoidance. Finally, findings were specific to Colorado, where a higher prevalence of daily cannabis use group may have influenced the results.

Despite these limitations, our study provides valuable insight into the effects of edible cannabis use on driving performance using a validated driving simulator. It also reflects individuals’ typical cannabis use, making the findings more generalizable to real-life conditions. Our results contribute to the literature on edible cannabis use and driving performance, considering factors like use frequency and road settings, and offer evidence to help inform guidelines and policies on cannabis use and driving safety.

Conclusion

Our study evaluated driving performance after edible cannabis use and found significant decreases in speed, along with increases in SDLP and lane departures, which persisted for almost 2 h post-consumption. Notably, our SDLP findings were nearly double the levels that signaled impairment in previous literature. We also observed differences based on frequency of use, with occasional-use group showing greater impairment than daily-use group, suggesting that tolerance plays a role beyond just THC concentration. Additionally, driving performance varied between rural and urban settings. When assessing driving impairment from cannabis use, factors such as the method of consumption, onset time, prolonged effects, frequency of use, and road settings must all be considered. Given the delayed and prolonged effects of edible cannabis, along with the driving impairment observed in our study and previous research, clear public communication about their impact on driving is crucial, especially as the use of edible cannabis continues to rise.