Advertisement

Risk Factors Associated with Driving After Marijuana Use among US College Students During the COVID-19 Pandemic

      Abstract

      Purpose

      To assess the sociodemographic and behavioral risk factors associated with driving after marijuana use among US college students.

      Methods

      A secondary analysis used the Fall 2020 and Spring 2021 American College Health Association- National College Health Assessment III and the dataset was restricted to college students ≥ 18 years of age who reported recent driving and marijuana use. Associations between risk factors and driving after marijuana use were estimated using multivariable logistic regression.

      Results

      A total of 29.9% (n=4,947) of the respondents reported driving after marijuana use. Males (adjusted odds ratio [AOR]: 1.64, 95% confidence interval [CI]:1.48-1.82), non-Hispanic Black (AOR: 1.32, 95% CI:1.02-1.71), sexual minorities (AOR:1.19, 95% CI: 1.07-1.31), individuals with an alcohol or substance use disorder (AOR: 1.44, 95% CI:1.08-1.91), anxiety (AOR: 1.20, 95% CI: 1.06-1.36), higher suicidality (AOR: 1.18, 95% CI: 1.07-1.31), and those who also drank and drove (AOR: 3.18, 95% CI: 2.84-3.57) had a higher risk of driving after marijuana use.

      Conclusions

      Future research should focus on increasing awareness of driving after marijuana use and prevention programs and/or strategies on college campuses regarding driving after marijuana use for these groups to reduce this risky behavior.

      Keywords

      Abbreviation list: (alphabetical):

      Adjusted odds ratio (AOR), Alcohol, Smoking and Substance Specific Involvement Test (ASSIST), Alcohol use disorder (AUD), American College Health Association-National College Health Assessment III (ACHA-NCHA III), Confidence intervals (CIs), Coronavirus disease 2019 (COVID-19), Driving after marijuana use (DAMU), Odds ratios (OR), Suicide Behaviors Questionnaire-Revised (SBQ-R), Substance use disorder (SUD), United States (U.S.)
      Implications and Contribution
      Driving after marijuana use is an important public health issue among college aged students in the US. A national survey of US college students showed decreasing trend in driving after marijuana use. However, significant effort must be done to increase prevention programs on college campuses regarding driving after marijuana use.

      Introduction

      Marijuana is the most prevalent non-alcoholic drug identified among drivers involved in motor vehicle collisions(
      • Azofeifa A.
      • Rexach-Guzmán B.D.
      • Hagemeyer A.N.
      • et al.
      Driving Under the Influence of Marijuana and Illicit Drugs Among Persons Aged ≥16 Years - United States, 2018.
      ). The 2018 National Survey on Drug Use and Health found that 12 million (4.7%) United States (U.S.) individuals aged ≥ 16 years reported driving under the influence of marijuana in the past 12 months (
      • Azofeifa A.
      • Rexach-Guzmán B.D.
      • Hagemeyer A.N.
      • et al.
      Driving Under the Influence of Marijuana and Illicit Drugs Among Persons Aged ≥16 Years - United States, 2018.
      ). As of 2021, 36 states have enacted legislation to decriminalize marijuana for medical use. Among those, 18 states, two U.S. territories, and the District of Columbia have decriminalized marijuana for recreational use (

      National Conference of State Legislatures. State Medical Marijuana Laws. Available at: https://www.ncsl.org/research/health/state-medical-marijuana-laws.aspx2021.

      ). Additionally, past year marijuana use has increased from 38% in 2015 to 44% in 2020 among college students (
      • Schulenberg J.E.P.M.
      • Johnston L.D.
      • O'Malley P.M.
      • Bachman J.G.
      • Miech R.A.
      Monitoring the Future national survey results on drug use, 1975-2020.
      ).
      Compared to more experienced drivers, marijuana-impaired driving is particularly salient for young, inexperienced drivers who have a higher crash risk (
      • Shope J.T.
      • Bingham C.R.
      Teen driving: motor-vehicle crashes and factors that contribute.
      ). Drivers aged 21-25 years of age have the highest prevalence of driving under the influence of marijuana (
      • Azofeifa A.
      • Rexach-Guzmán B.D.
      • Hagemeyer A.N.
      • et al.
      Driving Under the Influence of Marijuana and Illicit Drugs Among Persons Aged ≥16 Years - United States, 2018.
      ,
      • Azofeifa A.
      • Mattson M.E.
      • Lyerla R.
      Driving Under the Influence of Alcohol, Marijuana, and Alcohol and Marijuana Combined Among Persons Aged 16-25 Years - United States, 2002-2014.
      ), and college students are a population at increased risk of substance-related impaired driving (
      • Carter A.C.
      • Brandon K.O.
      • Goldman M.S.
      The college and noncollege experience: a review of the factors that influence drinking behavior in young adulthood.
      ,
      • Jewett A.
      • Peterson A.B.
      • Sauber-Schatz E.K.
      Exploring substance use and impaired driving among adults aged 21 years and older in the United States, 2015.
      ). Although driving under the influence of alcohol has decreased in the U.S. among youth over the last few decades, rates of driving after marijuana use (DAMU) have stayed consistent or risen (
      • Azofeifa A.
      • Mattson M.E.
      • Lyerla R.
      Driving Under the Influence of Alcohol, Marijuana, and Alcohol and Marijuana Combined Among Persons Aged 16-25 Years - United States, 2002-2014.
      ). Additionally, many U.S. campuses have programs aimed at reducing alcohol impaired driving and underage drinking, but do not have drugged driving prevention programs (
      U.S. Department of Education OoSaD-FS
      Higher Education Center for Alcohol and Other Drug Prevention.
      ).
      Several studies have reported increased alcohol and drug consumption during the coronavirus disease 2019 (COVID-19) pandemic, including the use of marijuana (
      • Rogés J.
      • Bosque-Prous M.
      • Colom J.
      • et al.
      Consumption of Alcohol, Cannabis, and Tobacco in a Cohort of Adolescents before and during COVID-19 Confinement.
      ,
      • Clendennen S.L.
      • Case K.R.
      • Sumbe A.
      • et al.
      Stress, Dependence, and COVID-19-related Changes in Past 30-day Marijuana, Electronic Cigarette, and Cigarette Use among Youth and Young Adults.
      ). College students have suffered heightened stress, decreased social connectedness, and higher anxiety about academic performance and job pressure (
      • Son C.
      • Hegde S.
      • Smith A.
      • et al.
      Effects of COVID-19 on College Students’ Mental Health in the United States: Interview Survey Study.
      ) as a result of the sudden societal changes implemented during the COVID-19 pandemic, such as university closures. These factors may have significantly contributed to increased drug use and increased risk behaviors that are positively associated with increased drug use, including drug-impaired driving (
      • Hasan R.
      • Watson B.
      • Haworth N.
      • et al.
      A systematic review of factors associated with illegal drug driving.
      ).
      To date, no published studies have investigated DAMU among college students during the COVID-19 pandemic. To fill the gap in the extant literature, this study: 1) describes the prevalence of DAMU among US college students who drive and also report recent marijuana use, and 2) identifies sociodemographic and behavioral risk factors associated with DAMU among these college students during the COVID-19 pandemic. Findings from this study may inform the development of public health or campus prevention programs aimed at reducing DAMU.

      Methods

      Data Source

      The data source used for this cross-sectional study was the Fall 2019-Spring 2021
      Fall and Spring semesters are the two intakes for admission purposes among US universities. The Fall semesters usually starts in the late August or early September and ends in late December/early January. The Spring semesters begins in January and ends in early May.
      American College Health Association-National College Health Assessment III (ACHA-NCHA III). The ACHA-NCHA III is a voluntary national survey of college students administered by the ACHA that collects detailed information about students’ health habits and behaviors, including substance use, physical and mental health, and personal safety and violence, using sets of validated and reliable instruments (

      Association ACH. American college health association-national college health assessment II: Reliability and validity analyses 2011. Hanover, MD: American College Health Association 2013.

      ). These survey years were selected because the survey was redesigned to include DAMU starting in Fall 2019 and included COVID-19 related questions starting in Fall 2020. In total, 13,373 and 96,489 students completed surveys in the Fall 2020 and Spring 2021, respectively. The response rate was 13.9% in Fall 2020 and 12.8% in Spring 2021. The ACHA-NCHA III is currently administered as a web-based survey via sending an email invitation to a random sample of students identified by their institution. College students ≥18 years of age are eligible to participate. As this dataset was de-identified, it did not qualify as human subjects’ research by the Institution Review Board at West Virginia University.

      Study Population

      The sample was restricted to respondents who had both used marijuana within the last 30 days and driven a vehicle in the previous 30 days. Specifically, respondents who selected, ‘Within the last 2 weeks’, or, ‘More than 2 weeks ago but within the last 30 days,’ about cannabis/marijuana use (When, if ever, was the last time you used cannabis/marijuana? Please include medical and non-medical use) and driving history (When, if ever, was the last time you drove a car or other vehicle?) were included in the study population. The flow chart of the study population is shown in Figure 1.
      Figure thumbnail gr1
      Figure 1Flow chart of study population. *The driving after marijuana use question was available for participants to answer if they drove in the last 30 days and used cannabis within the last 30 days. During the Spring 2020 semester, schools that began data collection after March 16, 2020, were not included in the reference group due to the COVID-19 pandemic

      Variables

      Dependent variable

      Engaging in DAMU

      If respondents reported having both used marijuana and driven a vehicle in the previous 30 days, they were presented with the question, “Within the last 30 days, did you drive within 6 hours of using cannabis/marijuana?”. If they answered ‘yes’, they were considered to have engaged in DAMU. Those who answered ‘no’ to this question were coded as not having engaged in DAMU. Participants who skipped this question were excluded from the sample.

      Independent variables

      The independent variables were categorized as demographic characteristics, driving characteristics, substance use behaviors, behavioral health problems, and mental health concerns. The ACHA-NCHA III questions used to classify each variable in these analyses are shown in Appendix Table 1. These variables were considered as confounders of DAMU (
      • Begg D.J.
      • Langley J.D.
      • Stephenson S.
      Identifying factors that predict persistent driving after drinking, unsafe driving after drinking, and driving after using cannabis among young adults.
      ,
      • McCarthy D.M.
      • Lynch A.M.
      • Pederson S.L.
      Driving after use of alcohol and marijuana in college students.
      ,
      • Berg C.J.
      • Daniel C.N.
      • Vu M.
      • et al.
      Marijuana Use and Driving Under the Influence among Young Adults: A Socioecological Perspective on Risk Factors.
      ,
      • Li L.
      • Hu G.
      • Schwebel D.C.
      • et al.
      Analysis of US Teen Driving After Using Marijuana, 2017.
      ,
      • Li K.
      • Simons-Morton B.
      • Gee B.
      • et al.
      Marijuana-, alcohol-, and drug-impaired driving among emerging adults: Changes from high school to one-year post-high school.
      ,
      • Carpino M.
      • Langille D.
      • Ilie G.
      • et al.
      Cannabis-related driving and passenger behaviours among high school students: a cross-sectional study using survey data.
      ,
      • Cantor N.
      • Kingsbury M.
      • Hamilton H.A.
      • et al.
      Correlates of driving after cannabis use in high school students.
      ).

      Demographics

      Year in school (Undergraduate, Masters, Doctorate) was categorized based on the recorded education variable in ACHA-NCHA III. ‘Undergraduate’ included participants attending college for undergraduate from 1st year to 5th years or more. ‘Masters’ included attending college for master’s degrees (i.e., MA, MS, MFA, etc.). ‘Doctorate’ included attending college for doctoral degrees (i.e., PhD, EdD, etc.). Sexual orientation was based on answers to the question, “What term best describes your sexual orientation?” Those who identified as ‘Bisexual’, ‘Gay’, ‘Lesbian’, ‘Pansexual’, ‘Queer’, or ‘Questioning’ were classified as ‘sexual minorities’. Ethnicity/Race (Non-Hispanic White, Non-Hispanic Black, Non-Hispanic Other, Multiple races, Hispanic/Latino) used the recorded race variable. ‘Non-Hispanic Other’ included non-Hispanic American Indian or Native Alaskan, non-Hispanic Asian or Asian American, non-Hispanic Middle Eastern/North African or Arab Origin, and non-Hispanic Native Hawaiian or Other Pacific Islander Native. Greek organization member was based on answers to the question, “Are you a member of a social fraternity or sorority?” If they answered “yes”, they were considered as a member of the Greek organization. Those who answered “no” to this question was coded as not a member of the Greek organization.

      Driving characteristics

      The variable indicating previous collisions was based on the answers to the question, “Within the last 12 months, how many times have you been involved in an accident when you drove a car or other vehicle?” If a respondent answered, ‘no accidents,’ in this question were identified as ‘no’. Those who answered ‘one accident’, ‘two accidents’, ‘three or more accidents’, were identified as ‘yes’. Driving while drinking was categorized based on participants who answered, “Within the last 30 days, did you drive after drinking any alcohol at all?” This question was only available for participants to answer when they drove a car or vehicle in the last 30 days and selected, ‘Within the last 2 weeks’ in question, “When, if ever, was the last time you drank alcohol?”. A continuous variable, days of driving with the past 2 weeks, was based on the answers to the question, “Within the last 2 weeks, how many days did you drive a car or other vehicle?”.

      Hazardous or Harmful Alcohol or Drug Use

      The item, “Over the last two weeks, how many times have you had five or more drinks (males) or four or more drinks (females) containing any kind of alcohol at a sitting?”, was used to assess binge drinking as per the Centers for Disease Control and Prevention’s (CDC) definition (

      Centers for Disease Control and Prevention. Binge Drinking. Available at: https://www.cdc.gov/alcohol/fact-sheets/binge-drinking.htm Accessed September 5 2022.

      ). Those who answered, ‘none’, were categorized as ‘no’; those who answered > 1 time were categorized as ‘yes’. The Alcohol, Smoking and Substance Specific Involvement Test (ASSIST) scores (

      Humeniuk R, Henry-Edwards S, Ali R, et al. The Alcohol, Smoking and Substance involvement Screening Test (ASSIST): manual for use in primary care / prepared by R. HumeniukƯ [et al]. Geneva: World Health Organization, 2010.

      ) was used to measure hazardous or harmful alcohol or drug use and were coded in ACHA-NCHA III. For hazardous or harmful alcohol use, ASSIST scores ‘0-10’ indicate ‘low risk’, ’11-26’ indicate ‘moderate risk’, and ‘>=27’ indicates ‘high risk’. For hazardous or harmful marijuana use, ASSIST scores ‘0-3’ indicate ‘low risk’, ’4-26’ indicates ‘moderate risk’, and ‘>=27’ indicate ‘high risk’.

      Behavioral health problems

      Diagnosed alcohol or substance use disorder (AUD/SUD), diagnosed anxiety, and diagnosed depression were categorized based on self-reported chronic conditions variables. Participants were asked whether they were ever diagnosed by a healthcare or mental health professional for these conditions. Those who answered ‘yes’ for each variable were identified as diagnosed with ‘alcohol or drug-related abuse or addiction’, anxiety, or depression.

      Mental health

      Suicide risk was assessed in the ACHA-NCHA III using the Suicide Behaviors Questionnaire-Revised (SBQ-R) instrument(
      • Osman A.
      • Bagge C.L.
      • Gutierrez P.M.
      • et al.
      The Suicidal Behaviors Questionnaire-Revised (SBQ-R): validation with clinical and nonclinical samples.
      ); total scores 3-6 were indicate no suicide risk and scores 7-18 indicate at-risk of suicide. Financial stress due to COVID-19 was defined based on the answers to the question, “How has your current financial situation been affected by the COVID-19 pandemic?” Those who answered, ‘a lot more stressful’ or ‘somewhat more stressful’ were coded as ‘more stressful’; those who answered, ‘somewhat less stress’ or ‘a lot less stressful’ were coded as ‘less stressful’, while those who answered, ‘no significant change’ were coded as ‘no change’. Overall stress level due to COVID-19 was identified based on the answers to the question, “How has your current overall level of stress been impacted by the COVID-19 pandemic?”. Those who identified as ‘significantly increased my level of stress’ or ‘somewhat increased my level of stress’ were classified as ‘increased’; those who identified as ‘somewhat decreased my level of stress’ or ‘significantly decreased my level of stress’ were classified as ‘decreased’; those who answered, ‘no change in my level of stress’ were classified as ‘no change’.

      Statistical analysis

      The percentage of respondents who reported DAMU from Fall 2019 to Spring 2021 were plotted using Microsoft Excel. Both descriptive statistics and logistic regression analyses were restricted to Fall 2020 and Spring 2021 as these periods asked about COVID-19 induced stress, which could have been a confounder of DAMU. Descriptive statistics were used to summarize the data and frequencies and percentages of all the independent variables were used to compare the respondents with and without DAMU. All binary variables were analyzed using Chi-Square tests and all ordinal variables were analyzed using Cochran-Armitage Trend tests with Modified Ridit scoring. Bivariate and multivariable logistic regression analyses were used to determine which independent variables were associated with DAMU using complete case analysis for missing values. A bivariate model was first conducted between each independent variable and the outcome to calculate unadjusted odds ratios (OR) and 95% confidence intervals (CIs). Variable inclusion in the multivariable logistic regression model was based on the bivariate regression model results; if an independent variable was associated with DAMU at alpha ≤ 0.20, it was included in the multivariable model to calculate adjusted odds ratios (AOR) and 95% CIs. Collinearity was assessed using tolerance and variance inflation factor prior to adding the independent variable to the multivariable regression model. Any variable with a tolerance <0.2 or variance inflation factor >5 was considered highly correlated (
      • O’brien R.M.
      A Caution Regarding Rules of Thumb for Variance Inflation Factors.
      ). No collinearity was observed. In addition, in order to verify whether selection bias was present, a sensitivity analysis was conducted to estimate the association between covariates and DAMU by adding participants who drove a vehicle but skipped the question about DAMU (Appendix Table 2). Those participants were considered not engaged in DAMU as they did not use marijuana. All analyses were conducted using SAS Version 9.4. Two tailed hypothesis tests were utilized with α=0.05.

      Results

      Over the four-semester period, the trends of DAMU among US college students changed, and the data suggested a decreasing trend in DAMU (Appendix Figure 1). However, the mean days of driving in the past 2 weeks among students who reported DAMU from Spring 2020 to Spring 2021 did not change significantly (Appendix Figure 2). Descriptive demographic and driving characteristics of DAMU are summarized in Table 1. Overall, 16,531 respondents in the Fall 2020 and Spring 2021 surveys met the inclusion criteria of having both used marijuana and driven within the past 30 days, and 29.9% (n=4,947) reported DAMU. Nearly 31.2% of undergraduates reported DAMU. In respect to gender, 36.2% of males reported DAMU, while only 27.2% of females reported DAMU. Although the majority of respondents were Non-Hispanic White (71.7%), Non-Hispanic Black were more likely to report DAMU compared with non-Hispanic Whites (37.7% vs. 30.9%). Over half of respondents who reported drinking while driving also reported DAMU, but only 25.5% of those who did not drink while driving reported DAMU. Descriptive statistics on substance use behaviors, behavioral health problems, and mental health characteristics of DAMU are summarized in Table 2. DAMU was reported by 43.2% of respondents that had high alcohol risk and 72.1% of respondents with high marijuana risk. Over 50% of respondents who were ever diagnosed with alcohol or drug-related abuse reported DAMU. Respondents with a positive screen for suicide were more likely to report DAMU compared with those who had a negative screen for suicide (35.8% vs. 26.3%). Most respondents experienced increased financial or overall stress during the pandemic, with 66.8% of respondents who answered the DAMU question reported increased financial stress due to COVID-19 and 92.2% of those reported increased overall stress due to COVID-19.
      Table 1Demographic and Driving Characteristics of DAMU on Both Fall 2020 and Spring 2021 Semesters (N=16531)
      Descriptive analysis used to compare DAMU vs. not DAMU counts and row percentages. The “total” column used column percentages. Percentages may not equal to 100% because of rounding. Sample size for each independent variable varies based on the missingness.
      VariablesTotalDAMUP-value
      P-value for Chi-square test statistics was used for binary variables and Cochran-Armitage trend test for ordinal variables. Variables with p-value <0.2 in bold are selected for multivariable analysis.
      16531 (100%)YesNo
      4947 (29.9%)11584 (70.1%)
      Demographic Characteristics
      School year<.0001
       Undergraduates12349 (74.8%)3857 (31.2%)8492 (68.8%)
       Masters2285 (13.9%)647 (28.3%)1638 (71.7%)
       Doctoral1601 (9.7%)360 (22.5%)1241 (77.5%)
       Others265 (1.6%)74 (27.9%)191 (72.1%)
       Missing31922
      Gender<.0001
       Male4940 (29.9%)1789 (36.2%)3151 (63.8%)
       Female11556 (70.0%)3147 (27.2%)8409 (72.8%)
       Others2 (0%)0 (0%)2 (100%)
       Missing331122
      Sexual identity<.0001
       Straight/heterosexual10719 (65.1%)3060 (28.6%)7659 (71.5%)
       Sexual minorities
      Includes asexual, bisexual, gay, lesbian, pansexual, queer, and questioning
      5753 (34.9%)1871 (32.5%)3882 (67.5%)
       Missing591643
      Enrollment Status0.0928
       Full-time14825 (90.1%)4407 (29.7%)10418 (70.3%)
       Part-time1553 (9.4%)493 (31.8%)1060 (68.3%)
       Other81 (0.5%)26 (32.1%)55 (67.9%)
       Missing722151
      Ethnicity/Race<.0001
       Non-Hispanic White11708 (71.7%)3616 (30.9%)8092 (69.1%)
       Non-Hispanic Black632 (3.9%)238 (37.7%)394 (62.3%)
       Non-Hispanic Other1631 (10.0%)376 (23.1%)1255 (77.0%)
       Multiple races421 (2.6%)122 (29.0%)299 (71.0%)
       Hispanic/Latino1941 (11.9%)530 (27.3%)1411 (72.7%)
       Missing19865133
      Greek Organization Member0.7332
       Yes1783 (10.8%)527 (29.6%)1256 (70.4%)
       No14712 (89.2%)4406 (30.0%)10306 (70.1%)
       Missing361422
      Cumulative grade average<.0001
       A9131 (56.4%)2424 (26.6%)6707 (73.5%)
       B5860 (36.2%)1975 (33.7%)3885 (66.3%)
       C or less1187 (7.3%)448 (37.7%)739 (62.3%)
       Missing353100253
      Driving Characteristics
      Previous Collisions<.0001
       Yes1431 (8.7%)530 (37.0%)901 (63.0%)
       No15081 (91.3%)4411(29.3%)10670 (70.8%)
       Missing
      The missing value might include participants who drove a car longer 2 weeks because the previous collision variable only captured participants who drove a car within 2 weeks.
      19613
      Driving after drinking<.0001
       Yes2373 (14.4%)1283 (54.1%)1090 (45.9%)
       No12342 (74.7%)3149 (25.5%)9194 (74.5%)
       Missing1815 (11.0%)515 (28.4%)1300 (71.6%)
      Days of driving within the past 2 weeksMeanSdMeanSdMeanSd<.0001
      8.574.429.624.048.104.50
      Descriptive analysis used to compare DAMU vs. not DAMU counts and row percentages. The “total” column used column percentages. Percentages may not equal to 100% because of rounding. Sample size for each independent variable varies based on the missingness.
      P-value for Chi-square test statistics was used for binary variables and Cochran-Armitage trend test for ordinal variables. Variables with p-value <0.2 in bold are selected for multivariable analysis.
      § Includes asexual, bisexual, gay, lesbian, pansexual, queer, and questioning
      The missing value might include participants who drove a car longer 2 weeks because the previous collision variable only captured participants who drove a car within 2 weeks.
      Table 2Substance Use Behaviors, Behavioral Health Problems, and Mental Health Characteristics of DAMU on Both Fall 2020 and Spring 2021 Semesters (N=16531)
      Descriptive statistics used to compare DAMU vs. not DAMU counts and row percentages. The “total” column used column percentages. Percentages may not equal to 100% because of rounding. Sample size for each independent variable varies based on the missingness.
      VariablesTotalDAMUp-value
      P-value for Chi-square test statistics was used for binary variables and Cochran-Armitage trend test for ordinal variables. Variables with p-value <0.2 in bold are selected for multivariable analysis.
      16531 (100%)YesNo
      4947 (29.9%)11584 (70.1%)
      Hazardous/harmful alcohol or drug use
      Alcohol risk
      Score with 0-10 for low risk, 11-26 for moderate risk, and >=27 for high risk for alcohol based on Alcohol, Smoking and Substance Specific Involvement Test (ASSIST)
      <.0001
       Low11106 (67.2%)3114 (28.0%)7992 (72.0%)
       Moderate3746 (22.7%)1338 (35.7%)2408 (64.3%)
       High389 (2.4%)168 (43.2%)221 (56.8%)
       Missing1290 (7.8%)327 (25.4%)963 (74.7%)
      Marijuana risk
      Score with 0-3 for low risk, 4-26 for moderate risk, and >=27 for high risk for marijuana based on ASSIST
      <.0001
       Low2920 (17.7%)206 (7.1%)2714 (93.0%)
       Moderate11421 (69.1%)3896 (34.1%)7525 (65.9%)
       High1687 (4.2%)495 (72.1%)192 (28.0%)
       Missing1503 (9.1%)350 (23.3%)1153 (76.7%)
      Binge drinking
      Participants can only answer this question when they drank alcohol within the last 2 weeks. According to the CDC’s definition, binge drinking is considered as 5 or more drinks for male or 4 or more drinks for females containing any kind of alcohol at a sitting.
      <.0001
       Yes7298 (44.2%)2354 (32.3%)4944 (67.7%)
       No5776 (34.9%)1556 (26.9%)4220 (73.1%)
       Missing3457 (20.9%)1037 (30.0%)2420 (70.0%)
      Behavioral health problems
      Diagnosed AUD/SUD<.0001
       Yes409 (2.5%)212 (51.8%)197 (48.2%)
       No16000 (97.5%)4699 (29.4%)11301 (70.6%)
       Missing1223686
      Diagnosed anxiety<.0001
       Yes6825 (41.6%)2316 (33.9%)4509 (66.1%)
       No9596 (58.4%)2600 (27.1%)6996 (72.9%)
       Missing1103179
      Diagnosed depression<.0001
       Yes5956 (36.3%)2088 (35.1%)3868 (64.9%)
       No10453 (63.7%)2825 (27.0%)7628 (73.0%)
       Missing1223488
      Mental health
      Suicide risk
      Cut-off point of =>7 for at-risk of suicide based on the Suicide Behavior Questionnaire-Revised (SBQ-R)
      <.0001
       Positive screening6302 (38.3%)2255 (35.8%)4047 (64.2%)
       Negative screening10151 (61.7%)2667 (26.3%)7484 (73.7%)
       Missing782553
      Financial stress due to COVID-190.0199
       More stressful11038 (66.8%)3414 (30.9%)7624 (69.1%)
       No change4516 (27.3%)1211 (26.8%)3305 (73.2%)
       Less stressful968 (5.9%)319 (33.0%)649 (67.1%)
       Missing936
      Overall stress due to COVID-190.0061
       Increased15223 (92.2%)4520 (30.0%)10703 (70.3%)
       No change876 (5.3%)270 (30.8%)606 (69.2%)
       Decreased419 (2.5%)152 (36.3%)267 (63.7%)
       Missing1358
      Abbreviation: COVID-19=coronavirus 2019; AUD= alcohol use disorder; SUD= substance use disorder
      Descriptive statistics used to compare DAMU vs. not DAMU counts and row percentages. The “total” column used column percentages. Percentages may not equal to 100% because of rounding. Sample size for each independent variable varies based on the missingness.
      P-value for Chi-square test statistics was used for binary variables and Cochran-Armitage trend test for ordinal variables. Variables with p-value <0.2 in bold are selected for multivariable analysis.
      § Score with 0-10 for low risk, 11-26 for moderate risk, and >=27 for high risk for alcohol based on Alcohol, Smoking and Substance Specific Involvement Test (ASSIST)
      Score with 0-3 for low risk, 4-26 for moderate risk, and >=27 for high risk for marijuana based on ASSIST
      ∗∗ Participants can only answer this question when they drank alcohol within the last 2 weeks. According to the CDC’s definition, binge drinking is considered as 5 or more drinks for male or 4 or more drinks for females containing any kind of alcohol at a sitting.
      †† Cut-off point of =>7 for at-risk of suicide based on the Suicide Behavior Questionnaire-Revised (SBQ-R)
      The AORs and 95% CIs from the final multivariable logistic regression are shown in Table 3. In the multivariable model, the odds of DAMU were higher among respondents who were undergraduates (AOR: 1.68; 95% CI: 1.42-1.98) and master students (AOR: 1.47; 95% CI: 1.21-1.79) compared to doctoral students, males (AOR: 1.64; 95% CI: 1.48-1.82) compared to females, non-Hispanic Black (AOR: 1.32; 95% CI: 1.02-1.71) compared to non-Hispanic Whites, and sexual minorities (AOR: 1.19, 95% CI: 1.07-1.31). Additionally, the odds of DAMU were higher among students who reported having a diagnosed AUD/SUD (AOR: 1.44, 95% CI: 1.08-1.91) or anxiety (AOR: 1.20, 95% CI: 1.06-1.36), and those that were a positive screen for suicidality (AOR: 1.18, 95% CI: 1.07-1.31). Compared to respondents with low marijuana risk, the odds of DAMU were nearly 7 times higher among those who had moderate marijuana risk and 36 times higher among those who had high marijuana risk. There was a significant association between DAMU and some risky driving behaviors, including drinking while driving (AOR: 3.18, 95% CI: 2.84-3.57) and being involved in a collision within the previous 2 weeks (AOR: 1.20, 95% CI: 1.03-1.40). For every one-day increase in driving, there were 7% higher odds of DAMU. However, there were no significant associations between financial stress or overall stress due to COVID-19 and DAMU.
      Table 3Sociodemographic and behavioral risk factors associated with driving after marijuana use among US college students
      Significant ORs (95% CIs) are bold as they have a p-value <0.05 and OR does not include 1.
      Bivariate modelMultivariable model
      The final multivariable model was adjusted for all eligible variables. The sample size of multivariable logistic regression is (n=10810) after using complete case analysis for missingness.
      VariablesOR (95% CI)AOR (95% CI)

      (n=10,810)
      School year
      Undergraduates1.57 (1.38, 1.77)1.68 (1.42, 1.98)
       Masters1.36 (1.17, 1.58)1.47 (1.21, 1.79)
       Doctoralrefref
       Others1.34 (1.00, 1.79)1.73 (1.18, 2.54)
      Gender
       Male1.52 (1.41, 1.63)1.64 (1.48, 1.82)
       Femalerefref
      Sexual identity
       Straight/Heterosexualrefref
       Sexual minorities
      Includes asexual, bisexual, gay, lesbian, pansexual, queer, and questioning
      1.21 (1.13, 1.29)1.19 (1.07, 1.31)
      Enrollment status
       Full-timeref--
       Part-time1.10 (0.98, 1.23)--
       Others1.12 (0.70, 1.78)--
      Ethnicity/race
       Non-Hispanic Whiterefref
       Non-Hispanic Black1.35 (1.15, 1.60)1.32 (1.02, 1.71)
       Non-Hispanic Other0.67 (0.59, 0.76)0.73 (0.61, 0.87)
       Multiple races0.91 (0.74, 1.13)0.95 (0.71, 1.29)
       Hispanic/Latino0.84 (0.76, 0.94)0.93 (0.80, 1.07)
      Greek organization member
       Yes0.98 (0.88, 1.09)--
       Noref--
      Cumulative grade average
       Arefref
       B1.41 (1.31, 1.51)1.23 (1.12, 1.36)
       C or less1.68 (1.48, 1.90)1.09 (0.91, 1.31)
      Previous Collisions <= 2 weeks
      Yes1.42 (1.27, 1.59)1.20 (1.03, 1.40)
      No
      The missing value might include participants who drove a car longer 2 weeks because the previous collision variable only captured participants who drove a car within 2 weeks.
      refref
      Driving after drinking
       Yes3.44 (3.14, 3.76)3.18 (2.84, 3.57)
       Norefref
      Alcohol risk
      Scores were 0-10 for low risk, 11-26 for moderate risk, and >=27 for high risk of alcohol risk based on ASSIST
       Lowrefref
       Moderate1.43 (1.32, 1.54)0.92 (0.83, 1.01)
       High1.95 (1.59, 2.40)0.65 (0.49, 0.85)
      Marijuana risk
      Scores were 0-3 for low risk, 4-26 for moderate risk, and >=27 for high risk of marijuana risk based on ASSIST
       Lowrefref
       Moderate6.82 (5.89, 7.90)7.61 (6.34, 9.13)
       High33.97 (27.29, 42.27)36.8 (27.66, 48.94)
      Binge drinking
      According to the CDC’s definition, binge drinking is considered as 5 or more drinks for male or 4 or more drinks for females containing any kind of alcohol at a sitting.
       Yes1.29 (1.20, 1.39)1.10 (1.00, 1.21)
       Norefref
      Diagnosed AUD/SUD
       Yes2.59 (2.12, 3.15)1.44 (1.08, 1.91)
       Norefref
      Diagnosed anxiety
       Yes1.38 (1.29, 1.48)1.20 (1.06, 1.36)
       Norefref
      Diagnosed depression
       Yes1.46 (1.36, 1.56)1.05 (0.93, 1.20)
       Norefref
      Suicide risk
      Cut-off point of =>7 for at-risk of suicide based on the SBQ-R
       Positive1.56 (1.46, 1.67)1.18 (1.07, 1.31)
       Negativerefref
      Financial stress due to COVID-19
       More stressful1.22 (1.13, 1.32)1.05 (0.94, 1.17)
       No changerefref
       Less stressful1.34 (1.16, 1.56)1.16 (0.95, 1.42)
      Overall stress due to COVID-19
       Increased0.95 (0.82, 1.10)0.97 (0.79, 1.20)
       No changerefref
       Decreased1.28 (1.00, 1.63)1.34 (0.95, 1.90)
      Days of driving within the past 2 weeks1.08 (1.08, 1.09)1.07 (1.06, 1.08)
      Abbreviations: OR=unadjusted odds ratio; AOR=adjusted odds ratio; COVID-19=coronavirus 2019; AUD= alcohol use disorder; SUD= substance use disorder
      Significant ORs (95% CIs) are bold as they have a p-value <0.05 and OR does not include 1.
      The final multivariable model was adjusted for all eligible variables. The sample size of multivariable logistic regression is (n=10810) after using complete case analysis for missingness.
      § Includes asexual, bisexual, gay, lesbian, pansexual, queer, and questioning
      The missing value might include participants who drove a car longer 2 weeks because the previous collision variable only captured participants who drove a car within 2 weeks.
      ∗∗ Scores were 0-10 for low risk, 11-26 for moderate risk, and >=27 for high risk of alcohol risk based on ASSIST
      †† Scores were 0-3 for low risk, 4-26 for moderate risk, and >=27 for high risk of marijuana risk based on ASSIST
      §§ According to the CDC’s definition, binge drinking is considered as 5 or more drinks for male or 4 or more drinks for females containing any kind of alcohol at a sitting.
      ¶¶ Cut-off point of =>7 for at-risk of suicide based on the SBQ-R
      A sensitivity analysis was used to investigate the association between eligible covariates and DAMU among respondents who drove a vehicle but skipped or did not see the DAMU question (due to skip patterns in the questionnaire) (Appendix Table 2). After adjusting for covariates, the results were similar. However, binge drinking had a significant association with DAMU (AOR: 1.19, 95% CI:1.09-1.31). Additionally, belonging to a Greek organization was not associated with DAMU (AOR: 0.96, 95% CI:0.84-1.09) after adjusting for all covariates in the final multivariable model.

      Discussion

      This study sought to identify sociodemographic and behavioral risk factors associated with DAMU among US college students during the COVID-19 pandemic. This study found that DAMU was associated with being an undergraduate student, male, non-Hispanic Black, sexual minority, and with some mental health (i.e., anxiety, suicide), or substance use issues (i.e., AUD/SUD, marijuana use dependence). Risky driving behaviors, such as drinking and driving and being involved in a previous collision, were associated with DAMU as well.
      Although the number of respondents changed each semester, the percentage of students who reported DAMU steadily declined after Spring 2020 when the COVID-19 response began in the US. This finding can be partly explained by public health measures implemented to reduce the spread of COVID-19. Many colleges and universities limited social interactions by changing from in-person classes to online formats. States also implemented social distancing and stay-at-home orders. Due to this reason, students also might have lived with their parents and could not access marijuana under their parents’ supervision. Therefore, students were not driving or may have had reduced access to marijuana. However, they likely began to socialize and their access to marijuana increased when restrictions were eased, which may have led to them DAMU(
      • Becker S.J.
      • Curry J.F.
      Testing the effects of peer socialization versus selection on alcohol and marijuana use among treated adolescents.
      ).
      In addition to the observed trends, this study found that males had a higher risk of DAMU compared to females, a finding consistent with previous studies(
      • McCarthy D.M.
      • Lynch A.M.
      • Pederson S.L.
      Driving after use of alcohol and marijuana in college students.
      ,
      • Carpino M.
      • Langille D.
      • Ilie G.
      • et al.
      Cannabis-related driving and passenger behaviours among high school students: a cross-sectional study using survey data.
      ,
      • Cantor N.
      • Kingsbury M.
      • Hamilton H.A.
      • et al.
      Correlates of driving after cannabis use in high school students.
      ). This finding can likely be explained by the differences in risk perception between the sexes. Previous studies have shown that male adolescents are less risk adverse compared to females (
      • Reniers R.L.
      • Murphy L.
      • Lin A.
      • et al.
      Risk Perception and Risk-Taking Behaviour during Adolescence: The Influence of Personality and Gender.
      ), and males may be more likely to engage in a variety of high-risk behaviors (
      • Byrnes J.P.
      • Miller D.C.
      • Schafer W.D.
      Gender differences in risk taking: A meta-analysis.
      ). The present study also showed that sexual minorities have a higher risk of DAMU compared to heterosexuals; previous studies identified that sexual minorities are more likely to drive under the influence of illicit drugs (
      • Rudisill T.M.
      • Smith G.S.
      Risk factors associated with driving under the influence of drugs in the USA.
      ).
      The present study also found a strong association between driving after drinking and DAMU, which is consistent with previous studies (
      • Begg D.J.
      • Langley J.D.
      • Stephenson S.
      Identifying factors that predict persistent driving after drinking, unsafe driving after drinking, and driving after using cannabis among young adults.
      ,
      • McCarthy D.M.
      • Lynch A.M.
      • Pederson S.L.
      Driving after use of alcohol and marijuana in college students.
      ,
      • Li L.
      • Hu G.
      • Schwebel D.C.
      • et al.
      Analysis of US Teen Driving After Using Marijuana, 2017.
      ,
      • Cantor N.
      • Kingsbury M.
      • Hamilton H.A.
      • et al.
      Correlates of driving after cannabis use in high school students.
      ,
      • Whitehill J.M.
      • Rivara F.P.
      • Moreno M.A.
      Marijuana-using drivers, alcohol-using drivers, and their passengers: prevalence and risk factors among underage college students.
      ). The data analyzed in this study did not assess DAMU with alcohol simultaneously. Although the present study showed that 54.1% of respondents who reported DAMU were also involved in driving after drinking, this potentially indicates that these substances were used concomitantly. Among the risk factors identified in the analysis, marijuana use dependence was identified as the strongest predictor of DAMU, which is consistent with other studies (
      • Begg D.J.
      • Langley J.D.
      • Stephenson S.
      Identifying factors that predict persistent driving after drinking, unsafe driving after drinking, and driving after using cannabis among young adults.
      ,
      • Cantor N.
      • Kingsbury M.
      • Hamilton H.A.
      • et al.
      Correlates of driving after cannabis use in high school students.
      ,
      • Rudisill T.M.
      • Smith G.S.
      Risk factors associated with driving under the influence of drugs in the USA.
      ). Also, anxiety was associated with DAMU which can be partly explained by those students who may use marijuana as “self-medication” to deal with symptoms of anxiety (
      • Sarvet A.L.
      • Wall M.M.
      • Keyes K.M.
      • et al.
      Self-medication of mood and anxiety disorders with marijuana: Higher in states with medical marijuana laws.
      ). The current study does not indicate that financial and overall stress due to the COVID-19 were associated with DAMU. This finding is likely because the majority of respondents experienced increased financial or overall stress during the pandemic.
      While this study found several sociodemographic and behavioral risk factors associated with DAMU among college students, it is not without limitations. First, the NCHA-NCHA III does not contain items about the perceived beliefs and dangerousness of DAMU and the age of first marijuana use; previous studies have found that social norms (
      • Berg C.J.
      • Daniel C.N.
      • Vu M.
      • et al.
      Marijuana Use and Driving Under the Influence among Young Adults: A Socioecological Perspective on Risk Factors.
      ,
      • Cantor N.
      • Kingsbury M.
      • Hamilton H.A.
      • et al.
      Correlates of driving after cannabis use in high school students.
      ) and the age of first marijuana use (
      • Whitehill J.M.
      • Rivara F.P.
      • Moreno M.A.
      Marijuana-using drivers, alcohol-using drivers, and their passengers: prevalence and risk factors among underage college students.
      ) were associated with DAMU. Second, the data used in this study were self-reported by the students and thus subject to limitations of recall bias and reporting bias. Therefore, it is unknown whether respondents who reported DAMU were more likely to skip the question. Third, another limitation is the timing of when both driving and marijuana use occurred; that is, the dependent variable only applies to recent DAMU, and it may not reflect DAMU more than 30 days ago. Additionally, the primary outcome was driving within six hours of using marijuana. However, it is difficult to identify the true duration of marijuana’s impact on driving with limited knowledge of dose and quantity used; it is not clear if these drivers were actually considered impaired, although it is supported by a report that THC may persist in blood from a single administration for more than 6 hours (

      Compton R. Marijuana-Impaired Driving - A Report to Congress. (DOT HS 812 440). . Washington, DC: National Highway Traffic Safety Administration. ; 2017.

      ). The effects might vary more between individuals than they do with alcohol because of tolerance, differences in administration route (smoking versus oral ingestion), and varying Δ9-tetrahydrocannabinol (THC) absorption rates(
      • Sewell R.A.
      • Poling J.
      • Sofuoglu M.
      The effect of cannabis compared with alcohol on driving.
      ). Next, the causal relationships could not be determined due to the cross-sectional nature of the data. Lasty, a list of institutions that participated in the ACHA-NCHA III each semester could not be obtained. It is possible that not all institutions participated each year or even each semester and not all colleges and universities are members of the NCHA, which may limit generalizability. Additionally, female respondents were over-represented (i.e., 70% of respondents who reported recent driving and marijuana use were females). Also, sample size may not be as representative of all US college students with an average 13% response rate each semester.
      The number of motor vehicle crash fatalities in the U.S. increased 7.2% from 2019 to 2020 despite decreases in driving in 2020 due to stay-at-home orders enacted during the COVID-19 pandemic (

      National Center for Statistics and Analysis. Early estimate of motor vehicle traffic fatalities in 2020 (Crash•Stats Brief Statistical Summary. Report No. DOT HS 813 115). National Highway Traffic Safety Administration.; 2021.

      ). Marijuana is one of the most common substances detected in fatal motor vehicle crashes, particularly among young drivers. The main psychoactive component of marijuana, active Delta-9-tetrahydrocannabinol (THC), had a significantly higher prevalence among seriously or fatally injured road users involved in motor vehicle collisions who presented at five US high-flow level 1 trauma centers and medical examiners during the pandemic compared to pre-COVID-19, and it was more prevalent among those road users during the COVID-19 pandemic compared to alcohol (

      Thomas FD, Berning A, Darrah J, et al. Drug and Alcohol Prevalence in Seriously and Fatally Injured Road Users Before and During the COVID-19 Public Health Emergency. 2020.

      ). Although research has shown that marijuana use impairs driving performance, including cognition, attention, reaction time, and vigilance (
      • Lenné M.G.
      • Dietze P.M.
      • Triggs T.J.
      • et al.
      The effects of cannabis and alcohol on simulated arterial driving: Influences of driving experience and task demand.
      ), the effects of marijuana on driving performance are not predictable due to different individual characteristics (i.e., tolerance, frequent vs. occasional users) (
      • Greene K.M.
      Perceptions of driving after marijuana use compared to alcohol use among rural American young adults.
      ,
      • Brands B.
      • Ciano P.D.
      • Mann R.E.
      Cannabis, Impaired Driving, and Road Safety: An Overview of Key Questions and Issues.
      ) and marijuana itself (i.e., dosage, administration route) (
      • Brands B.
      • Ciano P.D.
      • Mann R.E.
      Cannabis, Impaired Driving, and Road Safety: An Overview of Key Questions and Issues.
      ). Additionally, in contrast to alcohol, which is more easily excreted from the body and has known limits for impairment, we have yet to determine a gold-standard for driving impairment following marijuana use and is still detectable in body fluids long after impairment has stopped (

      Compton R. Marijuana-Impaired Driving - A Report to Congress. (DOT HS 812 440). . Washington, DC: National Highway Traffic Safety Administration. ; 2017.

      ,
      • Van der Linden T.
      • Silverans P.
      • Verstraete A.G.
      Comparison between self-report of cannabis use and toxicological detection of THC/THCCOOH in blood and THC in oral fluid in drivers in a roadside survey.
      ). Assessment of DAMU might only be established with collection and testing of a biological sample and trained drug recognition experts to help identify drug impairment (
      • Johnson O.E.
      • Miskelly G.M.
      • Rindelaub J.D.
      Testing for cannabis intoxication: Current issues and latest advancements.
      ). From a public health perspective, it is probable that concerns regarding DAMU will become more common as states enact recreational and medical marijuana laws (

      National Conference of State Legislatures. State Medical Marijuana Laws. Available at: https://www.ncsl.org/research/health/state-medical-marijuana-laws.aspx2021.

      ,

      Centers for Disease Control and Prevention. Marijuana and Public Health. Available at: https://www.cdc.gov/marijuana/index.htm.

      ,

      Tefft BCA, L.S. Cannabis Use Among Drivers in Fatal Crashes in Washington State Before and After Legalization (Research Brief). . Available at: https://aaafoundation.org/cannabis-use-among-drivers-in-fatal-crashes-in-washington-state-before-and-after-legalization/ Accessed Aug 22 2021.

      ). It is critical that drivers, as well as young drivers, are made aware of these concerns.
      Significant effort must be done to increase prevention programs on campus to focus on DAMU. Additionally, extra efforts to improve education and knowledge of the negative consequences of DAMU on public health, transportation, as well as the general public, particularly younger generations, who have frequent misconceptions and inconsistent beliefs associated with DAMU, are also necessary (
      • Greene K.M.
      Perceptions of driving after marijuana use compared to alcohol use among rural American young adults.
      ). Compared to alcohol, it may be difficult for young adults to quantify marijuana-related impairment as there are no clear dosing guidelines and potency of marijuana can vary (
      • Greene K.M.
      Perceptions of driving after marijuana use compared to alcohol use among rural American young adults.
      ). Additionally, young adults might be confused about the laws governing driving under the influence of marijuana given the uncertain legal limit for marijuana, the significant diversity of policies among states, and the lack of reliable and valid tests for detecting marijuana-related driving impairment (
      • Greene K.M.
      Perceptions of driving after marijuana use compared to alcohol use among rural American young adults.
      ). Future research should be conducted to increase awareness among younger generations and identify effective strategies for changing attitudes against DAMU among drug-using peer networks. Lastly, increased enforcement of existing laws regarding DAMU may be warranted in college communities. In addition, prevention efforts on campuses should aim at reducing DAMU among either the entire campus or high-risk groups such as those who have mental health or existing substance use issues. Student health or mental health clinics may be a viable option for implementing such interventions.
      DAMU is an important public health issue especially among college aged students. This study found that approximately 30% of respondents to a national survey of US college students reported recent DAMU during the COVID-19 pandemic. Various risk factors associated with DAMU were identified as well in this study, such as college students who are male, non-Hispanic Black, sexual minorities, had alcohol or substance use disorder, anxiety, higher suicidality, and involved in drinking and driving. While various interventions exist for drinking and driving (
      U.S. Department of Education OoSaD-FS
      Higher Education Center for Alcohol and Other Drug Prevention.
      ), very few interventions target DAMU on college campuses. Therefore, future research should focus on increasing awareness among those college-aged students who are considered most likely to engage in DAMU, and those groups can be the intervention targets. Prevention efforts and/or strategies, including laws or policies regarding drinking and driving, DAMU and driving, mental health and substance use interventions, to reduce this potentially life-threatening behavior on college campuses.

      Uncited reference

      [
      • Becker S.J.
      • Curry J.F.
      Testing the effects of peer socialization versus selection on alcohol and marijuana use among treated adolescents.
      ].

      Acknowledgements

      Not applicable

      Supplementary Data

      References

        • Azofeifa A.
        • Rexach-Guzmán B.D.
        • Hagemeyer A.N.
        • et al.
        Driving Under the Influence of Marijuana and Illicit Drugs Among Persons Aged ≥16 Years - United States, 2018.
        MMWR Morb Mortal Wkly Rep. 2019; 68: 1153-1157
      1. National Conference of State Legislatures. State Medical Marijuana Laws. Available at: https://www.ncsl.org/research/health/state-medical-marijuana-laws.aspx2021.

        • Schulenberg J.E.P.M.
        • Johnston L.D.
        • O'Malley P.M.
        • Bachman J.G.
        • Miech R.A.
        Monitoring the Future national survey results on drug use, 1975-2020.
        Volume II: College students and adults ageds. 2021; : 19-60
        • Shope J.T.
        • Bingham C.R.
        Teen driving: motor-vehicle crashes and factors that contribute.
        Am J Prev Med. 2008; 35: S261-271
        • Azofeifa A.
        • Mattson M.E.
        • Lyerla R.
        Driving Under the Influence of Alcohol, Marijuana, and Alcohol and Marijuana Combined Among Persons Aged 16-25 Years - United States, 2002-2014.
        MMWR Morb Mortal Wkly Rep. 2015; 64: 1325-1329
        • Carter A.C.
        • Brandon K.O.
        • Goldman M.S.
        The college and noncollege experience: a review of the factors that influence drinking behavior in young adulthood.
        J Stud Alcohol Drugs. 2010; 71: 742-750
        • Jewett A.
        • Peterson A.B.
        • Sauber-Schatz E.K.
        Exploring substance use and impaired driving among adults aged 21 years and older in the United States, 2015.
        Traffic Inj Prev. 2018; 19: 693-700
        • U.S. Department of Education OoSaD-FS
        Higher Education Center for Alcohol and Other Drug Prevention.
        Safe Lanes on Campus:A Guide for Preventing Impaired Driving and Underage Drinking. 2003;
        • Rogés J.
        • Bosque-Prous M.
        • Colom J.
        • et al.
        Consumption of Alcohol, Cannabis, and Tobacco in a Cohort of Adolescents before and during COVID-19 Confinement.
        Int J Environ Res Public Health. 2021; : 18
        • Clendennen S.L.
        • Case K.R.
        • Sumbe A.
        • et al.
        Stress, Dependence, and COVID-19-related Changes in Past 30-day Marijuana, Electronic Cigarette, and Cigarette Use among Youth and Young Adults.
        Tob Use Insights. 2021; 14 (1179173x211067439)
        • Son C.
        • Hegde S.
        • Smith A.
        • et al.
        Effects of COVID-19 on College Students’ Mental Health in the United States: Interview Survey Study.
        J Med Internet Res. 2020; 22e21279
        • Hasan R.
        • Watson B.
        • Haworth N.
        • et al.
        A systematic review of factors associated with illegal drug driving.
        Accident Analysis & Prevention. 2022; 168106574
      2. Association ACH. American college health association-national college health assessment II: Reliability and validity analyses 2011. Hanover, MD: American College Health Association 2013.

        • Begg D.J.
        • Langley J.D.
        • Stephenson S.
        Identifying factors that predict persistent driving after drinking, unsafe driving after drinking, and driving after using cannabis among young adults.
        Accid Anal Prev. 2003; 35: 669-675
        • McCarthy D.M.
        • Lynch A.M.
        • Pederson S.L.
        Driving after use of alcohol and marijuana in college students.
        Psychol Addict Behav. 2007; 21: 425-430
        • Berg C.J.
        • Daniel C.N.
        • Vu M.
        • et al.
        Marijuana Use and Driving Under the Influence among Young Adults: A Socioecological Perspective on Risk Factors.
        Subst Use Misuse. 2018; 53: 370-380
        • Li L.
        • Hu G.
        • Schwebel D.C.
        • et al.
        Analysis of US Teen Driving After Using Marijuana, 2017.
        JAMA Netw Open. 2020; 3e2030473
        • Li K.
        • Simons-Morton B.
        • Gee B.
        • et al.
        Marijuana-, alcohol-, and drug-impaired driving among emerging adults: Changes from high school to one-year post-high school.
        J Safety Res. 2016; 58: 15-20
        • Carpino M.
        • Langille D.
        • Ilie G.
        • et al.
        Cannabis-related driving and passenger behaviours among high school students: a cross-sectional study using survey data.
        CMAJ Open. 2020; 8: E754-e761
        • Cantor N.
        • Kingsbury M.
        • Hamilton H.A.
        • et al.
        Correlates of driving after cannabis use in high school students.
        Prev Med. 2021; 150106667
      3. Centers for Disease Control and Prevention. Binge Drinking. Available at: https://www.cdc.gov/alcohol/fact-sheets/binge-drinking.htm Accessed September 5 2022.

      4. Humeniuk R, Henry-Edwards S, Ali R, et al. The Alcohol, Smoking and Substance involvement Screening Test (ASSIST): manual for use in primary care / prepared by R. HumeniukƯ [et al]. Geneva: World Health Organization, 2010.

        • Osman A.
        • Bagge C.L.
        • Gutierrez P.M.
        • et al.
        The Suicidal Behaviors Questionnaire-Revised (SBQ-R): validation with clinical and nonclinical samples.
        Assessment. 2001; 8: 443-454
        • O’brien R.M.
        A Caution Regarding Rules of Thumb for Variance Inflation Factors.
        Quality & Quantity. 2007; 41: 673-690
        • Becker S.J.
        • Curry J.F.
        Testing the effects of peer socialization versus selection on alcohol and marijuana use among treated adolescents.
        Substance use & misuse. 2014; 49: 234-242
        • Reniers R.L.
        • Murphy L.
        • Lin A.
        • et al.
        Risk Perception and Risk-Taking Behaviour during Adolescence: The Influence of Personality and Gender.
        PLoS One. 2016; 11e0153842
        • Byrnes J.P.
        • Miller D.C.
        • Schafer W.D.
        Gender differences in risk taking: A meta-analysis.
        Psychological Bulletin. 1999; 125: 367-383
        • Rudisill T.M.
        • Smith G.S.
        Risk factors associated with driving under the influence of drugs in the USA.
        Inj Prev. 2021; 27: 514-520
        • Whitehill J.M.
        • Rivara F.P.
        • Moreno M.A.
        Marijuana-using drivers, alcohol-using drivers, and their passengers: prevalence and risk factors among underage college students.
        JAMA Pediatr. 2014; 168: 618-624
        • Sarvet A.L.
        • Wall M.M.
        • Keyes K.M.
        • et al.
        Self-medication of mood and anxiety disorders with marijuana: Higher in states with medical marijuana laws.
        Drug and alcohol dependence. 2018; 186: 10-15
      5. Compton R. Marijuana-Impaired Driving - A Report to Congress. (DOT HS 812 440). . Washington, DC: National Highway Traffic Safety Administration. ; 2017.

        • Sewell R.A.
        • Poling J.
        • Sofuoglu M.
        The effect of cannabis compared with alcohol on driving.
        Am J Addict. 2009; 18: 185-193
      6. National Center for Statistics and Analysis. Early estimate of motor vehicle traffic fatalities in 2020 (Crash•Stats Brief Statistical Summary. Report No. DOT HS 813 115). National Highway Traffic Safety Administration.; 2021.

      7. Thomas FD, Berning A, Darrah J, et al. Drug and Alcohol Prevalence in Seriously and Fatally Injured Road Users Before and During the COVID-19 Public Health Emergency. 2020.

        • Lenné M.G.
        • Dietze P.M.
        • Triggs T.J.
        • et al.
        The effects of cannabis and alcohol on simulated arterial driving: Influences of driving experience and task demand.
        Accid Anal Prev. 2010; 42: 859-866
        • Greene K.M.
        Perceptions of driving after marijuana use compared to alcohol use among rural American young adults.
        Drug Alcohol Rev. 2018; 37: 637-644
        • Brands B.
        • Ciano P.D.
        • Mann R.E.
        Cannabis, Impaired Driving, and Road Safety: An Overview of Key Questions and Issues.
        Frontiers in Psychiatry. 2021; 12641549
        • Van der Linden T.
        • Silverans P.
        • Verstraete A.G.
        Comparison between self-report of cannabis use and toxicological detection of THC/THCCOOH in blood and THC in oral fluid in drivers in a roadside survey.
        Drug Test Anal. 2014; 6: 137-142
        • Johnson O.E.
        • Miskelly G.M.
        • Rindelaub J.D.
        Testing for cannabis intoxication: Current issues and latest advancements.
        WIREs Forensic Science. 2022; 4e1450
      8. Centers for Disease Control and Prevention. Marijuana and Public Health. Available at: https://www.cdc.gov/marijuana/index.htm.

      9. Tefft BCA, L.S. Cannabis Use Among Drivers in Fatal Crashes in Washington State Before and After Legalization (Research Brief). . Available at: https://aaafoundation.org/cannabis-use-among-drivers-in-fatal-crashes-in-washington-state-before-and-after-legalization/ Accessed Aug 22 2021.